Quantitative Signal Detection in Pharmacovigilance

Eugène P. van Puijenbroek

QUANTITATIVE SIGNAL DETECTION IN PHARMACOVIGILANCE

Eugène Paul van Puijenbroek Title: Quantitative Signal Detection in Pharmacovigilance Thesis Utrecht– With ref. – With summary in Dutch ISBN 90-393-2796-3 © 2001 E.P. van Puijenbroek Printed by: Print Partners Ipskamp, Enschede

QUANTITATIVE SIGNAL DETECTION IN PHARMACOVIGILANCE Kwantitatieve Signaaldetectie in de Geneesmiddelenbewaking (met een samenvatting in het Nederlands) PROEFSCHRIFT Ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de Rector Magnificus. ingevolge het besluit van het College voor Promoties in het openbaar te verdedigen op maandag 1 oktober 2001 des middags te 16:15 uur door Eugène Paul van Puijenbroek geboren op 28 september 1960 te Tilburg . dr. WH Gispen. Prof.

dr. Utrecht Institute for Pharmaceutical Sciences The work presented in this thesis was performed at the Netherlands Pharmacovigilance Foundation Lareb and the Utrecht Institute for Pharmaceutical Sciences. dr. ACG Egberts Department of Pharmacoepidemiology and Pharmacotherapy.Promotores Prof. . HGM Leufkens Department of Pharmacoepidemiology and Pharmacotherapy. Utrecht Institute for Pharmaceutical Sciences Prof.

Linde and Sterre .In memory of my parents To Emmy.

.

1 A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions On the assessment of adverse drug reactions from spontaneous reporting systems.CONTENTS Introduction PART 1 METHODOLOGICAL ASPECTS OF QUANTITATIVE SIGNAL DETECTION 1.4 107 . fever and urticaria: symptoms or syndrome? 81 2.2 Different risks for NSAID-induced anaphylaxis Signalling possible drug-drug interactions in a spontaneous reporting system: delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole Detecting drug-drug interactions using a database for spontaneous adverse drug reactions: an example with diuretics and non-steroidal anti-inflammatory drugs Association between terbinafine and arthralgia.1 2. The influence of underreporting on odds ratios 9 21 23 1.2 41 65 67 PART 2 PRACTICAL APPROACHES OF QUANTITATIVE SIGNAL DETECTION 2.3 93 2.

2 141 155 171 177 183 187 191 193 197 General discussion Summary Samenvatting Addendum to Chapter 1.2 Addendum to Chapter 2.1 Determinants of signal selection in a spontaneous reporting system for adverse drug reactions Practical application of quantitative signal detection in the spontaneous reporting system for adverse drug reactions in the Netherlands 123 125 3.PART 3 IMPLEMENTATION OF THE QUANTITATIVE APPROACH 3.4 Dankwoord List of publications Curriculum vitae .

Introduction .

.

5 and the association between the use of practolol and the oculomucocutaneous syndrome.3 In the following years various other serious ADRs in suspected relation to a drug were detected after marketing. Concerning the latter aspect. Clinical trials.2 Consequently. it was not until 1961 that the first report of the possible association between thalidomide and phocomelia was published by McBride. For the detection of ADRs.8 introduced on the international market in 1989. and severe congenital malformations of the limbs of children whose mothers had used this drug in the early stages of pregnancy. clinical trials have certain limitations. some of them for drugs that had been on the market for quite some time. Examples are diethylstilbestrol (DES). is a rare condition in absence of the use of this drug. Others will be detected in the clinical trials conducted before marketing authorisation. The 1960s brought an awareness of the importance of a systematic surveillance of drugs for ADRs. many ADRs only become manifest after the drug has been marketed. efficacy as well as safety. Some of the ADRs can be predicted on the basis of experiences with pharmacologically related drugs. enabling a more systematic surveillance of drugs for possible ADRs. a reliable overview of the possible adverse drug reactions (ADRs) associated with the new drug is indispensable. Although the congenital malformation involved. however. they are generally difficult to detect or predict at this stage.6 Also.4.Introduction The need for pharmacovigilance The development of a new drug is a challenging but also costly process since it needs to focus on quality. more recently. and the risk of valve disorders associated with the use of fenfluramine-phentermine therapy for the treatment of obesitas. 11 . phocomelia. also after marketing. additional serious ADRs were detected.7. are primarily designed for and more specifically aimed at the efficacy assessments of new drugs. the latter having been exposed to DES in utero. with the detection of the association between thalidomide. Thalidomide was introduced in 1956.9 The dramatic experiences with thalidomide in the beginning of the 1960s gave rise to the establishment of national pharmacovigilance centres in many western countries. a hypnotic drug.1 As some ADRs are rare or have a long latency period and others only occur after prolonged use of the drug or are restricted to specific patient groups. Examples are the visual field defects occurring during the use of vigabatrin. which has been associated with an increased incidence of vaginal clear-cell adenocarcinoma in women and their daughters.

7%) between 1996 and 2000.5% of which 50% concerned approvals rescinded on the grounds of safety. safety issues (4% of the marketed products) were the primary cause of market withdrawal. still. Health professionals are urged (in most countries) or obliged by law to report their suspicions of possible associations between drugs and ADRs to these centres. 12 .12 Since these applications concern drugs that are not yet available on the market. Despite all the controversies that have arisen around the spontaneous reporting systems (SRS) over the past decades and the many debates held on their limitations. The number of withdrawals increased slightly during this period. In essence. the Food and Drug Administration (FDA) in the United States approved 543 new molecular entities.9%) were withdrawn and five (2. Between 1972 and 1994. it is not until a drug has been used in clinical practice for some length of time that unpredicted safety-related problems manifest themselves. One of the aims of the SRSs is the so-called signal detection as a first step in the analysis of a possible association between a suspected drug and an ADR. The detection of any (additional) ADRs may lead to changes in the official product information or.11 Unfortunately. as is reflected by the abovementioned withdrawal rates of approved and marketed drugs. more rarely.10 During the period of 1981 to 2000.6%) were rescinded for safety reasons. this may reflect a growing concern for safety aspects. Between 1986 and 1990 two drugs (1. Signals derived from these systems may either confirm or disprove the validity of the efficacy-safety tradeoff of the drugs involved. 59 pharmaceutical products were recalled from the British market from a total of 583 newly approved substances. Fourteen of these approvals (2. In addition to commercial reasons. the centres managing such ‘spontaneous reporting systems’ (SRSs) function as a diagnostic tool. Also the European Agency for the Evaluation of Medicinal Products (EMEA) observed a clear increase in the number of withdrawn applications in the centralised procedure in the period between 1995 and 1998. the approach has remained an important tool for pharmacovigilance. The average withdrawal rate in this period was 22. health professionals and industry gain more insight into the safety aspects of the drugs approved for marketing.Introduction These centres usually rely on the method of ‘spontaneous reporting’ to monitor for unexpected ADRs. even though pre-marketing research has been improved over the years. They are instruments that help health authorities. to the drug’s suspension or withdrawal from the market.

and on behalf of the Medicines Evaluation Board.Introduction Meyboom defined a signal as a set of data constituting a hypothesis that is relevant to the rational and safe use of a drug in humans. drug-food interactions and drug-related syndromes.15 the Pharmacovigilance Foundation Lareb is responsible for the management and maintenance of the spontaneous reporting system concerning drugs that have been approved for the Dutch market.000 reports.13 A signal was defined by the WHO Collaborating Centre for International Drug Monitoring as reported information on a possible causal relationship between an adverse event and a drug. patients usually obtain most of their drugs from the same pharmacist.14 Both definitions share the characteristic that new knowledge about the safety of a drug is generated. Consequently. The board subsequently forwards 13 . in the majority of cases a complete medication history of the patient is available and not only comprises the drugs prescribed by the patient’s general practitioner. however. As with other diagnostic systems. a total of 3. many signals will never be checked mainly as a result of limited resources. drugdrug interactions. to determine its incidence and to identify patients at high or low risk. causality assessment and quantification of the associations are closely related. In the Netherlands. In practice. Co-operation between pharmacist and physician therefore enables an optimal documentation of the SRS reports. but also those drugs that have been prescribed by specialists. In the Dutch healthcare system. an optimal signal-to-noise ratio must be established. but it also provides a database containing reliable and high-quality pharmaceutical information. The reports are sent to the MEB on a weekly basis. They explicitly cover a broad area that is not necessarily restricted to the detection of adverse events and a drug. Both physicians and pharmacists are urged to report suspected ADRs to Lareb. After a signal has been identified often other methods are required to confirm or repudiate the signal. but may also imply the detection of other aspects. In other words. an optimal balance between sensitivity and specificity.000 to 3. for instance risk factors in patients. This not only enhances the quality of the clinical documentation.500 reports are received and evaluated annually and the Lareb database now contains over 30. of which the relationship is unknown or incompletely documented previously. false positive and false negative signals must be found. Evidence supplied by an SRS often remains the only information for regulatory bodies to act on. Currently. Signal detection is a complex process in which the identification of unknown associations.

13 Statistical quantitative approaches can be efficient tools to analyse the data set of an SRS and help minimise the risk that possible signals are missed. Due to the ‘spontaneous’ character of the reporting.22 spontaneous reporting still is an important method of post marketing surveillance. for instance drug-drug interactions.20 Stephens.18 Physicians and pharmacists report to an SRS on a voluntary. has provided an overview of various factors responsible for underreporting. for instance ‘prescription event monitoring’ and ‘intensive monitoring’.19. the data set of an SRS does not necessarily constitute a valid representation of the ADRs occurring in daily practice. Although other techniques are available for the detection of ADRs for marketed drugs. the reports are sent to the WHO Centre for International Drug Monitoring (the Uppsala Monitoring Centre). but they may also be applied for the analysis of more complex relations.21 Unfortunately.Introduction the signals derived from them. to the EMEA. clustering of symptoms and the identification of risk factors in patients. among others. The most noticeable problem is (selective) underreporting. too many factors are involved in underreporting to make a reliable estimate of its extent. Since not all ADRs are reported. such as the case control design. The reports sent to an SRS are coded with different kinds of information. i. basis. Quantitative signal detection Methods applied in analytical epidemiology.17 Pharmacovigilance or ‘post marketing surveillance’ is mainly based on spontaneous reports of such observations.e. like patient characteristics and data concerning 14 . non-interventional. which are primarily used for the evaluation of signals. which obviously play a vital role in the detection of new associations. Methodology of spontaneous reporting First suspicions of possible ADRs are often generated and published by health professionals in the form of case reports.23 Data sets from spontaneous reporting systems are mainly used for the analysis of associations between a suspected ADR and a particular drug. may also serve as techniques for signal detection. and on a quarterly basis.16. the method has some limitations. In addition.

Unfortunately. The final step. Finney already described the basic principles of the majority of these methods in 1974. they all share the characteristic that they search the databases for disproportionality.24 although actual. hypothesis 15 . various statistical measures have been proposed for the application of computer-assisted quantitative signal detection procedures. In other words. computerisation of this process may provide a preliminary sifting of the large amount of information in the database. drug-related syndromes. Quantitative signal detection refers to the application of numerical procedures in pharmacovigilance. which can not only be used for signal detection and analysis of associations between ADR(s) and drug(s). practical application of the concepts was only made possible in recent years due to the introduction of and advances in information technology. they provide an answer to the question: Does the number of observed cases exceed the number of expected cases? The statistical measures of disproportionality all express the extent to which the reported ADR is associated with the suspected drug compared with the other drugs in the database. For 7 centres no specific method was specified. After a suspicion has been raised concerning a possible relation between a drug and an ADR it needs to be corroborated by the accumulation of additional data. Examples of such measures are the proportional ADR reporting ratio. methodological details about the automated approaches used in these centres were not available. but also for drug-drug interactions.26 the Reporting Odds Ratio. A survey by Olsson conducted in 1999 showed that 16 out of 43 national pharmacovigilance centres applied a combination of some kind of automatic screening method and a manual screening in the detection of ADRs.30-32 Although the rationale and the methodology of the various approaches differ.Introduction (concomitant) medication and the suspected ADR(s). Evaluation of such complex relations can be more efficiently carried out by means of a computerassisted quantitative analysis. 33 Signal assessment is often described as a sequential process.25. Given the possibility to generate signals based on a quantitative approach. In the past decades. Among these covariates different relationships relevant for signal detection may exist. The occurrence of ADRs related to other drugs in the database is used as a proxy for the background incidence of ADRs. risk factors in patients and time trends after marketing of the drug under investigation.27-29 and the Bayesian Propagation Neural Network analysis (BCPNN) of the Uppsala Monitoring Centre.

The results yielded by the quantitative approaches. involves the confirmation and quantification of the relation between drug and ADR by epidemiological methods. the relative importance of these approaches differs per step (Figure 1). signal detection is traditionally based on a ‘case-by-case’ approach. In this more or less subjective process quantitative. 16 . like other information in the reports. clinical and pharmacological aspects are taken into account. Trained assessors review each report sent to an SRS and evaluate if the reports show a pattern representing a possible signal. and the relative importance of information derived from case reports and from analytical techniques Hypotheses may be formulated in the literature based on observations in medical practice.Introduction testing. strengthening and testing. information from the literature. or they may follow from the analyses of datasets of spontaneous reporting systems. Although in every step of the process both individual case reports and analytical techniques are involved. With respect to reports sent to SRSs. The process of signal detection. The presence of a statistically significant result does not necessarily imply an actual causal relationship between the ADR and the drug. presented by a point estimate and confidence interval. may suggest that a clear and objective judgement can be made with respect to the existence of a true relationship. and can also be carried out by accumulating and weighing information from different sources. Signal strengthening does not always require the use of (pharmaco-) epidemiological techniques. nor does the absence of a statistically significant result necessarily disprove such a possible relationship. they should always be regarded as having a warning function and not as actual evidence of a possible relationship. However. Case reports Analytical techniques Signal detection Signal strengthening Signal testing Figure 1.34 Signals generated by quantitative approaches should be carefully evaluated in combination with additional information. etc.

in signal detection we are interested in the relationship between the patient (characteristics). Nevertheless. also other combinations may be of interest. Thanks to the availability of (new) statistical techniques spontaneous reporting systems can be used more efficiently. Actual testing of the association should be carried out by other methods. virtually no systematic surveillance of drugs for unexpected adverse drug reactions was performed after marketing. the clustering of two or more symptoms. Datamining techniques can be applied and more complex relationships can be analysed.36 All three approaches are used in the analyses of the reports of an SRS such as the Lareb data set. This most often concerns one drug and one event. Similarly. i. if a certain ADR occurs more often than expected in the event two drugs are used concomitantly.35 The primary goal of pharmacovigilance is to generate signals that indicate the presence of an association between a drug and an ADR.e. a clustering of two ADRs may indicate the existence of a drug-related syndrome. The relationship between patient. For example. Despite the methodological limitations inherent to the method of spontaneous reporting. The detection of ADRs depended to a large extent on the observation by physicians or pharmacists and subsequent communications and publications in the literature. this may indicate the existence of a drug-drug interaction. Patient data Association between drug and ADR Drug-drug interactions Syndromes ADR ADR Drug Drug Figure 2.Introduction Objectives and outline of this thesis Before 1960. it offers a relatively fast and efficient way of generating signals that can be used for balancing efficacy and safety. suspected drug(s) and clinical events. Reports sent to Lareb may concern one or more drugs and one or more ADRs (Figure 2).24. ADR and drug 17 . Generally.

Ioannidis JP. Lancet 1961:1358. Bergstrahl E. Therefore. 18 . In Part 2 of this thesis examples of quantitative signal detection are given and examples of new approaches are introduced. Thalidomide and congenital malformations. The first part of this thesis focuses on the methodology of quantitative signal detection. Increased incidence of cervical and vaginal 3. Barnes AB. Lancet 1998. Chapter 1. Gundersen J. Chapter 3. Noller KL.4. suggestions for future research are made. References 1. Chapter 2.Introduction The objective of this thesis is to evaluate the value and possibilities of quantitative signal detection in spontaneous reporting systems and to add further conceptual insight both into the analysis of the relationship between adverse drug reactions and a suspected drug as well as into complex relationships like drug-drug interactions and drug-related syndromes. Routledge PA (eds). 2. In the third part of the thesis the implementation of quantitative approaches in signal detection is discussed. 1998:197-253.1 the performance of various measures of disproportionality used by SRS centres for the analyses of data of spontaneous reports of ADRs is compared by applying these measures to the dataset of Lareb.2 focuses on the influence of underreporting when using Reporting Odds Ratios on the assessment of ADRs. practical aspects of the implementation of quantitative signal detection are described in Chapter 3. McBride WG. In Chapter 1. they are generally not routinely implemented. Although in various centres quantitative techniques are being developed. drug-drug interactions and drug-related syndromes. Robboy SJ. Talbot JCC. in the General discussion we will elaborate on the methodological considerations and practical implications of a quantitative approach.2. O'Brien P. McGorray S. Townsend D. Contopoulos-Ioannidis DG. Reporting of safety data from randomised trials. New York: Grove's dictionaries inc. Likewise. 4. The pre-marketing establishment of the side-effect profile of a new drug.3 data sent to an SRS also allow analyses of possible drug-drug interactions. Additionally. Stephens MDB.1 deals with the analysis of a possible association between a single drug and a suspected ADR. as is illustrated in Chapter 2. they can be used to study the possibility of clustering of symptoms. Lawrence WD. In: Stephens MDB. Kaufman RH. Detection of new adverse drug reactions.1 deals with the determinants responsible for the selection of signals that are disseminated to the Medicines Evaluation Board and the value of the results of the quantitative signal selection in this process.2 and 2. Finally. As illustrated in Chapter 2.352:1752-3.

In: Stephens MDB. II (continued): How were 18 important adverse reactions discovered and with what delays? Br Med J 1983. Untoward effects associated with practolol administration: oculomucocutaneous syndrome. Welch WR. Nieuwe regels voor het melden van bijwerkingen in Nederland na 1995. 13. Herbst AL. 336:156-8. Payne S .54:483-8. Lancet 2000. II-How were 18 important adverse reactions discovered and with what delays? Br Med J 1983. Lekkerkerker JFF. Begaud B. Experience of the National Collaborative Diethylstilbestrol Adenosis Project. BMJ 1997. Association with clear cell adenocarcinoma in diethylstilbestrol-exposed offspring. 17. Inman WHW.337:581-8.A challenge to the field. Under-reporting of adverse drug reactions. Hensrud DD. Patients taking vigabatrin should have regular visual field testing. Gill EP (eds). Vega T. Br Med J 1975. Wiholm BE. Atypical vaginal adenosis and cervical ectropion. 9. New York: Grove's dictionaries inc. Severe persistent visual field constriction associated with vigabatrin. Maryland. Leakey D. Adverse drug reactions: definitions. Monitoring for Drug Safety. Cancer 1984. Wong IC.140:1166-7. 8.1:595-8. Broekmans AW. Edwards IR. 1998:1-59. 15. Estimate based on a spontaneous reporting scheme and a sentinel system. Lewis JA. Edwards IR. Arias LH.Introduction dysplasia in 3. 14. Wright P. De Koning GHP. Venning GR. Scully RE. Hayllar J. 1999. N Engl J Med 1997. Epidemiologists and adverse event data . Principles of signal detection in pharmacovigilance. 22. Wanju SD. Meyboom RH.45:151-6. and management. Napke E. Lindquist M.54:869-75. Br J Clin Pharmacol 1998. Drug Saf 1997. Detection of new adverse drug reactions. 18. Post Marketing Surveillance 1993. 19. Reaction might be dose dependent. Venning GR. Expected number of adverse reactions in a sample.980 diethylstilbestrol-exposed young women. 12. EMEA/H/14994/99. New active substances authorized in the United Kingdom between 1972 and 1994. Pharmacoepidemiol Drug Saf 1999. Rawlins MD. Valvular heart disease associated with fenfluramine-phentermine. Edwards IR. de Koning FH. Egberts AC. Identification of adverse reactions to new drugs. Ned Tijdschr Geneeskd 1996. McGoon MD. improving public health through human drugs. Crary JL. Gribnau FW. Bortnichak EA. Carvajal A. 2001. 11. CDER 2000 Report to the nation. Edwards WD. Connolly HM. Alvarez-Requejo A. Robboy SJ. Severe persistent visual field constriction associated with vigabatrin. 7. Blackwell N. Rockville. Identification of adverse reactions to new drugs. 1986. Sander JW. Jefferys DB. Tubert-Bitter P. BMJ 1997. Truslow GY . Miremont F. Stephens MDB. 20. Report on withdrawn centralised applications 1995-1998.314:1694. Lancaster: MTP press. 16. Young RH.8:457-61. Introduction. Edwards BS. Moride Y. Food and Drug Administration. Aronson JK. 5. Mawer GE. Kelly G. Lancet 1990. Talbot JCC. 19 . 10. 21.286:289-92. Vree PH. Prat J. 6. Schaff HV.356:1255-9. JAMA 1984.799-118. Routledge PA (eds).286:365-8.252:2979-83. Eur J Clin Pharmacol 1998. Quality criteria for early signals of possible adverse drug reactions. diagnosis.16:355-65. Hekster YA. European Agency for the Evaluation of Medicinal Products. 23.314:1693-4. Begaud B.

Lindquist M.249:2226-8.44:513-8. Eur J Clin Pharmacol 1998. Orre R. 36. Knapp DE. Tijssen JGP. Br J Clin Pharmacol 1997. 25. Bate A. the Uppsala Monitoring Centre. Lansner A. Bakker A. 35. Begaud B. A comparison of selected phase IV studies with spontaneous reporting methods. Discovery of adverse drug reactions. JAMA 1983. Evans S. National Pharmacovigilance systems. O'Neill RT. 28. 32. de Koning GHP. Bate A. Computational Statistics and Data Analysis 2000.16:374-89. Olsson S (ed). 27. Stanley GR.45:1177-84. Meyboom RHB. Andrejak M.Introduction 24. Drug Saf 2000. Hekster YA. Reports of hypoglycaemia associated with the use of ACE inhibitors and other drugs: a case/non-case study in the French pharmacovigilance system database.54:315-21. Meyboom RHB. Serum sickness-like reactions to cefaclor.7(Suppl 2): S102-S102. Analysis of the information in a central ADE database. 26. Kreft-Jais C. Egberts ACG. Amery W. 31. Ollagnier M. Edwards IR. Graham CF.23:533-42. Anello C. 34. 1999.34:47393. Methods Inf Med 1974. Evans SJW. Proportional Reporting Ratios: The uses of epidemiological methods for signal generation. Egberts ACG. 29. Pharmacoepidemiol Drug Saf 1998. Orre R. Edwards IR.13:110. Bate A. Finney DJ. Davis S. 33. Leufkens HGM.5:123. Stricker BHCh.19:3199209. De Freitas RM. Pharmacovigilance: a science or fielding emergencies? Stat Med 2000. Bayesian neural networks with confidence estimations applied to data mining. Drug Saf 1997. Lindquist M. Ståhl M. Waller P. Uppsala. 20 . A retrospective evaluation of a data mining approach to aid finding new adverse drug reaction signals in the WHO international database. Causal or casual? The role of causality assessment in pharmacovigilance. country profiles and overview. Haramburu F. Non-puerperal lactation associated with antidepressant drug use. Br J Clin Pharmacol 1997. Olsson S. Lansner A. Noblet C. Meyboom RH. Mendelis PS. J Clin Epidemiol 1992.44:277-81. Rossi AC. A Bayesian neural network method for adverse drug reaction signal generation. Edwards IR. International Journal of risk & safety in Medicine 1993. Lindquist M. Systemic signalling of adverse reactions to drugs. Moore N. 30.

Like the classical approach.Part 1 METHODOLOGICAL ASPECTS OF QUANTITATIVE SIGNAL DETECTION The first part of this thesis consists of two studies concerning methodological aspects of quantitative signal detection. quantitative procedures can be used to sift data in different ways. One of the most noticeable problems is (selective) underreporting. For quantitative signal detection on the dataset of Lareb. Chapter 1. In addition to the classical approach. To focus attention of reviewers. however. where for each case it is assessed if the reported association represents a new ADR. a Reporting Odds Ratio is used as measure of disproportionality.2 focuses on the influence of selective and nonselective underreporting when using Reporting Odds Ratios in the detection of associations between drugs and ADRs. different measures are used to quantify the extent to which an association between clinical event and drug is reported disproportionally to the generality of the database. quantitative techniques are being developed to facilitate the signal detection process. The objective of the study in Chapter 1. these new techniques are also subject to some limitations. . drug-drug interactions and drug-related syndromes. In various centres.1 is to examine the level of concordance of the various estimates when applied to the dataset of the Netherlands Pharmacovigilance Foundation Lareb.

.

Utrecht Institute for Pharmaceutical Sciences. A Bate2.1.6 1 2 Netherlands Pharmacovigilance Foundation Lareb. Tilburg. Sweden 3 Department of Clinical Pharmacology. R Orre5 and ACG Egberts4. the Netherlands The Uppsala Monitoring Centre.1 A COMPARISON OF MEASURES OF DISPROPORTIONALITY FOR SIGNAL DETECTION IN SPONTANEOUS REPORTING SYSTEMS FOR ADVERSE DRUG REACTIONS EP van Puijenbroek1. Sweden 4 Department of Pharmacoepidemiology and Pharmacotherapy. the Netherlands SUBMITTED FOR PUBLICATION . Utrecht. TweeSteden Hospital and St. HGM Leufkens4. Stockholm University. the Netherlands 5 Department of Mathematics. Sweden 6 Hospital Pharmacy Midden-Brabant. 's-Hertogenbosch.3. Elisabeth Hospital. Umeå University. Uppsala. M Lindquist2.

96se. when applied to the dataset of the Netherlands Pharmacovigilance Foundation Lareb. To focus attention of human reviewers. The concordance increased dramatically when the number of reports per combination increased. Yule’s Q-1. is necessary to optimise signal detection. sensitivity was high with respect to the reference measure when a combination of point. Methods The Reporting Odds Ratio-1.1 Summary Objective A continuous systematic review of all reported combinations of drugs and suspected adverse reactions (ADRs) reported to a spontaneous reporting system. stratified according to the number of reports per combination. was examined. The objective of this study is to examine the level of concordance of the various estimates to the measure used by the WHO Collaborating Centre for International ADR monitoring. Additionally. Results In general. the Poisson probability and Chi-square test of all 17.Chapter 1. 24 .96 standard errors (se). In various centres. Conclusion This study shows that the different measures used are broadly comparable when 4 or more cases per combination have been collected.96se. the concordance of the various tests. the information component (IC). different measures are used to quantify the extent to which an ADR is reported disproportionally to a certain drug compared to the generality of the database. Proportional Reporting Ratio-1.and precision estimate was used.330 combinations were compared with the IC minus 2 standard deviations. quantitative procedures can be used to sift data in different ways.

Comparison of measures of disproportionality Introduction When medicinal products are marketed. because of the sheer load of information to be assessed. case reports of suspected adverse drug reactions (ADRs) are reported to spontaneous reporting systems on a national level. trained assessors regularly examine every incoming reported combination between a drug and a suspected ADR for possible signals in a case-by-case analysis. under the assumption that no relationship exists between the 25 .4 Subsequently these combinations must still be analysed and interpreted by the critical human mind.1 Usually. In quantitative signal detection.e. All case reports are filed in databases at the National Centres as well as sent to the WHO Collaborating Centre for International Drug Monitoring (the Uppsala Monitoring Centre). adequate signal detection without automated quantitative screening is becoming time consuming and inefficient. Several point estimates like the Reporting Odds Ratio (ROR). In contrast to hypothesis testing where quantitative estimates are used to express the frequency of a signal. may represent an important signal based upon a difference from the background frequency. combinations of a drug and a clinical event that are disproportionately high represented in the database.3 Because of increasingly large numbers of case reports being stored in databases. have been used. based on disproportionate reporting. The use of a measure of disproportionality is currently applied in various national spontaneous reporting centres as well as the Uppsala Monitoring Centre. in combination with additional estimators of the precision of point estimates such as the Chi square test. i. or the lower limits of the 95% confidence intervals of the point estimates. One task is to detect and investigate possible new side effects of these drugs. in spontaneous reporting systems they are used to determine the probability of a combination being a signal or not. Often.5-9 Furthermore. Proportional ADR Reporting Ratio (PRR) or Yule’s Q. a limited number of reports represent a signal. A systematic continuous review of the combinations present in the database is necessary to optimise the primary goal of spontaneous reporting systems. monitoring for unexpected or unknown ADRs or signal detection.2 The WHO defines a signal as: ‘Reported information on a possible causal relationship between an adverse event and a drug. the chance of the number of reports being reported on a certain combination. of which the relationship is unknown or incompletely documented previously’.

1 reported suspected ADR and the suspected medication. and to clarify the way they are related to each other. There is no true ‘gold’ standard to compare methods. and these terms should be seen in a relative sense. when applied to the dataset of suspected ADRs reported to the Netherlands Pharmacovigilance Foundation Lareb. instead of using a measure of concordance such as the kappa statistic.13 In order to obtain maximum information out of the comparison. positive predictive value and negative predictive values. Also the FDA is using a Bayesian approach. specifying the relation between the prior and posterior probability before and after linking data fields.12-14 This relationship is expressed as the ‘information component’ (IC).10. Calculations of measures of disproportionality are primarily based upon a two by two contingency table (Figure 1).15 The aim of this study is to examine the concordance of the various estimates. specificity. Two by two contingency table (for corresponding formulas see the overview at the end of this chapter) Since all the measures of disproportionality are based on the same principles of calculation using the 2x2 table. On the other hand it is important to know how the different methods perform in practice. and of adding new data to the database. results should be closely concordant. Reports with the event Reports with the suspected drug All other reports a c Reports with other events b d Figure 1. It is appreciated that the use of ‘sensitivity’ and ‘specificity’ under these circumstances can be misleading. particularly at low numbers of reports of a particular combination. currently being used for example by the Uppsala Monitoring Centre in the Bayesian Confidence Propagation Neural Network analysis (BCPNN).11 Another approach is the use of Bayesian logic. it was decided to express the level of concordance of the other methods with the IC-2sd in terms of sensitivity. can be calculated by means of the Poisson probability.Chapter 1. 26 . but the BCPNN has been tested for performance for signal determination against standard literature sources on a retrospective basis.

Comparison of measures of disproportionality Methods The dataset of the Netherlands Pharmacovigilance Foundation Lareb was used for the analysis. PRR-1. ROR minus 1. the missing combinations were excluded from the analysis. for the use of the various approaches described. specificity. Yule’s Q-1. since the use of a measure of disproportionality was not considered useful in the event that only one report has been received. Based on the 2x2 table and with respect to the background frequency of associations of drugs and suspected ADRs in the database. In this respect. even though measures of disproportionality can be calculated in the latter situation. Yule’s Q-1. Chi square (with Yates' correction) and the use of the Poisson probability. For the corresponding formulas see the overview at the end of this chapter. Lareb maintains the spontaneous adverse drug reaction reporting system in the Netherlands on behalf of the Dutch Medicines Evaluation Board.96se. positive predictive value and negative predictive value as well as the percentage of cases in which the various measures could be calculated were determined with IC-2sd as the reference measure. separate calculations were made for those situations in which number of reports per combination was greater or equal to 2. Since it is not clear whether these assumptions are always fulfilled. Chi square (with Yates' correction) and the Poisson probability.12-14. calculations might be less appropriate in the event of small numbers. a certain statistical distribution is assumed. The BCPNN analysis calculates IC and IC-2sd values for all drug-ADR combinations in its routine use.96se. sensitivity. In the event the measures could not be calculated for mathematical reasons. the following point and precision estimates were calculated for all combinations: IC-2sd.18.19 Furthermore. Only combinations with a minimum amount of 2 reports were selected. was also studied.4 or 6.96 standard error (se).3.96se.96se. For ROR-1.16 All reports received by January 1st 2000 were included in the analysis. PRR-1. Those combinations that have IC minus 2 standard deviation (sd) greater than zero are highlighted for review. a fairly small number of reports may be sufficient to generate a signal. A positive association was defined as IC-2sd>0 and 27 . For both reasons. the concordance of the various tests with the results of the BCPNN as the number of reports per combination varied.96se. The concordance of the different measures of disproportionality were compared with the results of the BCPNN analysis as reference measure.17 In spontaneous reporting.

96se could be calculated for 99.330 different combinations between a drug and a suspected ADR. Results On January 1st 2000.856. For calculating point and precision estimates. in the dataset used in this study. PRR and ROR-1. 2. the various methods were highly sensitive but had a rather low specificity. this subset is of particular interest for signal detection. 1. Nevertheless. In the event that 4 or more reports were received on a combination. 28 . In general sensitivity was high but the specificity was rather low for all measures applied. all approaches gave comparable results. calculating the ROR. all combinations could be analysed. Microsoft Excel 97 was used. for instance the ROR-1. no clear differences were found between the use of the various measures. two.555 reports were filed in the database of the Netherlands Pharmacovigilance Foundation Lareb. When 4 or more reports per combination were present.Chapter 1.947 respectively.2%). 26. three and four or more reports was 11. The number of combinations with one.95% of all the drug-ADR combinations reported 3 or more times. Only when the Poisson probability and Chi square were used.1 negative as IC-2sd<0. In all situations the concordance of combination of point estimate and precision estimate increased dramatically in case the number of reports per combination increases.455.96se was not possible in about 1-5% of the cases. The mean number of reports per combination was 2.790 reported suspected adverse drug reactions (ADRs) which concerned 17.072 and 1. These reports involved a total number of 39. In the majority of cases a combination of point and precision estimate could be calculated.3. Discussion In comparison with IC-2sd. The results of the comparisons for different numbers of reports per combination are presented in Table 1. and the use of the IC-2sd. In the event 2 or 3 cases per combination were reported. Although the percentage of combinations in the Lareb dataset with 4 or more reports were received is rather low (11. including the IC and IC-2sd and the indicators of the concordance.

85 0.00 1.00 1.00 1.0 100.70 0.80 0.9 100.88 0.05) For drug-ADR combinations in the dataset which are listed on less than four reports.91 0.0 100.89 0.0 99.84 1.85 0.81 0.9 99. For all measures sensitivity is still high with respect to IC-2sd.84 0.38 0.70 0.00 1.9 100.96se>1 Yule’s Q 1.0 100.88 0.00 1.0 ROR-1.05) a³2 a³3 a³4 a³6 a³2 a³3 a³4 a³6 1.46 0.41 0.80 0.0 94. specificity.00 1.00 1.0 100.73 0.8 99. regarding different numbers of reports per combination.34 0.96se>1 PRR-1.79 0.82 0. is provided.00 1.61 0.0 100.9 99.00 1.00 1.71 0.74 0.0 100.83 0.73 0.96se>1 Poisson (p<0.00 1.00 1.5 99.00 1.0 100.86 0.9 100. but specificity rapidly declines for such a low number of reports.90 0.00 1.00 1.00 94.00 1.00 1.00 0.86 0.00 1.83 0. Sensitivity.00 1.75 0. results of the various analyses show some differences.Comparison of measures of disproportionality Table 1.89 0.8 99. Also the percentage of combinations for which a point estimate could be calculated.60 0.00 1.00 1.73 0.53 0.44 0.00 1. positive predictive value and negative predictive value concerning the use of various point estimates and tests in comparison with IC-2sd.00 1.00 1. whereas not all combinations highlighted by the other 29 .00 1.00 1.00 1.00 1.00 1. This implies that all combinations highlighted as potential signals by use of IC-2sd are also highlighted by the other measures under investigation.81 0.00 1.00 1.00 1.00 1.00 1.0 100.00 1.0 100.66 0.00 1. Test Negative Percentage Number Sensitivity Specificity Positive of reports predictive predictive calculated value value a³2 a³3 a³4 a³6 a³2 a³3 a³4 a³6 a³2 a³3 a³4 a³6 Chi square (Yates' corr) (p<0.00 1.66 0.77 0.00 1.67 0.

1 measures have a positive IC-2sd. for each of the measures compared to the IC-2sd. For small numbers of reports the distribution may be skewed.g. For instance. Further detailed investigation is needed to attempt to determine which of these scenarios is most likely. Thus careful ongoing evaluation of how these potential signals develop over time may be the most appropriate method of investigation. the missing combinations were excluded from the analysis. or the potential positives highlighted by the other measures are in fact true positives which the BCPNN might go on to highlight later. Gaussian or Poisson distribution) will have a strong influence. on tables with more than a single degree of freedom. a minimum expected frequency of 10 is much safer. For all tests. concerning the use of Chi square.20 When the expected numbers are small. some basic assumptions should be applied. If there is only one degree of freedom (which is the case in our two by two contingency table). poor concordance was found at low counter values. Calculating the confidence interval of the odds ratio and Yule’s Q is also subject to limitations. although when using a Bayesian implementation.Chapter 1. but whichever method was used for comparison. and calculations based upon a Gaussian distribution cannot be applied without caution. This evaluation is made harder due to the lack of a true gold standard for discrimination of true and false signals. In general only a small number of reports per combination is necessary to trigger a signal. This lack of concordance may be explained by the fact that for a small number of reports the assumed type of distribution of the various classical methods (e. When the number of combinations for which a measure could be calculated was not 100%.21 Even so the results of the test should be interpreted cautiously. either the number of false positive signals increases for combinations of less than four reports. but greater than 5. For the tests used. as more information accumulates. calculation of the Poisson probability therefore is safer. In this situation. a minimum expected frequency of 5 can be regarded as adequate. the IC can also be calculated for small numbers. 30 . results are more comparable with IC-2sd when the number of reports per combination increases. These combinations should also be checked by other techniques to determine if they might represent true signals. another option is to apply continuity correction (Yates' correction). Although not presented here. other measures such as the Poisson probability were also studied as the reference measure. So.

the standard deviation was calculated from the IC distribution. making positive ICs less positive and negative ICs less negative than might be envisaged looking purely at the proportion of expected/observed cases. but in the Bayesian approach. PRR. Yule’s Q were calculated as the standard error. as well as other data in signal detection.3 In the event of small numbers of cases other aspects contribute increasingly in the selection of signals. Any combination selected in spontaneous reporting databases using purely statistical methods. when used for analysing the WHO database. we have chosen the information component of the WHO as being the reference measure for the following reasons. by evaluation of the original reports.22 Additional analysis of the signals. both point estimate (IC) and its probability interval can be calculated under all circumstances. in contrast to other measures. for calculating the lower limit of the confidence interval IC-2sd instead of IC-1. this approach yielded a positive predictive value of 44% and a negative predictive value of 85% in the detection of signals as compared with reference literature sources. albeit subject to limitations. An alternative method of comparison could have been used such as the kappa statistic. Secondly. and was used for the retrospective evaluation study outlined above. These differences between the quantitative measures serve to reiterate the crucial importance of clinical and pharmaceutical information.96sd has been used since it is routinely implemented in this way.23-25 Methodological considerations Although in quantitative signal detection no true gold standard is available.13 In this study the authors discuss the difficulty of defining a gold standard in signal detection. but a drawback is that this does not distinguish between a situation of 31 .Comparison of measures of disproportionality The prior assumption for the Bayesian derivation of the IC is of independence between the drug-ADR. should be carefully evaluated and confirmed by other means before making any kind of regulatory decision. We considered that the availability of the performance information in that study. Firstly.e. is therefore warranted. would allow the results of the current study to be placed in a useful context. Furthermore. i. The confidence intervals for the ROR. This causes a dampening effect of the IC at very low numbers of cases. such as the clinical information available and the level of documentation of the reports as the potential signal can be analysed on a case-by-case basis.

therefore. Apart from sensitivity.96 se to calculate disproportionality. also depends on the dataset available. occur more often than the expected frequency. advantages and disadvantages in which the various tests can be used are provided. If we wanted to look for combinations that occur less frequently in combination with a certain drug we should use for instance ROR+1. A major drawback with the ROR. An example of this type of combination is for instance the detection of a phocomelia associated with thalidomide. This situation may occur when rare ADRs are being reported. positive and negative predictive value. Choice of a measure of disproportionality In this study the level of concordance between the different measures used in quantitative signal detection have been examined.26 The choice of a suitable method. In Table 2 the conditions. Similarly the BCPNN approach can be adapted to adjust for covariates. 32 .96se and PRR-1. specificity.96se where the prior assumption is made that combinations we are looking for. The different implementations of these measures have not been considered. For example the MCA use a PRR>3 and Chi square>4 and 3 or more cases as a filter for signal detection. the Netherlands Pharmacovigilance Foundation Lareb presently uses the ROR-1.1 high sensitivity and low specificity. and one of poor sensitivity and high specificity. as these are necessarily dependent on other factors such as the data set used.96se. In the Poisson probability or the Chi-square test only the chance that the observed frequency differs from the expected frequency is provided. since in logistic regression analysis these adjustments can easily be made. or practolol associated with the oculomucocutaneous syndrome.96se and ROR-1. In our analysis we did not take these differences into account. For this reason. is the fact that in case an ADR is specifically associated with a certain drug. This situation differs from the other tests like Yule’s Q-1.Chapter 1. however. there is a risk that the number of reports in cell b or c of the contingency table (Figure 1) contains no reports and subsequently the odds ratio cannot be calculated. the possibility for correcting for covariates can be useful.

b.c and d of the contingency table Standard error cannot always be calculated Difficult to interpret -Standard error cannot always be calculated -Difficult to interpret Correction for different covariates can be easily established in Poisson regression -Always applicable -Large numbers of calculations can be made efficiently -Can be used for pattern recognition in higher dimensions Only p-value provided PRR-1.96se 1 Cells a and c -Easy interpretation have to contain reports Always applicable Chi square (Yates' corr.) Yule’s Q-1.c and d have to contain reports Only for rare events Poisson IC-2sd 0 none Relatively non transparent for people not familiar with Bayesian statistics 33 . advantages and disadvantages of different measures of disproportionality Measure ROR-1. interaction terms can be used for the analysis of drug interactions and syndromes Disadvantage -Odds ratio and standard error can not be calculated if denominator zero (specific ADRs) -Interpretation difficult -Results not always reliable in the event of small numbers in cells a. Conditions.96se 0 cells a.96se Expected Conditions ‘null value’ 1 Cells a.b. analysis.b.Comparison of measures of disproportionality Table 2.c and d have to contain reports Advantage -Easy applicable -Different adjustments possible in logistic regression analysis -In logistic regr.

a case-by-case approach will remain necessary both as an adjunct and an alternative. No individual approach to detect signals is adequate and the concurrent use of other methods is therefore essential.1 The ROR is a transparent measure. An additional advantage of using the odds ratio is the fact that non-selective underreporting of a drug or adverse drug reaction has no influence on the value of the ROR compared with the population of patients experiencing an ADR. easily interpretable which can be easily programmed in database programmes or spreadsheet programmes. Since quantitative signal detection cannot take into account clinical aspects. The heterogeneity of the data collected in databases of spontaneous reporting systems and the variety of biases influencing data (such as underreporting) are likely to have more influence on the potential for signal detection than the small behavioural differences between the measures detected in this study. Conclusion Statistical analyses have been shown to be useful tools in aiding signal detection in spontaneous reporting systems.5 Disproportionality is simply one way of selecting drug-ADR combinations that may be interesting for clinical review. are comparable when more than 4 or more reports constitute the drug-ADR combination. Although no ‘gold standard’ is available.Chapter 1. 34 . The various measures that are being applied in quantitative signal detection in various national centres. Acknowledgement The authors gratefully acknowledge Prof IR Edwards for his valuable comments. each method has its own advantages and disadvantages with respect to applicability in different situations and possibilities for implementation.

Comparison of measures of disproportionality

Overview of measures of disproportionality
Variables used in the different formulas correspond to the 2x2 contingency table of Figure 1 Reporting Odds Ratio (ROR) The ROR can be expressed as5 ROR =

(a / c) = ad (b / d ) bc

The standard error of ln(ROR) and 95 % confidence interval can be calculated by

se(lnROR = )

1 1 (a + b + 1 + d1 ) c

95%CI = eln(ROR)±1.96

æ1 1 1 1ö ç + + + ÷ èa b c dø

Proportional ADR Reporting Ratio (PRR) The PRR can be expressed as
PRR = a / (a + b ) c / (c + d )

The standard error of ln(PRR) and 95 % confidence interval can be calculated by27

se(lnPRR = )

1 1 1 (a - a+b + 1 - c+d ) c

95 % CI = e ln( PRR ) ±1 .96

1 1 1 ö æ1 + ç ÷ è a a +b c c+ d ø

Chi square (Yates' correction) Chi square tests for a 2 by 2 table, with Yates' correction can be expressed as
2 c = å

(O - E )
E

2

-

1 2

35

Chapter 1.1

The summation applies over all four cells of the contingency table. O is the observed frequency and E is the expected frequency of the reports. For example, in case the contingency table is used for the first cell O and E should be calculated as:

O=a

E=

(a + b )(a + c ) (a + b + c + d )

For the other cells it takes the value contained in the cell, i.e. b, c, d in turn Yule’s Q Yule’s Q can be expressed as28

Q=

ad - bc ad + bc

The standard error of Yule’s Q and the 95% CI is calculated by

se

Q

2 æ1 1 1 1ö = 1 æ1 - Q ö ç + + + ÷ ÷ 2ç è ø èa b c d ø

95%CI = Q ± 1.96 SE Q

Poisson probability The Poisson probability is calculated by10
a -1

p = 1- å

e

-m

k =0

´m k!

k

where m is the expected number of reports:
m=

(a + b )(a + c ) (a + b + c + d )

36

Comparison of measures of disproportionality

Information component, resulting from the BCPNN and its variance can be calculated as 17

E ( ICij ) = log 2

(cij + g ij )(C + a )(C + b ) (C + g )( ci + a i )( c j + b j )

V ( ICij ) =

C - cij + g - g ij C - c j + b - bi C - ci + a - a i + + (cij + g ij )(1 + C + g ) (ci + a i )(1 + C + a ) (ci + b j )(1 + C + b ) (log 2) 2

where

g = g ij

(C + a ) (C + b ) × (ci + a i ) (c j + b j )

and γij=1, αi=1, α=2, βj=1, β=2, C is the total number of reports in the database, Cij the number of combinations between a specific drug [i] and the suspected adverse drug reaction [j], Ci the total number of reports on drugs [i] in the database and Cj the total number of reports on the suspected ADR [j] in the database.

References
1. 2. 3. 4. 5. 6. 7. Olsson S. The role of the WHO programme on International Drug Monitoring in coordinating worldwide drug safety efforts. Drug Saf 1998;19:1-10. Finney DJ. The design and logic of a monitor of drug use. J Chron Dis 1965;18:77-98. Edwards IR, Lindquist M, Wiholm B-E, Napke E. Quality criteria for early signals of possible adverse drug reactions. Lancet 1990;336:156-8. Finney DJ. Statisical logic in the monitoring of reactions to therapeutic drugs. Methods Inf Med 1971;10:237-45. Stricker BHCh, Tijssen JGP. Serum sickness-like reactions to cefaclor. J Clin Epidemiol 1992;45:1177-84. Evans SJW, Waller P, Davis S. Proportional Reporting Ratios: The uses of epidemiological methods for signal generation. Pharmacoepidemiol Drug Saf 1998;7(Suppl 2):S102. Egberts ACG, Van der Hofstede JW, Meyboom RHB, de Koning GHP, Bakker A, Leufkens HGM. Transformation of a database of spontaneously reported suspected adverse drug

37

Edwards IR. Haramburu F. Lindquist M. A retrospective evaluation of a data mining approach to aid finding new adverse drug reaction signals in the WHO international database. In: Hayes WL. Edwards IR.8:301-4. the Netherlands. inc. Bate A. Edwards IR. Bayesian Neural Networks with confidence estimations applied to data mining. Waller P and Lee EH. Pharmacoepidemiologic approaches to the evaluation of antidepressant drugs. Noblet C. 13.7:119-37. Fort Worth: Holt. Lindquist M.54:315-21. Meyboom RH. Hekster YA. Reports of hypoglycaemia associated with the use of ACE inhibitors and other drugs: a case/non-case study in the French pharmacovigilance system database.8:535-52. 16. De Freitas RM. Pharmacoepidemiol Drug Saf 1999. Bayesian data mining in large frequency tables. Egberts AC. Pharmacovigilance in perspective. Pere JC. 22. Ståhl M. 9. Computational Statistics and Data Analysis 2000. Egberts ACG. In: Egberts ACG. Broekmans AW. 10. 1988:144-6. The Poisson distribution.21:429-47. Drug Saf 2000. 14. Ståhl M. Olsson S. Post Marketing Surveillance 1993. Utrecht. 20. The assumptions in chi-square tests of association In: Hayes WL. (Thesis) Utrecht University. Lindquist M.Chapter 1. A Bayesian neural network method for adverse drug reaction signal generation.45:283-6. Amery WK. Statistics. 18. Lansner A. 24. From association to alert -a revised approach to International signal analysis. and Gribnau FW. Bate A. 12. Br J Clin Pharmacol 1997. Responding to drug safety issues. Bate A. False-positives in spontaneous reporting: should we worry about them? Br J Clin Pharmacol 1994. Nieuwe regels voor het melden van bijwerkingen in Nederland na 1995. 19. De Koning GHP. Begaud B. Vree PH. Statistics. Tubert-Bitter P. Fucik H. and Nunes AM. Evans SJW. 8. de Koning GHP. Bégaud B. Lellouch J. J Clin Epidemiol 1992. Lindquist M. Drug Saf 2000. Ned Tijdschr Geneeskd 1996. Kreft-Jais C. Moride Y. 1997:111-24. Pharmacoepidemiol Drug Saf 1999.16:355-65. 38 .23:533-42. The American Statistician 1999. Hekster YA. Gribnau FW. Begaud B. Orre R. Haramburu F. Rinehart and Winston inc. 11. Orre R. 1988:780-1. Bate A. Tubert-Bitter P. Drug Saf 1999. with an application to the FDA spontaneous reporting system. Edwards IR. Ollagnier M. 15. Comparing safety of drugs. 21. Meyboom RHB. Power and weakness of spontaneous reporting: a probabilistic approach. Lekkerkerker JFF. Meyboom RH. DuMouchel W. Pharmacoepidemiol Drug Saf 1999. Tubert P.1 reactions and its use as a tool in signal detection. Begaud B.8:S15-S25. Rinehart and Winston.44:513-8. Chaslerie A. Haramburu F.140:1166-7. Assessment of signals generated form spontaneously reported adverse events. Pharmacovigilance: a science or fielding emergencies? Stat Med 2000. 19:3199-209. 23.53:177-89. Eur J Clin Pharmacol 1998. Moore N. 17. Andrejak M.34:47393.38:401-4. Principles of signal detection in pharmacovigilance. Fort Worth: Holt. Lansner A.

28. Proceedings of 20th Annual Conference of the International Society for Clinical Biostatistics 1999. Recognition of signals from spontaneous reporting systems . Philadelphia: Lippincott-Raven.119(abstract). Rothman KJ (eds). 39 . Amery WK. Evans SJW. neural networks. Introduction to Categorial Statistics In: Greenland S. Pharmacoepidemiol Drug Saf 1999.Comparison of different approaches: Poisson model. 2001:231-52.8:147-50. Griffin C. Greenland S. Rothman KJ. 27. Modern Epidemiology.Comparison of measures of disproportionality 25. The Advanced Theory of Statistics. Signal generation from spontaneous event reports. Proportional Reporting Ratios. London: 1995. 26.

.

Leiden. EP van Puijenbroek3.1. THE INFLUENCE OF UNDERREPORTING ON ODDS RATIOS PGM van der Heijden1. ′s-Hertogenbosch. Utrecht. the Netherlands 3 Netherlands Pharmacovigilance Foundation Lareb. the Netherlands SUBMITTED FOR PUBLICATION .2 ON THE ASSESSMENT OF ADVERSE DRUG REACTIONS FROM SPONTANEOUS REPORTING SYSTEMS. S van Buuren1 and JW van der Hofstede1 1 2 TNO Prevention and Health. Department of Methodology and Statistics. the Netherlands Utrecht University.2.

If there is no socalled joint underreporting effect of two drugs on the ADR. i. then SRS data can be useful for assessing the presence of drug-drug interactions. the problem that not all occurrences of ADRs are reported to the SRS. Similar results hold for covariate-drug interactions. Results If the assumption holds that there is an underreporting problem for a first drug. Conclusion In contrast to general underreporting effects.Chapter 1. 42 . but that these two underreporting problems do not influence each other. paired underreporting effects result in biases in the odds ratios estimated from the SRS.2 Summary Objective A well-known problem in spontaneous reporting systems (SRSs) for Adverse Drug Reactions (ADRs) is underreporting. We look at the question of how to draw statistical conclusions from analyses of SRS data using Reporting Odds Ratios. and an underreporting problem for a second drug. The influence of underreporting on the Reporting Odds Ratio is illustrated by two examples.e. Methods We will show that certain underreporting problems play no role in assessing ADRs from SRSs: the results from the analyses turn out to be biased by some specific underreporting problems. but not by others. then reporting odds ratios estimated from SRSs are useful for signalling drug-drug interactions in the ADR experiencing population.

it is not always clear if the ADR is caused by the new drug. Since trials are carried out under controlled circumstances. SRSs can only be used to signal the possible existence of new or unexpected ADRs. among all other medication. and subsequently reported to an SRS. the two situations where only one of the suspected drugs is used.1 Detection of ADRs in clinical trials is hampered by the fact that rare ADRs and ADRs with a long time to onset are difficult to detect. rather small population. Spontaneous reporting systems (SRSs) are commonly used to detect new or unexpected ADRs after the marketing of drugs. such as selective underreporting. 43 .Influence of underreporting on odds ratios Introduction Before marketing a new drug. SRSs can be used to detect such drug-drug interactions. patients with chronic diseases or patients with multiple drug use are even more difficult to detect. like the elderly. however.2-4 A special case is the detection of drug-drug interactions. In these trials.5-7 It is essential in this respect to compare four different situations: reports of patients who. Because of methodological reasons. and finally the control situation where neither drug is used. a drug-drug interaction is usually detected when it is suspected by a physician or pharmacist. or by some other cause. When a patient who already uses one or more drugs is administered another drug. many adverse drug reactions (ADRs) may either be suspected from chemical similarity to known drugs or detected in clinical trials. the detection of drug-drug interactions is more difficult. sometimes only 1 in 70 ADRs were reported. The amount of absolute underreporting can be quite large. Since in clinical trials drugs are used in a specific population.11-13 acknowledges the additional complexities created by underreporting. drugs are used in a selected. but at the same time tends to ignore these complexities in the quantitative analysis. By comparing the ’true’ ADR rate and the reported ADR rate.8-10 Much recent statistical work on the analysis of SRS data. Until now. and multiple drug use is often a criterion for exclusion. the drugdrug interaction. Further (pharmacoepidemiological) studies are needed to evaluate these ADRs in more detail. the detection of ADRs in specific populations. use both drugs suspected of causing a possible drug-drug interaction. an interaction between two different drugs is often detected by the occurrence of an ADR. The present paper gives conditions under which this strategy is appropriate. In daily practice. Based on the idea that in the event of drug-drug interactions the chance for an ADR to occur is increased.

and that not all underreporting problems are bothersome in the analysis of SRS data. however. How can spontaneous reporting systems (SRS) help us to answer this question? It is clear that usually only some of the patients who take the specific drug and experience the specific ADR report to the SRS. the number of reports concerning a possible association between a drug and an ADR may also indicate the presence of a true relationship. the clarity of the causal relation between the drug and the ADR. In section 2 we further motivate the problem by giving three typical examples and do a typical analysis of the SRS data. We are interested in whether the drug is associated with the specific ADR. Because of various confounders and underreporting. Motivating examples Suppose we are interested in the question of whether some specific drug leads to a specific adverse drug reaction (ADR). In the population that takes the drug there is a group of patients who experience this ADR. This underreporting problem can be more or less severe according to the seriousness of the ADR in relation to the indication for use. We summarize and discuss our results in section 6. Apart from qualitative aspects of the reports. 44 . and there is a group of patients who do not experience this ADR. This is called the underreporting problem. The number of reports necessary to generate a signal.8. or due to other possible reasons. For each of the typical examples the answer will then be given in a separate section. a causal relationship cannot be determined. For the underreporting problems that can be problematic.Chapter 1. with special attention to the underreporting problem.15 SRSs are known for their signalling function.14.10. Further studies under controlled circumstances are necessary to demonstrate any possible association.2 In this paper we discuss aspects of the statistical study of SRS data. depends on the total number of reports in the database. The question that remains to be answered is what can be concluded from the analysis in the light of underreporting problems. 2. we discuss what can be concluded under what type of assumptions. We will show that there are different types of underreporting problems. namely sections 3 to 5. the total number of ADRs in the database and the number of reports concerning the association.

Influence of underreporting on odds ratios In SRSs this problem is often tackled by comparing the different profiles of ADRs reported.19 The Netherlands Pharmacovigilance Foundation Lareb collects and analyzes reports of suspected adverse drug reactions from health professionals in the Netherlands. but regarding the population of patients who actually experience an ADR. However. not only reports involving both the specific ADR and the specific drug. the use of the Reporting Odds Ratio may provide a valid estimate.6.15-17.16.14. Usually a cross-classification of all reports available from the SRS is constructed. This Proportional Reporting Ratio can be considered as a representation of the safety profile of a drug.e. Moreover. due to the increasing amount of reports.18 In other words. For this reason. which implies that reporting of the suspected ADR by a physician or pharmacist is not compulsory. based on the same 2x2 contingency table. Based on this crossclassification for instance a Reporting Odds Ratio can be calculated.20 For every possible association present in the database. Indeed the underreporting factors cannot account properly for the size of the exposed population. a statistical approach enables the identification of more complex relationships like drug-drug interactions and the analysis of syndromes. but also reports involving other ADRs and other drugs. We discuss three examples using data provided by Lareb. 45 . The Reporting Odds Ratio offers advantages in the sense that in a logistic model adjustments for various confounders can be made and statistical interactions between various covariances can be analyzed in more detail. underreporting is inherent in this approach and signals of possible ADRs from SRSs should be considered in this perspective. i. Every report is assessed on a regular basis.1 Diuretics and possible ADRs Assume that we would like to investigate whether there is an association between the presence of diuretic drugs on the report forms and ADRs possibly representing the presence of congestive heart failure (see Table 1).7 These reports have a ’spontaneous’ character. the influence of underreporting on various factors should be carefully weighted. analysis by the human mind alone becomes more difficult and statistical analysis of the data may be helpful. 2. other reports in the database provide a proxy of the ’background incidence’ of the ADRs.17 Another option would be applying a ’Proportional Reporting Ratio’.

Cases were defined as reports in which one of the following WHO preferred terms were present: oedema.21 drugs are coded according the ATC terminology.2 Table 1. is identical to this probability in the reports where diuretics are absent.58 (95 percent confidence interval is 1. Exposure categories were the presence of diuretics among the medication used (ATC code beginning with C03) versus no diuretics. or oedema legs. Presence of a diuretic drug. oedema dependent. given the fact that diuretics are commonly used in the treatment of congestive heart failure. oedema generalized.820 1. This model is rejected (likelihood ratio chi-square is 10. which is not surprising. Since we are interested in the population of patients that actually uses the drugs in question. cardiac failure right. This is equivalent to fitting the independence model. df is 1. a total number of 9.822 reports concerning patients older than 50 year were received by Lareb.7 A selection was made of WHO-preferred terms that might indicate the presence of signs of congestive heart failure. showing that the use of diuretics and signs of possible congestive heart failure are related in the SRS data. oedema peripheral. A report may be used to report one or more suspected drugs and one or more suspected adverse drug reactions. or a lack of efficacy of the drugs concerned. A typical way to analyze the data is to fit a model in which the probability of signs of congestive heart failure in the reports where diuretic drugs are present. cardiac failure.9. Observed frequencies Diuretic drug absent present Congestive heart failure absent present 7.21-2. Non-cases were defined as all other reports. and signs of congestive heart failure as an ADR.05).697 227 78 The suspected ADRs of reports to Lareb are coded by means of the WHO adverse drug reaction terminology. The observed odds ratio is 1.Chapter 1. we subsequently want to evaluate whether this relation in the SRS data is a close representation of the relation in the ADR experiencing population? For this we 46 . Between January 1st 1990 and January 1st 1999. The reports concerned therefore might either represent the background incidence of oedema or congestive heart failure. p<0. cardiac failure left.001). of which sex of the patient involved and the type of reporting health professional (physician of pharmacist) were known.

but such a model fits poorly (LR chi-square is 7. Congestive heart failure as ADR. Two drugs: Presence of a diuretic drug and presence of an NSAID.293 253 Congestive heart failure present 185 53 42 25 A typical analysis would again involve the use of odds ratios. the concomitant use of NSAIDs and diuretics leads to an increase in signs of congestive heart failure. in the SRS data. p<0. We conclude that. ATC code beginning with M01A) can lead to an increased risk of developing signs of congestive heart failure. showing that these two observed odds ratios differ significantly.23 due to a decreased efficacy of the diuretics involved. In a loglinear model with no three-factor interaction these two odds ratios would be restricted to be equal. databases of SRSs may also be used to generate signals of possible drug-drug interactions or identifying risk factors in patients. Observed frequencies NSAID absent present Diuretic drug absent absent present absent present 6. Table 2.Influence of underreporting on odds ratios need a better understanding of the underreporting problem. df is 1. 2.04. For the data in Table 1 we now would like to investigate whether there is an indication of a drug-drug interaction of diuretics and NSAIDs with signs associated with the presence of congestive heart failure. whereas this odds ratio with NSAID prescription is 3. Consider the following example.444 1.22. Several case-reports and studies suggest that concomitant use of diuretics and Non-Steroidal AntiInflammatory Drugs (NSAID.7 Exposure categories were the use of NSAIDs or diuretics versus the use of neither of these drugs (see Table 2). 47 . The odds ratio for diuretics and congestive heart failure without NSAID prescription is 1.527 1. and we will elaborate this problem in section 3. But what can be said about the ADR experiencing population? This problem will be addressed in section 4.01).29.2 Drug-drug interaction of diuretics and NSAIDs Besides generating signals concerning ADRs.53.

0 (95% CI 7. The estimated odds ratio estimated in this way is 10.399 reports received by Lareb between 1990 and 1999 of patients older than 10 years were included. Acute allergic reactions are associated with various NSAIDs including diclofenac. 48 . We analysed this association using a case-control design.8.s.108 181 10.21 Cases were defined as all reports coded with ‘anaphylactic shock’ or ‘anaphylactoid reaction’. and sex as covariate.8) and for females 10.7-17. anaphylactic reactions as ADR. controlling for sex. But in what way might underreporting influence these results? This will be discussed in section 5. All 17. All other reports were considered as non-cases.Chapter 1. which would suggest an increased risk of these reactions during the use of diclofenac.6-13.658 331 30 11 61 19 male female absent present absent present The observed odds ratio for males is 12. we investigate whether the two observed odds ratios differ significantly. Diclofenac as drug. Observed frequencies Sex Diclofenac Anaphylactic reactions absent present 6. for males as well as females. All reports were coded using the WHO adverse drug terminology. The likelihood ratio chi-square is 0.3 Sex differences for ADRs of diclofenac Controlling for covariates is illustrated by a last example. By fitting the loglinear model without interaction term for sex by drug by ADR. See Table 3.24 Although these ADRs are mentioned in the Dutch Summary of Product characteristics. n. Table 3. These results show that.4 (95% CI 8. Lareb received a substantial number of reports on anaphylactoid reactions or anaphylactic shock associated with diclofenac.) showing that the data do not provide evidence for a difference.22 (df is 1.1). anaphylactic reactions are disproportionally reported on diclofenac as compared to other NSAIDs.2 2.

However. We distinguish four types of problems of underreporting: 49 . one for each frequency.Influence of underreporting on odds ratios 3 The underreporting problem for one drug 3. the number n11 is in most cases much greater than the other three numbers. namely by distinguishing between four separate processes that cause these four frequencies to be underreported. different forms of underreporting have effects on these numbers. congestive heart failure) having levels ’present’ (in report) and ’absent’. each of the frequencies is plagued by underreporting. Table 4 shows the observed counts in the SRS population. However. where i=1. If the specific ADR is more common with the specific drug than with other drugs. It is clear that. Therefore we want to know in what way the elements in Table 4. Naturally. When looking for ADRs in the SRS population. Thus we could say that there are four underreporting problems. diuretics) having levels ’present’ (in report) and ’absent’. showing the information in the SRS population. The underreporting problem in data from a spontaneous reporting system Specific drug absent absent present n11 n21 Specific ADR present n12 n22 Assume a variable ’specific ADR’ (for example. a relative high count of an ADR reported for a certain drug is used as a signal for a more detailed study. denoted by nik.1 Theory We will now discuss the question raised at the end of section 2. Table 4. and k=1. assume further a variable ’specific drug’ (for example. since we have four frequencies in Table 4. provide us with information of the ADRs in the ADR experiencing population.2 indexes the levels of the specific drug. one would expect n22/( n21+n22) to be larger than n12/( n11+ n12).1: in what way does underreporting distort the relation between estimates from the SRS data and the ADR experiencing population? Table 4 will help to answer this problem.2 indexes the levels of the specific ADR. it is more insightful to approach these four underreporting problems differently. since this embraces all the reports where neither the specific drug nor the specific ADR play a role (compare Table 1).

Thus the observed frequencies nik are realisations of the expected frequencies mik from the SRS population.Chapter 1. which is usually a cosmetic indication.18 Since this antifungal drug is used frequently for onychomycosis. for which we will denote the true frequencies by tik.2 (i) there is an overall underreporting problem. which pertains to all cells of Table 4. The ADR is more easily accepted by physicians and patients because of the necessity for the use of these drugs. but not from the ADR experiencing population. hair loss caused by chemotherapeutic drugs is a commonly occurring ADR. from the SRS population. The problem that we study in this paper is that we would like to make statements about the ADR experiencing population on the basis of the SRS population. There are different ways to work out this 50 . We will now explore this further using appropriate notation. For instance: death due to a serious allergic reaction is more likely to be reported than mild gastro-intestinal side effects. (ii) there is an underreporting problem for the specific drug compared to the other drugs. in the sense that the overall factor in (i) does not show that some ADRs are more often reported than others. (iii) there is an underreporting problem for the specific ADR compared to the other ADRs. Under (iv) therefore it is the severity of the ADR compared to the severity of the indication that seems to play a role. in the sense that the overall factor in (i) does not show that for some drugs the underreporting problem is more severe than for others. We distinguish the ADR experiencing population. the occurrence of hair loss is reported relatively frequently to Lareb. for which we will denote the expected frequencies by mik. Media attention might also increase the reporting rate of a specific drug. For instance. The question is: under what assumptions are we allowed doing this? We assume that the true frequencies tik from the ADR experiencing population are related to the expected frequencies mik from the SRS population by the four underreporting problems (i) to (iv). that is rarely reported. (iv) there is an underreporting problem for each combination of levels of the specific ADR and the specific drug that describes a deviation from the overall effect (i). the specific drug effect (ii) and the specific ADR effect (iii). For instance: recently introduced drugs are more likely to be reported. Hair loss is a relatively rare ADR of terbinafine.

and ADR absent and drug present. We will now look at the way these unknown constants bother us when we want to 51 . when we go from the ADR experiencing population tik to the SRS population with elements mik. expressed in terms of true frequencies tik used for the ADR-experiencing population Specific drug absent absent present Specific ADR present ccda t11 cd ca ccd t 21 ca cda cca t12 cd cda ccd ca cda t 22 Using the data from an SRS we are unable to estimate the constants c. each element tik has to be multiplied with c. Table 5.the general effect (i) will be denoted by an overall underreporting factor c that is identical for each of the four frequencies. Table 5 shows that the expected frequency m22 in the SRS population is related to the expected frequency t22 by m22=ccdcacdat22. In Table 5 the expected frequencies mik of the SRS population are expressed in terms of the true frequencies tik of the ADR experiencing population taking the four underreporting problems into account. . will both be 1/cda. Expected frequencies mik for the SRS population. .the effect for the combination (iv) will be denoted by cda if both the specific drug and ADR are present. One drug. cd. That is. ca and cda to derive the true frequencies of the ADR experiencing population. which leads to the following notation: .Influence of underreporting on odds ratios relation in mathematical terms. . and the effect for both ADR and drug absent by cda. In order to let the product of combination effects over the rows and over the columns equal one. the effect for ADR present and drug absent.the ADR effect (iii) will be denoted by ca if the specific ADR is present and 1/ca if not. We use deviation coding.the drug effect (ii) will be denoted by cd if the specific drug is present and 1/cd if not (note that the product of cd and 1/cd equals one).

the conclusion as to whether the estimate of θ is an over. therefore when a particular contingency table like Table 4 is studied. The underreporting problems (i). many of the constants vanish: m11 m22 = ( )4 t11t 22 = ( )4 t (2) q = cda c da q m12 m21 t12 t 21 This shows that estimate θ for the SRS population gives a biased account of the odds ratio of interest θ t due to cda. It should be born in mind that the factor cda should be interpreted as a factor which corrects the more general correction effects c. i.or an underestimate of θ t can be drawn only from substantive knowledge of the processes that play a role in reporting.2 determine from SRS data whether the specific drug causes the specific ADR in the ADR experiencing population. We now make a remark on the estimation of the odds ratio by the loglinear model and by logistic regression as this is helpful in understanding the generalisation of our findings to drug-drug interactions. We only need concern ourselves with the specific combination underreporting (iv). it is difficult to say for particular applications whether cda = 1.16.Chapter 1. Basically.26 Without detailed knowledge of the subject matter. we do not need to bother about the overall underreporting discussed in (i). In what way would this estimate provide a biased account of the odds ratio in the ADR experiencing population? The odds ratio θ t for the ADR experiencing population is t11 t 22 t q = t12 t 21 (1) and when we derive the odds ratio θ for the SRS population by using the relations provided in Table 5. about general underreporting of the drug discussed in (ii). This is not surprising given the well-known property of the odds ratio that it is insensitive to marginal changes in a two-way contingency. and it enables us later in this paper to control for possible covariates such as gender and age. and general underreporting of the ADR discussed in (iii).e. Let us assume that we want to study a possible relation between the specific drug and the specific ADR by estimating the odds ratio from observed counts derived from an SRS. combined effect (iv). in deciding this. cd and ca. (ii) and (iii) are not important if odds ratios are used. The loglinear model for the 2 x 2 table with elements mik is 52 . some others already noticed this.25 In the context of underreporting in SRSs. But equation (2) shows that. cda >1 or 0< cda <1.

ui 2 = 2 u1 + 2 ui1 = b + bi . These parameters add up to zero over each index in order to identify the model. and u ik refers to their interaction effect. 3. namely the general effect (i) that holds for all four frequencies in Table 1.1 shows that three underreporting effects do not concern us here.e.2 Diuretics and possible ADRs revisited In section 2.Influence of underreporting on odds ratios log mik = u + u i + u k + u ik . and it follows that both the loglinear model as well as the logistic regression model can be used to test whether the odds ratio in the SRS departs significantly from 1. A D AD (3) D where. Only the combination effect (iv) ( ) 53 . The odds ratio estimated in the SRS data was 1. We can also rewrite the loglinear model into a logistic regression model: log mi1 mi 2 = u1 + ui1 . u k refers AD to the drug effect. Filling in (3) into (2) gives us the relation between the parameters uij and the odds ratio θ AD q = m11 m22 = exp(4u11 ) q m12 m21 (4) AD It follows that. the estimated odds ratio will be equal to one.1 we studied Table 1. and the ADR effect (iii) that the underreporting of congestive heart failure may be different from underreporting other ADRs. the drug effect (ii) that the underreporting of ADRs the group for which diuretics are prescribed differs from ADRs in those for which they are not prescribed. u iA refers to the ADR effect. if there is no interaction term u ik needed in the loglinear model for Table 4 (i. in the SRS data the specific drug and ADR are unrelated). Section 3. and the question is in what way this odds ratio is biased by underreporting. u refers to the overall mean effect.58. ik D AD D AD D AD A (5) This shows that q = exp 2b1A .u 2 .

We adopt the same approach as in section 2.e.3 we reconsider the example of drug-drug-ADR interaction of diuretics and NSAIDs. the odds ratio in the ADR experiencing population will be larger. i. possibly due to a confounding by indication. If the bias is positive. First we write the expected frequencies of the SRS in terms of the true frequencies for the ADR experiencing population using constants c to cdea for these eight types of underreporting problems. j (j = 1. 54 . i. Since the aim of an SRS is to generate signals for an existing drug-ADR relation. the odds ratio may be overestimated in the SRS population.2) refers to the presence or absence of the first specific drug in a report. There are thus eight cells.2).2) refers to the presence or absence of the specific ADR. the odds ratio in the ADR experiencing population will be smaller. In this particular case we may be looking at a situation in which a lack of efficacy of the diuretics was perceived. and the SRS is possibly generating a ’false’ signal. 4 The underreporting problem for two drugs First we will work out the general case of two drugs (section 4. 4.1 and 4. then proceed to make some further assumptions and study the consequences of these (section 4. the general case In this section we will study the underreporting problem in case of two drugs.1). This pertains to a table of 2 x 2 x 2 with elements mijk. the possible effect that physicians and pharmacists submit in more or fewer reports specifically for the occurrence of signs indicating congestive heart failure when diuretics are prescribed. and discuss the consequences of sections 4.2 for the interpretation. for these eights cells we distinguish eight types of underreporting problems.2) refers to the presence or absence of the second specific drug.Chapter 1. If the bias is negative. negative bias should not worry us in this example. where i (i = 1.1 Theory.58 positively or negatively. Such a possible effect will bias the observed odds ratio of 1. Then we will show in what way the odds ratios calculated on the SRS data are biased as a result of these underreporting problems. However. in general the bias may be such that it reduces the odds ratio to a small value that then is not a large enough signal to stand out from the noise. In section 4.2 should concern us here. Nevertheless. since this situation is not to be expected. and k (k = 1.e.

55 . We will denote the odds ratios derived from the SRS between i and j for k = 1. denoted by ca. See section 4.2 for a thorough discussion of the joint underreporting effect.e.Influence of underreporting on odds ratios We distinguish the following eight types of underreporting problems: (i) the general underreporting effect that pertains to each of the eight cells. For example.2. denoted by ce. In Table 6 we relate the expected frequencies mijk in the SRS population to the true frequencies tijk in the ADR experiencing population using the constants c to cdea. i. (iii) the general underreporting effect for the second drug. denoted by cd. denoted by cdea.2. denoted by c. (viii) the joint underreporting effect for both drugs with the specific ADR. and lastly. above the general and paired underreporting effects. denoted by cea.for example. this statistical interaction is unrelated to the column variable ADR. (iv) the general underreporting effect for ADR. cdea can be smaller than 1 when a drug-drug-ADR interaction is well known so that health professionals do not take the trouble to file a report. (v) the paired underreporting effect between both drugs denoted by cde. denoted by cda (this effect is comparable to combination effect (iv) in section 2). and two odds ratios between j and k for k = 1. This underreporting effect shows the effect that a combination of two drugs has on reporting an ADR. two odds ratios between i and j for k = 1. In three-way tables. the possible relations between the variables can be summarised by so-called conditional odds ratios. and the corresponding odds ratios derived from the ADR experiencing population by q tij k =1 and q tij k = 2 . two odds ratios between i and k for k = 1.2. Thus we can answer the question of how the SRS estimates provide a biased account of the odds ratios in the ADR experiencing population. but only shows whether the first drug is found relatively more or less often together with the second drug. in the situation that a new drug is marketed with properties similar to existing drugs for which a drug-drug-ADR interaction is proven. (ii) the general underreporting effect for the first drug.2 as q ij k =1 and q ij k = 2 . (vii) the paired underreporting effect for the second drug and the specific ADR. it can be larger than 1 when a drug-drug-ADR interaction is suspected but not proven yet . (vi) the paired underreporting effect for the first drug and the specific ADR.

Two drugs. expressed in terms of true frequencies tijk used for the ADR experiencing population Specific drug 1 absent Specific drug 2 absent absent present present absent present Specific ADR present ccde cda cea t111 cd ce ca cdea cce cda cdea t121 cd ca cde cea ccd cea cdea t 211 ce ca cde cda ccd ce cde t 221 ca cda cea cdea cca cde cdea t112 cd ce cda cea cce ca cea t122 cd cde cda cdea ccd ca cda t 212 ce cde cea cdea ccd ce ca cde cda cea cdea t 222 The odds ratios for the other variables are denoted in similar ways. Expected frequencies mijk for the SRS population. If we want to relate the odds ratios derived from the SRS population to the odds ratios derived from the ADR experiencing population we get the following results (compare equations (1) and (2)): m111 m221 = (c de ) t111 t 221 = (c de ) t q ij k =1 = 4 4 q ij k =1× 4 4 m121 m211 4 (c dea ) 4 t121 t 211 (c dea ) (6) t q ij k =2 = (c de ) (c dea ) q ij k =2× (7) (c da )4 t q ik j =1 = q (c dea )4 ik j =1× q ik j =2 (8) = (c da ) (c dea ) q tik 4 4 j = 2× (9) q jk i =1 = (c ea )4 t q (c dea )4 jk i =1× (10) 56 .2 Table 6.Chapter 1.

three different situations can be distinguished. and their interaction. 57 . ce and ca do not appear in (6) to (11). the chance that this drug-drug interaction is reported to an SRS is probably less likely. cd. 3. In this situation.Influence of underreporting on odds ratios t q jk i =2 = (c ea ) (c dea ) q jk i =2× 4 4 (11) We conclude from these equations that 1. the odds ratios estimated from SRS data are biased by paired underreporting effects. further assumptions Regarding cdea. general underreporting effects (i). 2. Secondly a drug-drug interaction may not yet be associated with a specific drug but may be expected. the odds ratios estimated from SRS data are biased by the joint underreporting effect for both drugs. the odds ratios q ij k and q jk i can be obtained from the parameters of a logistic regression model where the logit of the specific ADR is predicted by the first and the second drug. j jk (12) where the parameters add up to zero over each index. and the specific ADR. (iii) (for the second drug) and (iv) (for the specific ADR) do not lead to bias in the odds ratios derived from the SRS population. 4. (ii) (for the first drug). denoted by the constants cde. Firstly. the loglinear model for the 2 x 2 x 2 table with elements mijk is A DE DA DEA log mijk = u + uiD + u E + u k + u ij + u ik + u EA + u ijk . Therefore the relation between the loglinear parameters and. by the constant cdea. for example. the conditional odds ratio q ik j =1 is q ik j =1 DA DEA = m111 m221 = exp 4u11 + 4u111 × m121 m211 ( ) (13) In a similar way. cda and cea . This follows from the fact that the constants c. so in this situation we might expect that cdea <1. for example in the event that a related drug is known to cause a similar drug-drug interaction.2 Theory. Similar to equation (3). We will indicate shortly how these odds ratios can be obtained by means of parameters of a loglinear model. in the event that the drug-drug interaction is well known in the literature and is common knowledge of physicians and pharmacists.

The estimates of both odds ratios are biased by an unknown factor cda. For example. due to cde. the bias problem is simple in the sense that the pairs of odds-ratios in (6) and (7). in the event that an interaction is not known in the literature and is not to be suspected.0 and qˆik j = 2 = 4. cda and cea. that in the SRS population DEA there is no drug-drug-ADR interaction. i. 4.e. i. For example. This shows that the combination of both drugs coincides with an ADR in the SRS population. For clarity of exposition.0. da 3. if we find that t t qˆik j =1 = 1. generating signals of previously unknown ADRs or drug-drug interactions.e. Regarding the main purpose of SRSs.Chapter 1. however. and therefore we do not know the estimate of these odds ratios in the ADR experiencing population. in equations (6) to (11) cdea = 1. we 58 . 2. Finally. this implies that. i. in the SRS population. the ratio of odds ratios q ik j =1 / q ik j =2 = q ik j =1 / q ik j =2 . i. This is a useful result. However. After all. it holds for each pair of odds ratios that the ratio of the elements of the pair in the SRS population is unbiased for the corresponding ratio of odds ratios in the ADR experiencing population. and in (10) and (11). for q ik j =1 and q ik j=2 this factor is c4 . are biased by the same factor.e. an ADR should be more or less frequently reported. in addition to the assumption cdea = 1. the odds ratio between the first drug and the specific ADR is estimated as 1 if the second drug is absent (so then there is no relation) but as 4 when the second drug is present. Assume now that the joint underreporting effect for both specific drugs with the specific ADR can be ignored. one may assume that cdea =1.e. so we might expect cdea >1.. u ijk = 0 .2 the interaction is probably more easily reported. These equations then simplify considerably. Under this assumption we can make the following observations: 1. Because of 2. Assume now. the odds ratios are only biased because of the paired underreporting effects. this situation need not necessarily be unfavourable. in (8) and (9). since q ik j =1 / q ik j =2 = q tik j =1 / q tik j =2 both in the SRS population as well as in the ADR experiencing population the odds ratio between the first drug and the ADR is four times greater when the second drug is present than when it is absent. For example. generally there is no reason why in the event the specific combination of both drugs is used.

by substituting this in equations (8) and (9). this implies the next result: if cdea = 1 and there is no drug-drug- ADR interaction in the SRS population.29 and 3. if there is no drug-drug-ADR interaction in the ADR experiencing population. then the ratio of the odds ratios estimated from data in the SRS is equal to the ratio of the odds ratios in the ADR experiencing population (see result (iii) and the Addendum). we can use data from an SRS to assess this. and therefore.04. but these results also hold for the other odds ratios. then there will be no drug-drug-ADR interaction in the SRS population. 4. Therefore we can carry out a loglinear analysis in the SRS population (or.2. we find q ik j =1 = q ik j=2 DA = exp 4 u11 . equivalently. By working out (13) for both q ik j =1 and q ik j=2 . If in the ADR experiencing t population the interaction (u ) = 0 . We now show that the reverse also holds. Second. The factor 59 . having values 1.Influence of underreporting on odds ratios focus our discussion now on q ik j =1 and q ik j=2 (compare (8). it will also be zero in the SRS population. Since t t q ik j =1 = q ik j =2 . if the drug-drug-ADR interaction is zero in the ADR experiencing population. we find q ik j =1 = q ik j=2 4 = cda q tik j =1 4 = cda q tik j =2 . If there is such an interaction.3 Drug-drug interaction of diuretics and NSAIDs revisited In the notation of Section 4. are q ik j =1 and q ik j=2 . The factor cda reflects a possible paired underreporting of the diuretics – congestive heart failure combination. Other underreporting effects play no role. Equations (8) and j =1 (9) show us that these give biased accounts of the odds ratios q tik and q tik j =2 in the ADR experiencing population by the underreporting factors cda and cdea. the odds ratios estimated for the SRS population in section 2. and therefore t DA q ik j =1 = q ik j =2 = c da q ik j =1 = c da q ik j =2 = c da exp 4 (u11 ) -4 -4 -4 t This shows that q ik j =1 = q ik j=2 . DEA t ijk then t t DA q ik j =1 = q ik j =2 = exp 4 (u11 ) .2. Under the assumption that cdea = 1. Underreporting leads to multiplication with a factor cda t -4 (the reverse of what happens in (8) and (9)). a logistic regression) to assess whether there is a drug-drug-ADR interaction in the ADR experiencing population. (9) and (13)).

and the Appendix shows that if cdea = 1 this u ijk ˆ DEA ˆ DEA estimated in the SRS population is unbiased for the ADR experiencing population. The approach of this simple case can also be used for more complicated cases.04 / 1. with levels male and female. The odds ratios estimated in the SRS population are biased with respect to the ADR experiencing population using these constants.35 = exp 4 u ijk . for the case of one drug and one categorical covariate. However.29 and 3.Chapter 1.e. but without further information it is unclear in what direction and in what magnitude.2.1 Theory Here we discuss the assessment of interaction between a specific drug and a specific ADR in the presence of covariates one would like to control for. and it is possible to find equations similar to equations (6) to (11). the relation between diuretics and congestive heart failure is 2. csa. The situation is equivalent to the approach for section 4. As in section 4.35 times as strong when a NSAID is used compared with that NSAID is not used. It follows is 3. viz. In other words.29 = 2.2 shows that this value 2. leading to the eight constants c. We do this for a simple case only. csd. indexed by s. cda and csda. cs. Then q ik j =1 and q ik j=2 are only biased by the same factor cda. whatever size of the bias factor cda. This assumption generally is true when underreporting of congestive heart failure due to a diuretic drug is not influenced by the presence or absence of a NSAID. 5 Covariates 5. i. Let us now assume that we can ignore cdea.2 cdea reflects joint underreporting of diuretics and NSAID on congestive heart failure. in the general case this gives odds ratios that are difficult to interpret due to biases which cannot be estimated from the SRS. it is 60 . To make the exposition simple. These two factors may cause the estimates 1. cdea = 1. Some manipulation of equations from Section 3. in the ADR experiencing population the odds ratio q tik ratio q tik j =1 j =2 . with the second drug replaced by the covariate. cd.35 times as large as the odds that we can say that.04 to be biased. such as when there are more drugs and more covariates. Therefore we keep this section brief. ca. Just as in section 4 there are eight underreporting problems. as is discussed in section 3. we take as covariate sex.

We found an odds ratio of 12. gender and ADR is not equal to one. If this drug-sex-ADR interaction can be deleted from the model fitted on the SRS data. If we make the assumption that there is no joint underreporting effect. cs. cda and csda. but csa = cda = csda = 1. ca. However. then csda = 1. ADRs affecting someone’s ability to work may be more likely to be reported if they occur in men (assuming men make up a larger proportion of the work-force). this means that there is no evidence for this drug-sex-ADR interaction in the ADR experiencing population either.3. if this reporting is irrespective of the drug. if this underreporting for men is irrespective of drug and ADR. So we conclude that. 5. 61 . and the estimated odds ratio in this model is 10.2 for details). but csa = cda = csda = 1. then cs ¹ 1. sex and ADR may give rise to a different level of reporting. We give a few examples to see this. Both are more likely to be reported if they occur in women than in men. there is wide recognition that women consult their doctor more frequently than men. As in section 4. then csa ¹ 1 but csda = 1. As a last example. thus leading to increased opportunities for ADRs to be reported in women. Examples of csda ¹ 1 are difficult to imagine. generally however it is unlikely that the specific underreporting factor concerning the combination of both drug.4 for males and 10. the conditional odds ratios are biased. We now discuss the possible effects of underreporting on these results.2 Sex differences for ADRs of diclofenac revisited We now revisit the analysis of sex differences for ADRs of diclofenac discussed in section 2. and all equations simplify considerably (compare section 4.8. but ratios of these conditional odds ratios are not (see section 4. however. Again.0 for females. if this underreporting for males is irrespective of drugs and ADRs. As a second example. Consider the appearance of facial hair or the disappearance of scalp hair. The drug-sex-ADR interaction in the loglinear model was not significant. If the drug-sex-ADR interaction is included in the model. although the specific combination of both drug. cd. csa.Influence of underreporting on odds ratios clear that the odds ratios are only biased because of paired and joint underreporting effects csd.2).2. sex and ADR in SRS data. and not because of general underreporting effects c. if the assumption that csda = 1 is realistic it is useful to fit a loglinear or logistic regression model to assess whether there is an interaction between drug. then this would lead to cs ¹ 1. However.

6 Conclusion Spontaneous Reporting Systems (SRS) are plagued by underreporting problems of ADRs. do not bias these odds ratios. secondly by discussing the existence of possible underreporting effects. We approach the analysis of SRS data by firstly conducting loglinear analyses of data. 62 .Chapter 1.e. Notice that general underreporting effects cs for sex. not the size) drug . The only problem to worry about is the underreporting effect cda. and thirdly by discussing whether the odds ratios found in the SRS data are likely to be affected by these underreporting effects to such an extent that conclusions about the existence (i. It turns out that general underreporting effects for specific drugs and for specific ADRs are harmless. we have focused on the question of which types of underreporting problems plague SRS data and which types are harmless when we want to make statements about the ADR experiencing population using data from the SRS population. as well as the paired underreporting effect for sex and drug csd and for sex and ADR csa. If we assume that the joint underreporting effect csda can be neglected (i. If there is no so-called joint underreporting effect of two drugs on the ADR. then SRS data can be very useful for assessing the presence of drug interactions. and their relative size. but paired underreporting effects result in biases in the odds ratios estimated from the SRS. then the loglinear model without interaction term for sex by drug by ADR also holds in the ADR experiencing population. In particular.8 in the SRS population to 1 in the ADR experiencing population.ADR . drug and ADR. since there is no reason for anaphylactic reactions due to diclofenac to be reported more frequently in either men or women.relations would have to be modified. csda = 1). but in this example it is unlikely that this underreporting effect is so large that it will bring down the estimated odds ratio of 10. The aim of this paper was to clarify the different ways in which underreporting can play a role. cd and ca.2 These odds ratios estimated from the SRS data give possibly biased accounts of the corresponding odds ratios in the ADR experiencing population because of the paired underreporting cda of drug by ADR and the joint underreporting effect csda for sex.e.

Alert systems for post-marketing surveillance of adverse drug reactions. Egberts ACG. Acknowledgements We gratefully acknowledge the support from the Dutch Fund for Prevention (number 28. Analysis of the information in a central ADE database. Systemic signalling of adverse reactions to drugs. Estimation of underreporting. (Thesis) Utrecht University. J Biopharm Stat 1995. Finney.56:733-8. Tsong Y. Methods Inf Med. Leufkens HGM.7:39-49. Br J Clin Pharmacol 1999. 2. 1992. Statistical logic in the monitoring of reactions to therapeutic drugs.10:237-45.1:153161. Detecting drug-drug interactions using a database for spontaneous adverse drug reactions: An example with diuretics and non-steroidal anti-inflammatory drugs. International Journal of Risk and Safety in Medicine 1993. Lee EN. Schindel F. Post Marketing Surveillance 1993. Signalling possible drug-drug interactions in a spontaneous reporting system: Delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole. DJ. Haramburu F.5:95-114. References 1. Comparing reporting rates of adverse events between drugs with adjustment for year of marketing and secular trend in total reporting. 4. Stat Med 1982. American Statistician 1999. the Netherlands. 6. Van Puijenbroek EP. Methods Inf Med 1971. 12. Praus M.2632). 10. Finney DJ. Eur J Clin Pharmacol. 5.53:177202. de Koning. Waller PC. but also of the validity of drawing conclusions from the analysis of SRS data.47:689-693. 1994:76-96. 8.Influence of underreporting on odds ratios We think this paper provides a better understanding of the possible pitfalls. London: MacMillan Publishers Ltd.5:123. Schwarz S. Heerdink ER. The detection of adverse reactions to therapeutic drugs. Utrecht. 1974. Fescharek R. Amery W. Bayesian data mining in large frequency tables. GHP. 7. Van Puijenbroek EP. Finney DJ. Detection of new adverse drug reactions. Pharmacoepidemiol Drug Saf 1999:535-552. 63 . with an application to the FDA Spontaneous Reporting System (with discussion). 2000.12:2383-2393. Leufkens HGM. 3. 13. Meyboom RHB. Responding to drug safety issues. 11. A Regionalized Spontaneous Surveillance Program for Adverse Drug Reactions as a Tool to Improve Pharmacotherapy. 9.13:110. Egberts ACG. Stephens MDB. DuMouchel W. Stat Med 1993.

Br J Clin Pharmacol 1999. Martin RM. 1990. John Wiley and Sons.2 14. Van der Hofstede JW. Serum sickness-like reactions to cefaclor. Adverse cardiovascular effects of NSAIDs in patients with congestive heart failure. Wilton LV. Mosterd A. Meyboom RHB. J Clin Epidemiol 1992. Begaud B. 23. Bakker A. Treatment of onychomycosis with terbinafine. Agresti. 15. Br J Dermatol 1992:126:40-46. BMJ 1998. Kapoor KV. Meyboom. Non-puerperal lactation associated with antidepressant drug use. 64 . 21. Bakker A. (Thesis) Utrecht University. 1997:111-124. Ottervanger JP. Eland IA. Stricker BHC. Egberts ACG.17:166-180. Rakosi T. Sweden.47:329-31. Stricker BHC. In: Egberts ACG. Br J Clinical Pharmacology 1997. De Koning FH. 17. Utrecht. Moore ND. Leufkens HGM. Wili PB. 20. van Grootheest AC. 1995. 16. NSAIDs associated with increased risk of congestive heart failure in elderly patients taking diuretics. Pharmacovigilance: a science or fielding emergencies? Stat Med 2000. 22. 18. O’Brien WM. RH. Diclofenac compared with other nonsteroidal anti-inflammatory drugs. Egberts AC. Pharmacoepidemiological approached to the evaluation of antidepressant drugs. Rawlins MD. Leufkens HG. Bakker A. Feenstra J.317:119-20. Is reporting rate a good predictor of risks associated with drugs? Br J Clin Pharmacol 1999.158:1108-1112. Stricker BH. Uppsala. Herings RM. Anonymous. Mann RD. de Koning GHP. Stricker BH. 25. Evans SJW. WHO Adverse Drug Reaction Dictionary. Grobbee DE. Belton KJ. Heerdink ER. Pierfitte C.19:31993209. WHO Centre for International Drug Monitoring.Chapter 1. Adverse reactions to nonsteroidal anti-inflammatory drugs. Categorical data analysis. Baudraz-Rosselet F.48:623-627. Drug Saf 1997. the Netherlands.10:1177-1184. 26. Am J Med 1986. Transformation of a database of spontaneously reported suspected adverse drug reactions and its use as a tool in signal detection. Underreporting of suspected adverse drug reactions to newly marketed (”black triangle”) drugs in general practice: observational study. Leufkens HG. Attitudinal survey of voluntary reporting of adverse drug reactions.44:277281. Tijssen JPG. Kenzelmann R. Lagnaoui R. Arch Intern Med 1998.80:70-80. Meiners AP. 24. 19.

Likewise drug-drug interactions. examples of quantitative signal detection are given and two new approaches are introduced.4. As a second example. Chapter 2. The objective of this study is to analyse the clustering of the ADRs arthralgia.Part 2 PRACTICAL APPROACHES OF QUANTITATIVE SIGNAL DETECTION In the second part of this thesis.2 a new quantitative approach is introduced to analyse possible drug-drug interactions in datasets of SRSs.3 the influence of concomitant use of diuretics and non-steroidal anti-inflammatory drugs (NSAIDs) on symptoms indicating a decreased efficacy of diuretics is examined. datasets of SRS can be used to study the possibility of clustering of symptoms.1 deals with the analysis of a possible association between a single drug and a suspected adverse drug reaction (ADR). . as is illustrated in Chapter 2. in Chapter 2. As an example we analyse the ADR 'delayed withdrawal bleeding' resulting from a possible interaction between itraconazole and oral contraceptives in reports received by the Netherlands Pharmacovigilance Foundation Lareb. Drug-drug interactions are relatively rarely reported to spontaneous reporting systems (SRSs) for adverse drug reactions. In Chapter 2. statistically in order to determine whether or not these symptoms are interrelated. urticaria and fever during the use of the antimycotic drug terbinafine. The aim of this study is to investigate whether the risk of anaphylactic reactions being reported during the use of various non-steroidal antiinflammatory drugs (NSAIDs) is greater than with other classes of drugs and whether differences among NSAIDs exist.

.

4 and HGM Leufkens3 1 2 Netherlands Pharmacovigilance Foundation Lareb.3.1 DIFFERENT RISKS FOR NSAID-INDUCED ANAPHYLAXIS EP van Puijenbroek1. Sweden THE ANNALS OF PHARMACOTHERAPY (IN PRESS) . Utrecht. ACG Egberts2. Utrecht Institute for Pharmaceutical Sciences. the Netherlands 3 Department of Pharmacoepidemiology and Pharmacotherapy. Uppsala. Elisabeth Hospital. the Netherlands Hospital Pharmacy Midden-Brabant. 's-Hertogenbosch. the Netherlands 4 The Uppsala Monitoring Centre. RHB Meyboom1.2. Tilburg. TweeSteden Hospital and St.

adjusted for age.1 (95% CI 5.5 (95% CI 2.9) and 5. The ROR.2 (95% CI 12. The corresponding Reporting Odds Ratios from logistic regression analysis.4 (95% Confidence Interval 6. 9.5). gender and source of the reports was 9.9-12. ibuprofen and diclofenac were reported disproportionally with respect to other drugs.1-24. The aim of our study was to investigate whether the risk of anaphylactic reactions being reported during the use of various NSAIDs is greater than with other classes of drugs and if differences among NSAIDs exist. These drugs are strongly associated with anaphylactic reactions.5-11. particularly diclofenac. Results Between January 1985 and November 2000. naproxen and ibuprofen were 17.2-15. the reference group consisted of all other reports.Chapter 2. Cases were defined as reports in which anaphylactic or anaphylactoid reactions were reported.9). Methods In a case/non-case design Reporting Odds Ratios (RORs) were calculated using logistic regression analysis. The index group consisted of reports on which NSAIDs were mentioned as the suspected medication. all other reports were considered as non-cases. Spontaneous Reporting Systems (SRS) can provide valuable information regarding the occurrence of suspected adverse drug reactions (ADRs). adjusted for age.1 Summary Objective After marketing of drugs. 68 .7). gender and reporting source for diclofenac. The Netherlands Pharmacovigilance Foundation received a substantial number of anaphylactic reaction reports related to the use of nonsteroidal anti-inflammatory drugs (NSAIDs). ibuprofen and naproxen. Anaphylactic reactions associated with the use of naproxen. respectively. the Netherlands Pharmacovigilance Foundation Lareb received 76 cases concerning anaphylactic reactions to NSAIDs. Conclusion The results of this study strengthen previous findings concerning the relative high risk of developing an anaphylactic reaction during the use of NSAIDs.

Furthermore. The aim of our study was to investigate whether the risk of anaphylactic reactions being reported during the use of various NSAIDs is greater than with other classes of drugs and if differences among NSAIDs exist.. Data concerning the 69 .4-20 it is not clear whether there are differences between individual agents. Rarely blood cell or liver function disorders can occur. limited evidence is available concerning the relative safety of these drugs. however. The most frequently occurring adverse drug reactions (ADRs) related to NSAID use concern the gastrointestinal tract and kidney. additional information concerning the clinical details of the report is retrieved.5 Although these potentially serious ADRs are reported with various NSAIDs. varying from urticaria and angioedema to anaphylactic reactions.4 Also in spontaneous reporting systems (SRS) for adverse drug reactions NSAIDs were frequently associated with anaphylactic reactions. have shown that among patients admitted to a hospital because of anaphylactic reactions.3 NSAIDs have also been associated with acute allergic reactions. but were reported because of the seriousness of the reaction involved or concern of the reporting health professional. NSAIDs were a frequent cause.2 NSAIDs. In this context a substantial number of reports of anaphylactic reactions related to the use of NSAIDs were received by the Netherlands Pharmacovigilance Foundation Lareb. Special attention is paid to the description of the ADR. Reports that are received are subject to review by qualified assessors. SRS can provide valuable information regarding possible new ADRs. After new drugs reach the market. are also associated with serious and non-serious complications.NSAIDs and anaphylactic reactions Introduction Non-steroidal anti-inflammatory drugs (NSAIDs) have a prominent place in the treatment of pain and arthritis.1. ADRs are reported by physicians and pharmacists on a voluntarily basis. Methods Setting The Netherlands Pharmacovigilance Foundation Lareb maintains the national spontaneous reporting system (SRS) on behalf of the Dutch Medicines Evaluation Board. With respect to the occurrence of anaphylactic reactions. datasets of SRS contain information about ADRs that are already known. Van der Klauw et al. If needed.

30 Characteristic symptoms of such a reaction may involve the skin. pharmacological and therapeutic properties. The ROR is defined as the ratio of the exposure odds among reported cases to the exposure odds among reported non-cases. Both preferred terms are considered to be a manifestation of an anaphylactic reaction. This meeting is attended by assessors with specific experience on the field of spontaneous reporting.31 In the event that allergic symptoms concerning two or more organ systems were reported.23 For signal detection every new report is reviewed on a regular basis in a discussion meeting in which the reported association between the suspected drug and ADR is assessed.Chapter 2. anaphylactic reaction. anaphylactoid reaction.29 Design The analysis included all reports received by Lareb between January 1st 1985 and November 1st 2000 in which data concerning age and gender of the patients were available. anaphylaxis and red neck syndrome. simultaneously involving several organ systems. and subsequently filed in a database. gastrointestinal system or neuropsychological symptoms.21 In the ATC classification system. In addition to this ‘case-by-case’ analysis. the ADRs were classified as an 70 . In a case/non-case design. because NSAIDs are rarely used in children in the Netherlands.24.24-29 We analysed the association between NSAIDs and anaphylactic reactions using a case/non-case design and expressing the strength of the association as the ADR Reporting Odds Ratio (ROR).24.29 Cases were defined as all reports coded with the WHO preferred term ‘anaphylactic shock’ or ‘anaphylactoid reaction’. the drugs are divided into different groups according to the organ or system on which they act and their chemical.1 suspected adverse drug reaction and the drugs involved are coded using the WHO adverse drug reaction terminology and the Anatomical Therapeutic Chemical (ATC) classification system respectively. which can be considered as an acute systemic adverse reaction. respiratory system. the extent to which a possible ADR is reported for a given suspected drug can be analysed statistically. RORs were calculated by means of logistic regression analysis. cardiovascular system. Patients younger than 10 years of age were excluded.22 The primary goal of an SRS is to give an early warning of a possible causal relationship between an ADR and a drug of which the relation was previously unknown or incompletely documented. These codes encompass the WHO included terms anaphylactic shock.

such as misoprostol. 841 reports were excluded because the patient was younger than 10 years. were expressed as point estimates with corresponding 95% confidence intervals. For all statistical analyses.910 reports. reports were classified as anaphylactic shock. NSAIDs can be used in combination with other drugs. Because of this large volume. the Netherlands Pharmacovigilance Foundation Lareb received reports on 23 different ATC codes concerning NSAIDs or combinations of NSAIDs and other drugs. Results Between January 1985 and November 2000. All cases referred to oral dosage form. if shock-like symptoms were reported and the blood pressure was known to have dropped. adjusted for age and gender of the patients. Additionally. Of these. except for one case concerning 71 . Theoretically one of the other drugs in the combination could be responsible for the anaphylactic reaction. The index groups consisted of reports on which an NSAID (ATC code beginning with M01A) was mentioned as the suspected medication.0 was used. 76 concerned a possible relationship between an NSAID and an anaphylactic reaction. RORs. however. either physician or pharmacist. The reference group consisted of all reports in which no NSAID was mentioned as the suspected medication. bronchospasm or shock. SPSS 10. In case combinations of NSAIDs and other drugs have been registered in the Netherlands. a total number of 28. the ADRs were coded as such. In the event where next to the NSAID another suspected medication is reported. If symptoms pertaining to only one system and organ class were reported. the reports are also regarded as reports on the NSAID involved. while 252 reports were excluded because gender or age of the patients were not reported. like urticaria. Since 1985. Of the remaining 26. year of reporting and source of the reports. results of the analysis of all separate NSAIDs were restricted to those drugs from which more than 100 reports were received. and subsequently were treated as non-cases. reports on these drugs were regarded as reports on the NSAID present in the combination concerned.003 reports of suspected ADRs were received. All other reports were considered as non-cases.NSAIDs and anaphylactic reactions anaphylactic reaction.

2) 448 (1.s. ** Pearson Chi square.7) 175 (0. Table 1.9) p<0.7) (78.5) 17.6) 121 149 (63.7 (15. *** in comparison with the number of pharmacists Among the cases.4) Non-cases (n=26.2 (17. Table 1 shows the distribution of age.7) 296 (1.1) 114 (60.9) 7 (3.8) 15 (7. n.1) 866 (3. gender and source of the reports among cases and non-cases Cases (n=190) n (%) Mean age (years) (standard deviation) Number of females (%) Number of reporting physicians (%)*** 52.9) 72 .05** *Student-t test. Distributions of age.** p<0.690 (64.420 (9.7) 1 (0. gender and source of the reports among cases and non-cases.720) n (%) 2. Table 2.294 18.s. Distribution of cases and non-cases among various NSAIDs and other suspected drugs Cases (n=190) n (%) All NSAIDs Diclofenac Naproxen Ibuprofen Piroxicam Ketoprofen Indomethacin Other NSAIDs Other suspected drugs 76 (40) 51 (26.0) Non cases (n=26.Chapter 2. the age of patients was statistically significant higher and they were also more frequently reported by physicians as compared with pharmacists.720) n (%) 47.1) 180 (0.7) (69.300 (90.01* n.5) 0 0 2 (1. The distribution between cases and non-cases for different NSAIDs are shown in table 2.1 the use of diclofenac (injection).7) 105 (0. not statistically significant.4) 350 (1.3) 24.

7 (5.2 (12.9) 5.9) 1.5) 1.5-11.0-9. For diclofenac. On ketoprofen and indomethacin no cases were reported.2 (95% CI 12.1-24.2 (0.7) 1. for naproxen 9. Table 3 shows the results of the univariate and multivariate analysis.4 (0.1-24.3) 5.0 (2. gender and source of the reports and the number of cases for different NSAIDs are shown in Figure 1. Anaphylactic reactions were disproportionally more reported for ibuprofen. there were 2 additional cases (sulindac and nabumetone).9) 1. anaphylactic reactions were reported on diclofenac. implying no ROR could be calculated for the latter drugs.1-12.4 (6.2 (0.2-8.2-15.9).5 (2.3-4. the RORs with corresponding confidence intervals.0) 12. Table 3. the adjusted ROR was 17. The total number of reported reactions.7) 17.5). adjusted for age.9) Multivariate analysis ROR (95% CI) 9. Results of univariate and multivariate logistic regression analysis Univariate analysis ROR (95% CI) All NSAIDs Diclofenac Naproxen Ibuprofen Piroxicam Other NSAIDs 6.3-10.9-12.0-17.2 (0.6) 7.3-5.5-11.2-15. adjusted for age.1 (4.5 (95% CI 2.7).2-11. 73 . naproxen. The ROR.1 (5. Among the NSAIDs on which less than 100 reports were received.5) 9.9) and for ibuprofen 5. naproxen and diclofenac. ibuprofen and piroxicam.1 (95% CI 5.5 (9.4 (95% confidence interval 6. gender and source of the reports was 9.9-12.NSAIDs and anaphylactic reactions Among the NSAIDs on which more than 100 reports were received.6) NSAIDs were strongly associated with reports of anaphylactic reactions.

Chapter 2.1
100 60 reporting odds ratio number of reported cases 50 reporting odds ratio (95% CI)

30

1

20

10

0,1 diclofenac naproxen NSAID ibuprofen piroxicam

0

Figure 1. Results of the analysis of anaphylaxis associated with various NSAIDs versus all other reported cases. The total number of reported cases and the Reporting Odds Ratios (semi-logarithmic scale) are shown, adjusted for gender and age of the patient, year of reporting and source of the reports, with corresponding 95% confidence intervals for diclofenac, ibuprofen, naproxen and piroxicam.

Discussion
The results of our study shows that risk of an anaphylactic reaction being reported is increased during the use of NSAIDs, notably on diclofenac, naproxen and ibuprofen. Among the NSAIDs, diclofenac appears to have the highest reporting rate. Interpretation of the results originating from spontaneous reporting systems should be done with great care. The method applied provides quantitative information about the extent of the reported associations between reported suspected drugs and ADRs with respect to other reports sent to the SRS. The rationale of the case/non case design is that the proportion of ADRs is relatively constant over time and for this reason, the reference group can be considered to be a measure of the ‘background frequency’ of the suspected ADRs. This implies that interpretation of the quantitative results based on data sets of SRS requires specific knowledge of the composition of the database. For this reason spontaneous reporting systems are primarily used for signal detection purposes and not for hypothesis testing. An estimation of the actual incidence of the ADR 74

number of reports

10

40

NSAIDs and anaphylactic reactions

in populations using the suspected drug cannot be made. Prescription data cannot be used to estimate the actual use of the drugs since some NSAIDs, like ibuprofen and naproxen, are also available without prescription in the Netherlands. A case cohort study in the Netherlands showed previously that diclofenac in particular was among the most frequent causes of anaphylactic reactions leading to hospital admission, the relative risk of anaphylaxis relative to all other drugs being 9.5 (95% CI 3.7-24.5).4 The results of our study are in accordance with these findings, but furthermore demonstrate that anaphylactic reactions are also reported disproportionally for naproxen and ibuprofen. Anaphylaxis is an immediate (type I) hypersensitivity reaction to an allergen, caused by its rapid cross-linking with specific IgE on tissue mast cells and peripheral blood basophils. It requires previous exposure to the foreign antigen. An anaphylactoid reaction, however, is not an IgE mediated response but, similarly, involves inflammatory mediators to be released from mast cells and basophils. This activation of immune cells may occur both directly and as the result of disturbances in arachidonic acid metabolism and immune complexmediated activation of complement.32 These non-IgE mediated reactions, or anaphylactoid reactions, do not require previous exposure and may also be caused by NSAIDs.20,32,33 Although pathophysiology differs to a certain extent, anaphylactic reactions and anaphylactoid reactions share the same clinical features and cannot be distinguished on clinical grounds.32,33 It is unclear whether anaphylactic or anaphylactoid reactions predominate. For this reason, no distinction could be made between both anaphylaxis and anaphylactoid reactions in this study. Among reported cases of an anaphylactic reaction during the use of a NSAID, the reaction was fatal in one case. This patient, a 62-year-old female, used diclofenac. There were four fatal cases among the patients in whom an anaphylactic reaction was reported in association with another drug. Fatal cases associated with an anaphylactic reaction were not statistically significant between NSAIDs and other drugs (Fisher’s exact test p>0.05). NSAIDs can be subclassified with respect to their chemical structure. Diclofenac, together with tolmetin and ketorolac belong to the heteroaryl acetic acids, ibuprofen, naproxen, flurbiprofen, ketoprofen, fenoprofen and oxaprozin to the arylpropionic acids.34 When NSAIDs that have a similar chemical structure are grouped, heteroaryl acetic acids have an adjusted ROR of 19.7 (95% CI 13.8-

75

Chapter 2.1

28.1) while the adjusted ROR for the arylpropionic acids was 6.7 (95% CI 4.210.6). This suggests that the risk for an anaphylactic reaction is higher with the use of heteroaryl acetic acids. In our study, of the 166 reports on the combination of diclofenac and misoprostol 2 reports were characterised as anaphylactic reactions and were labelled as reports on diclofenac. The distribution of cases and non-cases among diclofenac and the combination of diclofenac/misoprostol were similar (Fisher’s exact test p>0.05), showing that misoprostol was not likely to have an additional effect on the chance of an anaphylactic reaction being reported. In the event that combinations of NSAIDs and other drugs are considered to be non-cases, the ROR adjusted for year of reporting, age, gender and source of the reports for diclofenac referred to all reports on non NSAIDs was 17.0 (95% CI 11.7-24.6). Next to the suspected NSAID another suspected medication has been reported in 95 cases. In our study, these reports were not excluded, but regarded as reports on the NSAID involved. In two of these reports an anaphylactic reaction has been reported. The distribution among cases and non-cases between the reports in which two or more suspected medications were reported did not differ from reports were only an NSAID was reported (Fisher’s exact test p>0.05). This suggests that the anaphylactic reactions where likely to be caused by the NSAIDs involved. When reports with other suspected drugs beside NSAIDs and were regarded as non cases, the adjusted ROR for diclofenac was still 20.0 (95% CI 13.7-29.2). Non-selective reporting of either the suspected drug or the suspected ADR has a similar effect on numerator and denominator of the ROR. For this reason, nonselective reporting has no influence on the magnitude of the ROR.24 Selective reporting on the combination of drug and ADR, reflecting the concern of health care professionals involved, however, may influence the ROR. We believe that this non-differential bias, for instance precipitated by specific media attention to anaphylactic reactions on certain drugs, is not likely to have occurred. Another confounding factor can be the intermittent use of some types NSAIDs that enhances the chance of sensitisation. Elevated concentrations of leukotrienes can be found in tissues or exudates in several diseases, including asthma, diverse allergic states psoriasis, spondyloarthritis, and gout.35 It is unclear, if these elevated concentrations also enhances the chance for anaphylaxis and the indication for use subsequently

76

the results of our study lend support to previous findings concerning the relative high risk of developing an anaphylactic reaction with NSAIDs. however. did not differ significantly from other reports in the database. and that the connection is particularly strong for diclofenac. in the event of concomitant use of beta-blocking drugs.23. occur rarely and are not primarily related to the main pharmacological action of the drug. ibuprofen and naproxen.37 Cross-hypersensitivity among NSAIDs may occur. In an additional analysis we looked for the existence of possible risk factors among the users of NSAIDs. Unfortunately it was not possible to take intermittent use or the indication for use into account in this study. These ADRs are characterised by their unpredictable nature. and may act as a confounder.3 and a previous history of an anaphylactic reaction on an NSAID may be a reason to restrain from prescribing another NSAID in the future. Channelling refers to the phenomenon that a drug is prescribed preferably for a specific group of patients with certain recognised risk factors. Anaphylaxis is a rare ADR. however. 77 . Statistical analyses of reporting patterns support the view that diclofenac carries a higher risk for anaphylactic reactions than other NSAIDs. since this information is only available for a limited number of reports. a previous history of gastric complaints may act as a confounder. gender and use of betablocking agents among the users of NSAIDs. this condition is rare and therefore the risk of channelling is low. can be considered ‘type B’ or ‘idiosyncratic’ effects. Conclusion Spontaneous ADR reporting data in the Netherlands show that anaphylactic reactions are more frequently reported in association with NSAIDs than is expected form the background frequency in the database. Even if the risk for developing this ADR is increased by NSAID use. however. Anaphylactic reactions.39 Theoretically this may lead to a greater reporting rate. In the context of studying ADRs.36 For example in studying the risk for peptic ulcers among patients using NSAIDs.38 Finally. Age. the presence of contraindications is rather commonly predictive of the outcome criteria for ADRs. signs of an anaphylactic reaction may become more severe. Presumably. Although an estimation of the actual incidence of the ADR in populations using the suspected drug cannot be made.NSAIDs and anaphylactic reactions might have been a confounding factor.

Assessment and treatment of neuropathic cancer pain following WHO guidelines. Cistero A. Lehmann KA.47:576-8. Reports to a specialty-based system for spontaneous reporting of adverse reactions to drugs. Garcia-Moll M. Geli A. Czerniawska-Mysik G. Isr J Med Sci 1994. 5. Clinical patterns of hypersensitivity to nonsteroidal anti-inflammatory drugs and their pathogenesis. Cost WS. Anaphylaxis from tolmetin. Lehmann KA. 1977. Coronary artery spasm and acute myocardial infarction in naproxen-associated anaphylactic reaction. Loick G. Ibuprofen urticaria. A population based case-cohort study of drug-induced anaphylaxis. Nonsteroidal anti-inflammatory intolerance. 11. Stricker BH. Anaphylactic reaction to oral diclofenac sodium sustained-release tablet. 8. An expanded profile of cutaneous reactions to nonsteroidal antiinflammatory drugs. Arzneimittelforschung 1991.60:276-84. Wilson JH.79:15-20.1 the actual risk is still low. Settipane GA.60:107-10. Hertel D. Arch Intern Med 1980.141(3 Spec No):328-32. More study is needed to determine the actual incidence of NSAID-induced anaphylaxis. Biscarini L. Kimura K. Ueda F. 10.41:235-9. Shelley WB. Lynch J. Shelley ED.30:909-10. Drug-associated anaphylaxis: 20 years of reporting in The Netherlands (1974-1994) and review of the literature. Fujisawa H. Hermoni D. Szczeklik A. Arch Intern Med 1981. Bayes de Luna A. Adverse reactions of aspirin and related drugs. Mahmoud SF. Sabatowski R. Lleonart R . Bigby M. Milman U.Chapter 2. J Am Acad Dermatol 1987. Grond S. N Engl J Med 1980. 4. Meyler's Side Effects of Drugs. In: Dukes NMG. Yokota M. 15. 12. J Allergy Clin Immunol.252:1433-7.26:1355-63. Non-steroidal anti-inflammatory drugs. Valkenburg HA. Mechanism of anti-inflammatory action of etodolac. Wilson JH. Shibata Y. Ahmad S. Amsterdam: Elsevier Science. 9. 13. van der Klauw MM. 14. Nishimura T. Gryglewski RJ. Pain 1995. Guindo J. Motonaga A. Goldsmith DP. Grond S. Miki N. 6. 2. JAMA 1984. van der Klauw MM. Aronson JK (eds). Br J Clin Pharmacol 1993. Meuser T. Stern RS. Stricker BH. Inoue K. Clin Exp Allergy 1996. An anaphylactic reaction to tolmetin. Validation of World Health Organization Guidelines for cancer pain relief: a 10-year prospective study. Radbruch L. Alkhawajah AM. Eifawal M. 78 .140:1105-6. References 1. 2000:246-309. Herings RM.35:400-8. Zech DF.303:1417.63:65-76. Urias S. Forensic Sci Int 1993. Moore ME. Allergy 1992. 3. Fatal anaphylactic reaction to diclofenac. 7.17:1057-8. Pain 1999.

Anonymous. Tijssen JGP. Diclofenac compared with other nonsteroidal anti-inflammatory drugs. 17. Olsson S. 28. Bate A. Moore N.344:109-13.whocc. Rosenfeld JB. Bruppacher R. 27. 1995. WHO Centre for International Drug Monitoring. Ruddon RW. Reporting adverse drug reactions. Brankowski Z. Edwards IR. Limbird LE. Methods Inf Med 1971. Goldman DW. N Engl J Med 1996. Lancet 2000. 30. Br J Clin Pharmacol 1997. and management. Ann Allergy 1992. Levy JH. 1999. 29.45:1177-84. 21. 31. Analgesics-antipyretics and antinflammatory agents. Orre R. O'brien WM.69:87-96. Uppsala. Geneva: Council for International Organizations of Medical Sciences (CIOMS). 24. July 1st 2001. 1999:621.80:70-80. Goodman & Gilman's The Pharmacological Basis of Therapeutics. Anaphylaxis and anaphylactoid reactions. McKenna F. Finney DJ. Int J Clin Pract 2000. Bihari D. Adverse reactions to nonsteroidal anti-inflammatory drugs. Garty M.nmd. Lindquist M.4:79-84. Meyboom RHB. Kay AB. Brogden RN. Andrejak M. Edwards IR. 26. 20. de Koning GHP. Goetzl EJ.no/.NSAIDs and anaphylactic reactions 16. Anaphylaxis and coronary disease. 19. Murrant T.286:1861. In: Hardman JG. N Engl J Med 2001. Goodman Gilman A (eds). Leufkens HGM. Pakes GE. Avery GS.20:24-48. 33. Ollagnier M. Reports of hypoglycaemia associated with the use of ACE inhibitors and other drugs: a case/non-case study in the French pharmacovigilance system database.335:1925. A Bayesian neural network method for adverse drug reaction signal generation. Definitions of Terms and Criteria for their Use. WHO Adverse Drug Reaction Dictionary. Shanewise JS. De Freitas RM. 34. Dux S. Kreft-Jais C. 79 .10:1-8.44:513-8. 22. Allergy and allergic diseases: Allergic diseases and their treatment. Egberts ACG. The Design and Logic of a Monitor of Drug Use. Immunopathogenetic roles of leukotrienes in human diseases.356:1255-9. 25. Serum sickness-like reactions to cefaclor. Diclofenac sodium: a review of its pharmacological properties and therapeutic use in rheumatic diseases and pain of varying origin. Noblet C. Anaphylactic shock induced by diclofenac. 32.54:3228. Heel RC. Non-puerperal lactation associated with antidepressant drug use.45(Suppl 1):24-30. Molinoff BP. New York: McGraw-Hill. Finney DJ. Yunginger JW. Aronson JK. WHO Collaborating Centre for Drug Statistics Methodology. Am J Med 1986.54:315-21. J Clin Epidemiol 1992. Statistical aspects of monitoring for dangers in drug therapy. Br J Clin Pharmacol 1997. diagnosis. J Clin Immunol 1984. Eur J Clin Pharmacol 1998. Speight TM. Lansner A. Stricker BHC. Groslop I. Crusius Ieal. 35. Haramburu F. Efficacy of diclofenac/misoprostol vs diclofenac in the treatment of ankylosing spondylitis. 18.18:77-98. 23. Bakker A. Insel PA. Anaphylaxis. Adverse drug reactions: definitions. http://www. Payan DG. Begaud B. J Chron Dis 1965. Drugs 1980. Br Med J 1983.44:277-81. Drugs 1993.

Lindquist M.Med 1983. Channelling of controlled release formulation of ketoprofen (Oscorel) in patients with history of gastrointestinal problems. 80 . Bakker A. Stat.46:428-32. The need for randomization in the study of intended effects. Petri H.1 36. Miettinen OS. 38.22:415-23.Chapter 2. Urquhart J.81:1-5. Risk of anaphylaxis in patients receiving beta-blocker drugs. 39. 37. J Epidemiol Community Health 1992.2:267-71. Leufkens HG. J Allergy Clin Immunol 1988. An ABC of drug-related problems. Egberts ACG. Stricker BH. Toogood JH. Meyboom RHB. Drug Saf 2000.

the Netherlands BRITISH JOURNAL OF CLINICAL PHARMACOLOGY 1999.2.3. the Netherlands 3 Hospital Pharmacy Midden-Brabant. the Netherlands Utrecht Institute for Pharmaceutical Sciences. TweeSteden Hospital and St. Elisabeth Hospital.47:689-693 . RHB Meyboom1. and HGM Leufkens2 1 2 Netherlands Pharmacovigilance Foundation Lareb. Department of Pharmacoepidemiology and Pharmacotherapy. 's-Hertogenbosch. Utrecht.2 SIGNALLING POSSIBLE DRUG-DRUG INTERACTIONS IN A SPONTANEOUS REPORTING SYSTEM: DELAY OF WITHDRAWAL BLEEDING DURING CONCOMITANT USE OF ORAL CONTRACEPTIVES AND ITRACONAZOLE EP van Puijenbroek1. Tilburg. ACG Egberts2.

82 . Methods To illustrate this method. In addition. we analysed the adverse drug reaction ‘delayed withdrawal bleeding’ resulting from a possible interaction between itraconazole and oral contraceptives in reports received by the Netherlands Pharmacovigilance Foundation Lareb between 1991 and 1998.2 Summary Objective In spontaneous adverse drug reaction reporting systems. This approach might be a promising tool for the development of procedures for automated detection of possible drug-drug interactions in spontaneous reporting systems. Conclusion Since spontaneous reporting systems can only generate signals concerning possible relationships. nor itraconazole. age and source of the reports. We describe a method for detecting possible drug-drug interactions using logistic regression analysis to calculate ADR Reporting Odds Ratios. for a delayed withdrawal bleeding in women who used both drugs concomitantly compared with women who used neither oral contraceptives. was 85 (95% Confidence Interval 32-230).Chapter 2. this association needs to be analysed by other methods in more detail in order to determine the real strength of the relationship. The ADR Reporting Odds Ratio. adjusted for year of reporting. Results In total 5. special attention is needed for the detection of adverse drug reactions resulting from possible drugdrug interactions.503 reports were included in the study. there is a growing need for methods facilitating the automated detection of signals concerning possible adverse drug reactions.

19 reports of delayed withdrawal bleeding in women using OCs have been received. Because of the unexpected high number of reports concerning a delay in the withdrawal bleeding where OCs and itraconazole were used concomitantly. Special cases are those ADRs resulting from drug-drug interactions.1 Since 1996 Lareb is the national centre in support of the Medicines Evaluation Board. In 10 of these reports OCs and itraconazole were used concomitantly. there is a growing need for a computerised system that aids in this process. These signals are currently being detected by a systematic review of all individual reports that are received in a certain time span by a team of experts in this field. Rarely the presence of a suspected interaction is mentioned. ‘ADR Reporting Odds Ratios’ can be calculated. Reports to Lareb usually concern an event ascribed to only one suspected drug. which was previously described for analyzing reported ADRs in a case non-case design.3. Nine of these reports concerned OCs containing desogestrel. One of the major goals of this system is the timely detection of possible new adverse drug reactions (ADRs) and interactions.5. Clinical details of this possible interaction were recently described as a case series on the basis of reports submitted to Lareb. With the increasing number of reports. Since the approval for marketing of itraconazole in the Netherlands.6 The objective of the 83 . The detection of possible interactions is based on the following concept: when a suspected ADR is reported more frequently on the combination of two drugs as compared with the situation where these drugs are used in absence of one each other. This concept. our analysis was focussed on this association.4 is applied for the analysis of possible interactions in this study. which makes the detection of possible drug-drug interactions difficult.Drug-drug interaction between itraconazole and oral contraceptives Introduction The Netherlands Pharmacovigilance Foundation Lareb maintains the spontaneous adverse drug reaction reporting system in the Netherlands. and used for the detection of signals. this association might indicate the existence of a drug-drug interaction. contingency tables can be constructed and subsequently disproportional drug-ADR combinations are identified. Based upon all records in the database. One other woman used a cyproterone-acetate containing OC and experienced a delayed withdrawal bleeding. In 1974 Finney described the concept of ‘reaction proportion signalling’.2 By means of these contingency tables.

if necessary. -Age of the patients. Also detailed information is requested about all drugs that were used at the time the reaction occurred. which was licensed in the Netherlands in 1991. On 1 January 1998. recoded. In all other cases. In order to make an appropriate comparison between women who used itraconazole.2 present article was to illustrate this method by analyzing a possible interaction between oral contraceptives (OCs) and the antifungal agent itraconazole. One report of a woman reporting a ‘delayed 84 . the ADR was labelled as ‘delayed bleeding’. -Reported reactions: All menstruation disorders and disorders of the withdrawal bleeding in the database were checked and. only reports with a reporting date later than January 1991 were included. In February 1998 we published an article concerning the possible association between the use of itraconazole and oral contraceptives in one of the major Dutch Journals. also some cases of ‘amenorrhoea’ or ‘oligomenorrhoea’ were recoded as ‘delayed bleeding’. and subsequently reports are filed in the Lareb database. If the WHO preferred term ‘menses onset delayed’ or ‘missed withdrawal bleeding’ was recorded on a report form. by means of a limited number of questions on a special report form. information such as ADR(s) and indication for use is coded. This database can be used for further evaluation and analysis of the data. Information is collected about the patient and the suspected ADR.5 To avoid the risk of reporting bias due to this publication. the ADR was coded as ‘other ADRs’.Chapter 2. and women who did not use this antifungal drug. After being received by Lareb. Methods Source Health professionals send reports to Lareb on a voluntary basis. it consisted of a total of approximately 19. Selection of cases and non-cases The domain was restricted to reports of women between the age of 15 and 50 years.500 reports. only reports received before January 1998 were used. After checking the original description of the ADR. The following data were used in the study: -Source: A report might originate from either a physician or a pharmacist.

The ADR Reporting Odds Ratios were calculated as (a:c)/(b:d) (Figure 1) (4). and calculated by logistic regression analysis. These ratios are defined as the ratio of the exposure odds of reported cases of ‘delayed bleeding’ to the exposure odds of other ADRs. Reported suspected ADR 'Delayed bleeding' No 'delayed bleeding' reported reported Medication Index group Reference group a c b d Figure 1. ADR Reporting Odds Ratios were calculated. Analysis For the comparison of index and reference groups. The reference group consisted of patients who used neither OCs. -Year of reporting. because a miscarriage could not be ruled out. -Reports of women who used itraconazole. Two by two table used for the calculation of ADR Reporting Odds Ratios. In order to 85 . nor itraconazole. all drugs were taken into account in the present study. Cases were defined as patients who reported a ‘delayed bleeding’. but no OC at that time. -Although on the forms received by Lareb. while noncases consisted of all other reports. Odds ratios were expressed as point estimates with 95% confidence intervals. but no itraconazole at that time. the health professional makes a distinction between suspected and non-suspected medication. -Reports of women who used itraconazole and OC concomitantly. For the comparison of the age of the patients between the three groups. ADR Reporting Odds Ratios were adjusted for source. analysis of variance was conducted. The reports were divided into three index groups: -Reports of women who used OCs. -Presence of itraconazole among the drugs used. -Presence of OCs among the drugs used.Drug-drug interaction between itraconazole and oral contraceptives withdrawal bleeding’ in association with an initial transient positive pregnancy test was coded as ‘other ADR’. year of reporting and age.

16 women used itraconazole but no OCs. If only itraconazole was used none out of 16 patients reported the occurrence of this effect. All statistical analyses were performed using SPSS 8.2 77.7 29. 86 . The other 13 patients in this group did not mention this side effect.998 5. Results A total number of 5.3 35. If only OCs were used.001 Chi-square). There was a difference between these groups concerning the age of the patients (p<0.1 71.466 women used OCs but no itraconazole.5 Mean age (years) Source (percentage reporting by physicians) Number of reports 31.503 reports were included in the study.6 1.2 analyze the source of the reports a Chi-square test was used. Delayed bleeding was reported by 10 patients who used itraconazole and OCs concomitantly. source and number of reports for index and reference groups that were studied are presented in Table 1.6 Total Only Itraconazole as itraconazole well as oral contraceptive 40. Age and source of reporting of index groups. as well as the source of the reports ( p<0.503 Differences in reported ADRs between the three groups are shown in Table 2.8 43.0.466 16 23 3.7 56.5 75. reference group and total group A Only oral contraceptive Index groups B C Reference group Neither itraconazole nor oral contraceptives 37. Table 1. and 23 women used both drugs. Mean age.001 analysis of variance). 9 out of 466 patients mentioned ‘delayed bleeding’. Altogether 1.Chapter 2.

A special case is the detection of interactions. the ADR Reporting Odds Ratio for the risk of a ‘delayed bleeding’ in reports of women who used itraconazole and OCs concomitantly compared with the reports of women who used neither of these drugs was 85 (95% CI 32-230).8). Since this procedure is time consuming and the number of reports is increasing. there were no cases.2 (0. Odds ratios tend to overestimate the relative risk.4-1. age and source The ADR Reporting Odds Ratios and 95% confidence intervals for all comparisons are shown in Table 2. since they are often not reported.4-1. However.7) 0 (0-∞) 153 (60-389) Adjusted* OR (95% CI) 1 0. The level of the odds ratio however. all reports are individually reviewed and detection depends upon the skills and memory of the professionals involved. Adjusted for year of reporting. When only itraconazole was used. is of less importance. age and source. the use of an odds ratio offers 87 .464 (95% CI) 3978 1457 16 13 1 1. The analysis of possible drug-drug interactions is based on the concept that a suspected ADR is more often reported on the combination of two drugs compared to the situation where either of these drugs has been used in absence of the other one. Discussion A major function of spontaneous reporting systems is generating new ‘signals’. age and source in the group where only OCs were used was 0. there is a growing need for an automated method that facilitates detection of ADRs. The ADR Reporting Odds Ratio adjusted for year of reporting. Cases n=39 No itraconazole or OC OC/ no itraconazole Itraconazole/ no OC Concomitant use of itraconazole and OC 20 9 0 10 Controls crude OR n=5. but provides an indication of the degree of disproportionality.Drug-drug interaction between itraconazole and oral contraceptives Table 2.7 For this purpose.8 so interpretation is more difficult. Reporting Odds Ratios (OR) concerning the occurrence of a ‘delayed bleeding’ of reference and index groups.6-2.8) 0 (0-∞) 85 (32-230) *adjusted for year of reporting.8 (95% CI 0.8 (0.

The use of a computer in the detection of adverse drug interactions was already proposed by Amery. OC= oral contraceptive) The estimates resulting from this model does not differ from the previous method. Possible drug-drug interactions might be analysed in different ways.01). S= source. Secondly. A likelihood ratiotest can be used to evaluate the effect of adding this term.9 In his approach the computer is instructed to look for pairs of a specific concomitant drug with a specific ADR. In our approach. as was shown in this study.Chapter 2. These index groups can be compared with a reference group in which none of the suspected drugs is used. The fraction of patients showing the ADR under consideration is subsequently compared with the frequency of this ADR in all database patients. and furthermore offers a possibility of analysing the use of interaction terms in more detail.2 advantages since concomitant variables may be analysed by logistic regression analysis. Given the fact that the difference in –2 log likelihood follows a chi-square distribution with one degree of freedom. three index groups can be distinguished. the analysis could also have been done by using a logistic model in which next to the use of both itraconazole and oral contraceptives as covariates subsequently an interaction term for the concomitant use of both drugs is added. In our example the logistic model for calculating the adjusted odds ratios would then look like: log(odds) = b0 + b2A + b3S + b4Y + b5I + b6OC + b7OC*I (A = age. two groups where both drugs are used in absence of one each other and one group where both drugs are used concomitantly. In the first place. As preliminary experiments indicate. Y= reporting year. I= Itraconazole. The latter method can also be used for the automatic detection of drug interactions. The likelihood of a reaction being due to a suspected interaction is increased if the 95% confidence intervals of the group in which both drugs are used and the other two index groups are mutually exclusive. the use of logistic regression analysis offers the possibility for correcting for covariates. analysing the effect of adding an 88 . This model can be compared with the model in which no interaction term (b7OC*I) is present. adding the interaction term is this case was statistically significant (p< 0.

is seldom painful. This implies that there is little reason for the use of NSAIDs. For instance. reporting bias might occur. In our example selection bias might be present in the index groups where OCs are used in contrast to the index group where no OCs are used. Although it is unlikely that self-medication could have acted as a confounder in these reports. since certain ADRs are more likely to be reported than others. mycotic infections. Increased reporting due to a recent introduction of a drug or attention for an adverse drug reaction in the media does not necessarily influence the Reporting Odds Ratio. For instance analgesics such as naproxen may cause a delay or interrupt menstruation in women who are not taking oral contraceptives. a third drug may act as a confounder when it is associated with one of the drugs or the concomitant use of both suspected drugs. also used NSAIDs. a delay in the onset of the menstruation with a couple of days might be reported less likely because of acceptance of the normal physiological irregularity of the menstruation women might experience. but in the Netherlands these drugs are also available for self-treatment. A delay in the onset of the withdrawal bleeding during the use of OCs.Drug-drug interaction between itraconazole and oral contraceptives interaction terms by a likelihood ratiotest is a promising tool in the automated detection of drug interactions and will be further explored. Internal validity In general. None of the patients in the group who used both drugs. however. selection bias might be present. as with the analysis of adverse drug reactions. might be experienced as a more unusual situation.10 NSAIDs were not prescribed to any of the case patients reported to Lareb. since non selective reporting bias has a similar effect on both numerator and denominator. Selection bias. Another point of attention is confounding. In the latter situation. 89 .3 If there is a specific attention for an interaction however. is not likely to have played a more prominent role in cases where both OCs and itraconazole were used than in cases where only OCs were used. Despite the fact that a delayed withdrawal bleeding rarely occurs with oral contraceptives this ‘event’ might be reported more frequently. which might influence the occurrence of a withdrawal bleeding. The indication for the use of itraconazole. the use of NSAIDs for self-treatment could be definitely ruled out in only one case. This selection bias can cause an overestimation of the odds ratio for the occurrence of a ‘delayed bleeding’ in reports were OCs are used. however.

this estimation was not made. an enzyme system involved in the biotransformation of steroid hormones.14 Triazole compounds such as ketoconazole and itraconazole in vitro inhibit cytochrome P450-3A. Of the ten women who reported this suspected side effect. nine women used a desogestrel containing OC. Because of the various degrees of underreporting.2 External validity Like the analysis of ADRs in a spontaneous reporting system. For our analysis we focussed on a comparison of the chances of ‘events’ being reported in the different groups.Chapter 2. the strong association between the concomitant use of itraconazole and OCs and the occurrence of a ‘delayed withdrawal bleeding’ therefore is suggestive of an interaction. detection and confirmation of an association does not imply a causal relationship. the delay in withdrawal bleeding might have been caused by the use of itraconazole. Since there were no cases. For this reason. and one woman a cyproterone acetate containing OC. Additional studies need to be done to confirm a true pharmacodynamic or pharmacokinetic interaction and to determine the strength of this association in daily practice. different mechanisms may be responsible for the ‘delayed bleeding’ in these two groups.11-13 it is difficult to estimate the true incidence of a possible interaction. In our example where the signal power strongly points at a true interaction. The number of patients who used itraconazole in the absence of OCs is rather small. The chance of this ADR being reported was not increased in women who used itraconazole or an OC alone. The reason for the high number of desogestrel containing OCs is not clear. If a woman uses itraconazole without an oral contraceptive. the odds ratio must be zero. that are inextricably bound up with spontaneous reporting. The mechanism underlying this suspected interaction remains to be explained. Itraconazole and oral contraceptives The concomitant use of itraconazole and oral contraceptives might lead to an increase in reports of a ‘delayed bleeding’. but the corresponding confidence interval was large. Theoretically.15. but the reporting rate is much higher when oral contraceptives were used concomitantly.16 An increase in the levels of steroids might lead to a delay of the withdrawal bleeding. the ‘delayed bleeding’ is in fact a 90 . Itraconazole interacts with a variety of drugs. since data concerning the degree of underreporting and prescription rates of drug are not available. Nevertheless.

13:1-10.44:277-81. Since it is not clear if different mechanisms underlie these suspected ADRs. de Koning GHP. 2. Utrecht. Verstoring van de pilcyclus tijdens het gelijktijdig gebruik van itaconazol en orale anticonceptiva. 1994:431-47.316:98991.110:300. New York: John Wiley & Sons. Bakker A. Disturbance of withdrawal bleeding during concomitant use of itraconazole and oral contraceptives. 4. Leufkens HGM. 91 . Maislin G.Drug-drug interaction between itraconazole and oral contraceptives delayed menstruation. Meth Inform Med 1974. Meyboom RHB. Feenstra J. the Netherland.5:123. In women using oral contraceptives. Davies HT. it refers to a delayed withdrawal bleeding. 6. (Thesis) Utrecht University. Ned Tijdschr Geneeskd 1998. De Koning GHP. Conclusion In a spontaneous reporting system the automatic analysis of reports is usually limited to adverse drug reactions. 8. Analysis of the information in a central ADE database. Meyboom RHB. Stricker BHCh. Screening for unknown effects of newly marketed drugs. Br J Clin Pharmacol 1997. Tavakoli M. When can odds ratios mislead? BMJ 1998. Carson JL. Lastdrager CJ. Egberts ACG. the results strongly suggest that in women using OCs and itraconazole concomitantly the withdrawal bleeding may be delayed due to a drug interaction. the index and reference groups originating from the database can be used to signal possible drug-drug interactions. Crombie IK. Amery W. Tijssen JGP. 3. 5. Intern J Risk Safety in Med 1993. Finney DJ. both ADRs were treated as one variable in our analysis. 9. Van Puijenbroek EP.45:1177-84. J Clin Epidemiol 1992. 7. References 1. In our example. Vinks MHAM. 1994. Van Puijenbroek EP. In: Pharmacoepidemiology. In the method described. Serum sickness-like reactions to cefaclor. Meyboom RHB. N Z Med J 1997. Strom BL. Systematic signalling of adverse reactions to drugs. Non-puerperal lactation associated with antidepressant drug use. A regionalized spontaneous surveillance program for adverse drug reactions as tool to improve pharmacotherapy. and illustrates that the method described might be a useful tool in analyzing possible interactions in spontaneous reporting systems.142:146-9.

4:161-3. 11. Pharmacoepidemiol Drug Saf 1995. Bloomer JC. Post Marketing Surveillance 1993. J Clin Epidemiol 1996. 13. de Koning GHP. Comparing the toxicity of two drugs in the framework of spontaneous reporting: a confidence interval approach. Meyboom RHB. Back DJ.1:11-3. Disturbance of menstruation as a side effect of nonsteroidal antiinflammatory drugs (NSAIDs). 12. Chaslerie A. Comparing safety of drugs. Bégaud B.7:119-37. 1996. Begaud B.Chapter 2. 16. Ayrton AD. 14. Tubert-Bitter P. Van den Bemt PMLA. Heymeijer GWJ. Chenery RJ. Azoles and allylamines: the clinical implications of interaction with cytochrome P450 enzymes. Ketoconazole and sulfaphenazole as the respective selective inhibitors of P450 3A and 2C9. 15.49:121-3. Faich GA. Baldwin SJ. Drug interactions. Stockley IH. Tjia JF. Clarke SE. Drug Inf J 1986. Oxford: Blackwell Science. Tubert-Bitter P. Factors affecting physician reporting of adverse drug reactions.20:157-64. Milstien JB.25:261-70. Xenobiotica 1995. J Dermatol Treatment 1990. Smith GJ. 92 . Haramburu F.2 10. Hsu JP. Moride Y.

3 DETECTING DRUG-DRUG INTERACTIONS USING A DATABASE FOR SPONTANEOUS ADVERSE DRUG REACTIONS: AN EXAMPLE WITH DIURETICS AND NONSTEROIDAL ANTI-INFLAMMATORY DRUGS EP van Puijenbroek1. 's-Hertogenbosch.3. ER Heerdink2 and HGM Leufkens2 1 2 Netherlands Pharmacovigilance Foundation Lareb.2. Utrecht Institute for Pharmaceutical Sciences. Utrecht. the Netherlands Department of Pharmacoepidemiology and Pharmacotherapy. Tilburg. the Netherlands EUROPEAN JOURNAL OF CLINICAL PHARMACOLOGY 2000. Elisabeth Hospital. TweeSteden Hospital and St.56:733-8 . ACG Egberts2. the Netherlands 3 Hospital Pharmacy Midden-Brabant.

94 . Methods Reports received between 1 January 1990 and 1 January 1999 of patients older than 50 years were included in the study. Conclusion The findings illustrate that spontaneous reporting systems have a potential for signal detection and the analysis of possible drug-drug interactions. For this reason. utilising reports of adverse drug reactions in data sets of SRS.0. As an example. Results The odds ratio of the statistical interaction term of the combined use of both drugs was increased (adjusted odds ratio 2. the traditional approach for analysing data from SRS has major limitations for the detection of drug-drug interactions. The method described may enable a more active approach in the detection of drug-drug interactions after marketing. The influence of the combined use of both drugs was examined using logistic regression analysis. which are often not explicitly reported. which may indicate an enhanced effect of concomitant drug use. 95% confidence interval 1.Chapter 2. Cases were defined as reports with symptoms indicating a decreased efficacy of diuretics. Exposure categories were the use of NSAIDs or diuretics versus the use of neither of these drugs.7). non-cases as all other reports. We developed a method that may enable signalling of these possible interactions. the influence of concomitant use of diuretics and non-steroidal anti-inflammatory drugs (NSAIDs) on symptoms indicating a decreased efficacy of diuretics was examined using reports received by the Netherlands Pharmacovigilance Foundation Lareb.13.3 Summary Objective Drug-drug interactions are relatively rarely reported to spontaneous reporting systems (SRSs) for adverse drug reactions.

Usually a drug-drug interaction might be suspected in the event that similar substances have proven to cause a similar interaction. patients with multiple drug use are usually excluded. in reports received by the Netherlands Pharmacovigilance Foundation Lareb. spontaneous reporting systems (SRSs) have been used to detect adverse drug reactions (ADRs) after marketing of drugs. the concomitant use of both diuretics and 95 . In particular in the elderly.6 The basic principle of looking for disproportionality can be extended to the detection of drug-drug interactions. one drug influences the effect of another drug.2-4 For signal detection concerning possible unexpected ADRs.5. Furthermore. In the event of a drug-drug interaction. it is usually difficult to recall whether specific concomitant medication was also used in similar reports. In the event a drug-drug interaction is unexpected given the previous knowledge. automated signal generation may be a promising tool for selecting possible combinations of ADRs and drugs that might be worthwhile analysing in more detail. An automated statistical approach may be helpful in analysing large numbers of these reports and in revealing the existence of these complex relationships. By analysing individual reports.1 Since the size of data sets is increasing. these reporting systems play a major role in pharmacovigilance.Drug-drug interaction between diuretics and NSAIDs Introduction Since the early 1960s. This may subsequently cause an increase or decrease in the number of reported ADRs of the latter drug. it is usually not clear whether ADRs arise directly from the use of a certain drug or that the ADR concerned is in fact the result of an underlying pharmacodynamic or pharmacokinetic drug-drug interaction. as was shown in a previous study by our group. If this is not the case. including Reporting Odds Ratios (RORs). various measures of disproportionality can be used. enhancing the number of possible explanations of a certain unexpected clinical event. The influence of the combined use of drugs can be studied by introducing a statistical interaction term in a logistic model for the calculation of RORs.7 As an example of the approach described. which are generally more difficult to detect. detection becomes more complicated. in individual patients. additional factors such as co-morbidity and multiple drug use may be present. In pre-marketing trials. it is rarely reported to an SRS. which makes the detection of drug-drug interactions in the post-marketing period even more important. Nowadays.

the reported possible ADR is coded by a qualified assessor using the World Health Organization (WHO) Adverse Drug Reaction Terminology. ‘oedema peripheral’. ‘cardiac failure right’.8-21 Methods Setting and design The Netherlands Pharmacovigilance Foundation Lareb is the national centre for submitting spontaneous reports of suspected ADRs originating from health care professionals in the Netherlands. Data concerning ages and genders of the patients had to be available. Reports that mentioned one of the aforementioned ADRs were defined as cases. ‘pulmonary oedema’ and ‘oedema legs’.Chapter 2. 96 .27 This specific increase in reports is assumed to be reflected in the number of reports to Lareb. This drug-drug interaction has been described in several casereports and studies. All other reports were defined as non-cases. After being received by Lareb.28 The presence of one or more of the following WHO preferred terms on the reports was therefore considered as an indication for this situation: ‘oedema’.7. All reports submitted to Lareb between 1 January 1990 and 1 January 1999 of patients older than 50 years were included in the analysis. ‘cardiac failure’. ‘oedema dependent’. ‘cardiac failure left’.3 non-steroidal anti-inflammatory drugs (NSAIDs) was associated with the occurrence of symptoms that may point towards a reduction of the therapeutic effects of diuretics.22 These reports are considered to be a reflection of the ADRs that occur in daily practice.23-26 The analysis of drug-drug interactions is based on the assumption that specific ADRs may occur more frequently when both drugs are used concomitantly in comparison with separate use. Selection of cases and non-cases A decrease in the efficacy of diuretics may express itself as the occurrence of oedema or signs indicating the onset or worsening of congestive heart failure (CHF). ‘oedema generalised’. taking into account the various degrees of underreporting that are an inherent attribute of spontaneous reporting.

we therefore made no distinction between suspected and non-suspected medication. Exposure categories were the use of NSAIDs (WHO Anatomical Therapeutic Chemical (ATC) classification code M01A). All medication that the patient was using according the medication history on the calendar date of the event was considered. or diuretics (ATC code C03) versus the use of neither of these drugs. source and reporting year. peripheral vasodilatating drugs (ATC code C04).e.0 was used. age and gender of the patient involved. calcium channel blocking agents (ATC code C08) and drugs acting on the renin angiotensin aldosterone system (RAAS. the use of antidiabetic drugs (ATC code A10). Covariates used in the analysis were: type of health professional that reported the ADR (either pharmacist or physician). Cn-x = different covariates. If a drug was used to treat the ADR it is not listed as concomitant medication. For the vast majority of the reports Lareb has the patients’ drug dispensing history from community pharmacies.Drug-drug interaction between diuretics and NSAIDs Exposure categories Information about concomitant drug use is requested on the reporting forms. Statistical analysis For the analysis. ATC code C09). cardiac glycosides (ATC code C01). A statistically significant value of the interaction term b3 indicates an additional effect of concomitant use of diuretics and NSAIDs. age. Since we were looking for drug-drug interactions. antihypertensive drugs (ATC code C02). i. The reporting health professionals give an indication which drug is considered the suspected drug. only one drug is indicated. b-blocking agents (ATC code C07). D= diuretics. For all analyses the statistical package SPSS 8.05 or less were considered statistically significant. Usually. 97 . Probability (P) values of 0. All drugs in use at the moment the ADR occurred were considered possible causes of the ADR. year of reporting. however. the following logistic model was used: log(odds) = b 0 + b1N + b2D + b3N*D + bn-xCn-x where N= NSAIDs.

6%) 49 (16.4%) 6220 (65.92-1. Characteristics of cases and non-cases concerning age. ttest) is significantly higher.15 (0.17 (0.94-1.25 (0.4 224 (73.37-3.19) NSAIDs non-steroidal anti-inflammatory drugs. which were separated in 305 cases and 9.8%) 656 (6. source of the reports and the use of several cardiac drugs are provided in Table 1.1%) 1717 (18.4%) Non-Cases n (%) n=9517 65.64) 4.31) 1.5%) 1546 (16.83-1.5%) 209 (2.2%) 1697 (17.01.Chapter 2.23-3.58 (1.1%) 1246 (13. 98 . the number of females and the age of the patients (p<0.06) 0. 95% CI 95% confidence interval.74-6.9 5848 (61. peripheral vasodilatating drugs.50) 1.45 (1.517 non-cases.79 (0.6%) 17 (5.6%) 127 (41.91) 1.907 reports of patients aged over 50 years.22-2.34-2.0 %) 66 (21.49-1.64-1. antihypertensive drugs and calcium channel blocking agents was more frequent among the cases.6%) 43 (14.9%) 67 (22.00) 0. Table 1.0%) Odds ratio (95% CI) 1. cardiac glycoside drugs.6%) 3 (1.0%) 78 (25.74 (3.11-1. Characteristics of cases and non-cases Cases n (%) n=305 Mean age (years) Females Reports by physicians NSAIDs Diuretics Insulin and oral antidiabetic drugs Cardiac glycosides Antihypertensive drugs Peripheral vasodilatating drugs Beta-blocking agents Calcium antagonists Drug acting on the RAAS system 67.60) 2.78) 1. NSAIDs.19 (1. Eighty-five reports were excluded because age or gender of the patient was not known.73 (1.24) 1. There were no differences concerning the use of insulin and oral antidiabetic drugs. beta blocking agents. gender. RAAS renin angiotensin aldosterone system Among the cases. were included in the analysis. A total of 9.2%) 79 (0.8%) 1727 (18. Lareb received 9.73) 1.822 reports.1%) 14 (4.3 Results From January 1990 until January 1999.4%) 210 (68.14 (1.87 (0. and drugs acting on the RAAS in the medication history. Also the use of diuretics.9%) 1185 (12.

77) 95% CI 95% confidence interval. The distribution of drugs present in cases and non-cases.0. oedema pulmonary (n=3) and oedema legs (n=66). 95% CI 1. This is an indication for an 99 . If no NSAID was used.Drug-drug interaction between diuretics and NSAIDs Among the cases. the total number of suspected ADRs exceeds the number of cases. Distribution of drugs present in cases and non-cases. oedema dependent (n=90). is shown in Table 2. oedema generalised (n=14).29 (95% CI 0. Table 2. If an NSAID was used by the patient.29 (0. The corresponding odds ratios with 95% confidence intervals are also shown.08) 1. cardiac failure left (n=6).77).82-5.95-1.03 (1. However. the following suspected ADRs were mentioned on the reporting forms: oedema (n=68). cardiac failure (n=14). the crude odds ratio was 3.03 (95% CI 1.7). OR odds ratio Crude and adjusted odds ratios and 95% confidence intervals are shown in Table 3.517 6527 1444 7971 1293 253 1546 Total OR (95% CI) 6712 1497 8209 1335 278 1613 3. stratified for the use of non steroidal anit-inflammatory drugs (NSAIDs) and diuretics Cases n=305 NSAID not present Diuretic not present Diuretic present Total NSAID present Diuretic not present Diuretic present Total 185 53 238 42 25 67 Noncases n=9. Since more than one ADR can be attributed to one report.08). The WHO term ‘cardiac failure right’ was not noted on the report forms.95-1. oedema peripheral (n=45). stratified for the use of NSAIDs and diuretics.1-3. the odds ratio of the statistical interaction term NSAIDs*diuretics is statistically significant (adjusted odds ratio 2. The use of diuretics or NSAIDs alone is not statistically significant as an increased risk for onset or worsening of symptoms of CHF or oedema.82-5. the crude odds ratio for the use of diuretics was 1.

95% CI 95% confidence interval.75 (0.Chapter 2. year of reporting.86-1. diuretics and covariates on the occurrence of symptoms of congestive heart failure (CHF) during the use of NSAIDs and diuretics. associated with the combined use of both drugs.15 (0.77) 2.73) 1.01 (1. gender. bblocking agents.25) 0. Association between non-steroidal anti-inflammatory drugs (NSAIDs). can be detected using a spontaneous reporting system.09-6. peripheral vasodilatating drugs.68 (0. source.25-2.08-3.69) 1. ATC anatomical therapeutic chemical Discussion Methodology In the event drug-drug interactions had not been previously observed clinically or experimentally.93 (0.09-3. Covariates Adjusted odds ratio* (95% CI) 1.48-0.29 (0.3 enhanced chance of cases being reported.75-1. Although this method was developed to detect unknown drug-drug interactions.38-2.86) 1.41) 5.93 (0. in this case the 100 . the detection in the post-marketing phase is troublesome.06 (0.75) 0.71-1.00-1.83 (0. We developed a method by which drug-drug interactions. which were not reported explicitly.82-1.22 (0.29-4.85 (0. and drugs acting on the renin angiotensin aldosterone system (RAAS).45) 0.88-0.05 (0. and the use of antidiabetic drugs.80 (1.95-1. drugs (C04) b-blocking agents (C07) Calcium antagonists (C08) Drugs acting RAAS system (C09) *Odds ratio adjusted for age.15) Unadjusted odds ratio (95% CI) 1.98) 0. cardiac glycosides.98) 1. Table 3. antihypertensive drugs.22) 0.28) NSAIDs (ATC code M01A) Diuretics (ATC code C03) Interaction term NSAIDs*Diuretics Age (year) Gender (female) Source (physician) Year of reporting Antidiabetic drugs (A10) Cardiac glycosides (C01) Antihypertensives (C02) Peripheral vasodilat.94 (1.00 (1.35 (1.46) 2.25 (4. calcium-antagonists.45-1.60-1.79-1.02) 1.34) 0.61) 1.

a third drug may act as a confounder when it is associated with one of the drugs or the concomitant use of both suspected drugs. however. Increased reporting due to a recent introduction of a drug or attention for an ADR in the media does not necessarily influence the Reporting Odds Ratio since nonselective reporting bias has a similar effect on both numerator an denominator.35). These findings are supportive for an 101 . The observation of a disproportionate association relative to the whole database.Drug-drug interaction between diuretics and NSAIDs well-known interaction between diuretics and NSAIDs was used to illustrate this new approach. This odds ratio can be also be calculated as the product of the ROR of diuretics (1. The adjusted odds ratios for the latter interaction terms were not significantly increased. a reporting bias might occur. the odds ratio of the statistical interaction term of the combined use of both drugs was increased. As shown in Table 2. whereas the adjusted odds ratio for the interaction term NSAIDs*diuretics was 2. mentions a suspicion of a possible drug-drug interaction.7). In this way. does not imply a causal relationship but suggests an association and serves as a starting point for further analysis. which may indicate an enhanced effect of concomitant drug use.5 Another point of attention is confounding bias. the ROR expressing the magnitude of the association between diuretics and signs of oedema or CHF increased when NSAIDs were used among the concomitant medications. Therefore.0 (95% CI 1.0-3. If there is a specific interest in an interaction for instance. In contrast to the exposure categories diuretics or NSAIDs. in a separate analysis. The interaction term NSAIDs*diuretics also might have been increased due to confounding if a statistical interaction existed between the use of NSAIDs and a pre-existing CHF. Interpretation of the data should be done with great trepidation due to the spontaneous character of the reports giving rise to all possible sources of bias. The methodology is to be used for signalling interactions and not for signal testing. the calculated odds ratios approximate the incidence rate of the ADRs in the population of drug users. for which diuretics were used to treat this condition. For instance. even if different adjustments are made.29) and the ROR of the interaction term (2. The analysis rests upon the assumption that the data set of the SRS is a close representative of the event rate occurring in the population of drug users. None of the original reporting forms. the interaction terms NSAIDs*cardiac glycosides and NSAIDs*ACE-inhibitors were added to the logistic model.

since the change on CHF increases in elderly patients.6 (95% CI 1. for instance in case of diabetes mellitus and hypertension. the number of cases that were coded as cardiac failure was rather 102 . studying the different classes of NSAIDs would be possible. pharmacists might be more familiar with the drug-drug interaction under investigation.e. NSAIDs and diuretics Several reports in literature suggest that concomitant use of diuretics and NSAIDs might lead to a decrease in the effect of the diuretic drug. the analysis was not restricted to oedema only. Because of the various degrees of underreporting that are inherent to SRS.3). cannot be excluded. In larger databases that also offers the possibility of the use of concomitant medication.29 both antidiabetic drugs and antihypertensive agents were used as covariates.31 and oedema is a common ADR of these drugs.3 actual interaction between diuretics and NSAIDs and not for confounding due to an interaction between the use of NSAIDs and a pre-existing cardiac failure.8-21 This applies to sulphonamides as well as thiazide diuretics. In this age group the use of NSAIDs also might increase. enhancing the risk of using NSAIDs. i. Since the starting point in the analysis was a reduction in the effect of diuretics. If an additional analysis is done after exclusion of all reports that mention the presence of one of the calcium antagonists in the medication history. mainly oedema. In the logistic model we corrected for age of the patient. however.30. the influence of calcium antagonists on the occurrence of symptoms of CHF.33 In our study. Different covariates were used in the analysis. The source of the reports was also used as a covariate.Chapter 2. Nevertheless. The use of NSAIDs alone (in the absence of the use of diuretics) is also associated with signs of CHF20. the adjusted odds ratio of the interaction term NSAIDs*diuretics was still 2.25. Since pre-existing impairment of kidney function. Because of the small number of reports with combined use of NSAIDs and diuretics. In this study we used a limited number of ADR codes that might indicate a decrease in the effect of diuretics. because of rheumatological disorders. it is difficult to estimate the true incidence of a possible interaction.32 The use of calcium antagonists is also associated with leg oedema. was evident. since physicians might report signs of CHF more often in comparison to pharmacists. we did not made a distinction between different classes of NSAIDs and diuretics.

Practical considerations of detecting drug-drug interactions Detecting drug-drug interactions in data sets of SRS implies a meticulous process. This step might be executed automatically by a computer programme. The first selection is based on the presence of a disproportionate increase in the number of reports of a certain ADR in association with a combined drug use (unadjusted odds ratio). analysis of drug-drug interactions should be developed as an automated process. SRS may also be used for the actual detection of drug-drug interactions. Furthermore. This might be partially related to the original descriptions of the reporting health professional. In general. Due to an increase in the size and quality of the SRS databases. the interaction term of the covariates involved can be derived. For this reason. Our results were indeed supportive for this drug-drug interaction. they can not be used for automated signal detection. either selected at random. Since it was not always clear when codes for dyspnoea and related pulmonary problems should have been attributed to respiratory or cardiological problems. which implies a rather strong clinical effect. the drug-drug interaction should lead to an increase in the number of reports. A second step is the refinement of the signal using a dedicated correction for possible confounders present in the database (adjusted odds ratio). the use of larger databases may be promising. provided high quality reports are available and concomitant drug use is filed in the database. Using the code dyspnoea in the analysis will therefore result in a dilution of the effect. Although in our example we strengthened a signal that was already known in literature. For various combinations. In our database. which sometimes hampers assigning an appropriate diagnosis. Given the fact that this interaction was present in a relatively small database.5%). there is a chance that it will be reported less frequently since it might be considered as a well-known interaction. If the association is monitored for in pharmacist or physician information systems. 103 . The aim of the present study was to illustrate the possible use of an SRS as a tool for signalling drug-drug interactions. or selected based on a related drug-drug interaction.Drug-drug interaction between diuretics and NSAIDs low (6. In theory a large number of possible drug-drug interactions is possible. the majority of cases of dyspnoea is used for shortness of breath resulting from a pulmonary cause. In such cases only the symptoms can be coded. a sufficient number of cases where both drugs are used concomitantly is a prerequisite for detecting interactions in a database for ADRs.

45:1177-84. Prevention and Health subdivision. Lansner A.8:S15-S25. Edwards IR. Non-puerperal lactation associated with antidepressant drug use. Acknowledgements We acknowledge the Netherlands Organisation for Applied Scientific Research (TNO).6:217-21. Van Puijenbroek EP. Hind ID. Bate A. Edwards IR.16:355-65. Hysert PE. Olsson S. Orre R. The effect of flurbiprofen on the responses to frusemide in healthy volunteers. From association to alert A revised approach to international signal analysis. Meyboom RHB. Signalling possible drug-drug interactions in a spontaneous reporting system: Delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole. Br J Clin Pharmacol 1999. Nunes AM. Mookerjee BK. 104 . Meyboom RH. Scott Med J 1982.44:277-281. 2. Eur J Clin Pharmacol 1998. Ståhl M. 9. Lee JB. References 1. de Koning GHP. Stricker BHCh.Chapter 2. Int J Clin Pharmacol Ther Toxicol 1983. 11. 10. de Koning FH. Drug Saf 1999. Babej M. Bate A. J Clin Epidemiol 1992. Meyboom RHB. Hekster YA.3 detection and analysis of drug-drug interactions clearly offers a major challenge for pharmacovigilance in the near future. JGP Serum sickness-like reactions to cefaclor. Daskalopoulos G. Pharmacovigilance in perspective. Edwards IR. Am J Kidney Dis 1985. Rawles JM. Egberts AC. Principles of signal detection in pharmacovigilance.21:429-47. Antagonism between non-steroidal anti-inflammatory drugs and diuretics. In contrast to the present way of detecting drug-drug interactions. Gonzalez M. 3. De Freitas RM.27:37-40. 5. Rees JA. 4.47:689-93. Zipser RD. Laffi G. and Leufkens HGM. Lindquist M. Pharmacoepidemiol Drug Saf 1999. A Bayesian neural network method for adverse drug reaction signal generation. Gribnau FW. Lindquist M. Egberts ACG. Bakker A. for partly funding this study (project ZON 28-2632). the method described provides an active approach in the post marketing phase of drugs. Sulindac and indomethacin suppress the diuretic action of furosemide in patients with cirrhosis and ascites: evidence that sulindac affects renal prostaglandins. Fucik H.21:350-4. Katkov W. Br J Clin Pharmacol 1997. 6. Meyboom RH. Symmons DP.54:315-21. Tijssen. Antagonism of the effects of furosemide by indomethacin in normal and hypertensive man. Egberts ACG.10:649-59. 7. Drug Saf 1997. Bentzel CJ. Patak RV. Prostaglandins 1975. Leufkens HGM. 8. Kronborg I. Kendall MJ.

Uppsala. 13. 28. Van Ganse E. Matheson J. Phenylbutazone and pulmonary oedema. Letter: Pulmonary edema and salicylates. Boeynaems JM.2:1358. Drug Saf 1997. Post Marketing Surveillance 1993. Study on the possible interaction between tenoxicam and furosemide. Vallotton MB. Moore ND. Kapoor KV. Fries JF. Ned Tijdschr Geneeskd 1996. Lekkerkerker JFF. Nieuwe regels voor het melden van bijwerkingen in Nederland na 1995. 15. Douchamps J. Stricker BH. Davidman M. Turner P. Interaction of diuretics and non-steroidal antiinflammatory drugs in man. Identification of risk for renal insufficiency from nonsteroidal anti. 31. Flamenbaum W. Corwin N. Rose M. Interaction between nonsteroidal anti-inflammatory drugs and loop diuretics: modulation by sodium balance. Kraines RG. Vree PH. Kleinbloesem CH. 105 .21:663-7. Arzneimittelforschung 1987. Verniory A. Riondel A. International Journal of risk & safety in Medicine 1993. Pharm Med 1986. Lawson M. Chennavasin P. WHO Adverse Drug Reaction Dictionary. Amery W. Mosterd A. Interaction of ketoprofen and frusemide in man.47:329-31. Walker SR. Stricker BH.64:407-15. Ahmad S. 14. Nevins M. 143:1130-4. 16. Derenne F. Lyon L. Lumley CE. NSAIDs associated with increased risk of congestive heart failure in elderly patients taking diuretics. Jackson SH. Favre L. 22. The under-reporting of adverse drug reactions seen in general practice.81:274-5. Arch Intern Med 1998. 24.inflammatory drugs. Drug Intell Clin Pharm 1988. Anonymous. 1995. Underreporting of suspected adverse drug reactions to newly marketed ("black triangle") drugs in general practice: observational study. Lancet 1969.248:1175-81. Juvent M. J Pharmacol Exp Ther 1989. Indomethacin and the response to bumetanide. Feenstra J. Clin Sci 1983.158:1108-12. Comparing safety of drugs. Grobbee DE. Am J Med 1983. 18. Indomethacin-bumetanide interaction: an alert. Sweden.22:505-6. 26. Clin Pharmacol Ther 1980. Tubert-Bitter P. Bakker A. 17. 23.67:655-8.74:820-8. 25.317:119-20. Vetter G. Lucker PW. Bumetanide-induced diuresis and natriuresis: effect of prostaglandin synthetase inhibition.inflammatory drugs. Am J Cardiol 1984. Bégaud B. Deger F.1:205-12. 29. Berque S. Wilton LV. Coles LS.7:119-37. Adverse cardiovascular effects of NSAIDs in patients with congestive heart failure. 19. Is reporting rate a good predictor of risks associated with drugs? Br J Clin Pharmacol 1999. Roth SH. Martin RM.54:246-7. 20. Piroxicam-furosemide drug interaction. From experiment to experience: side effects of nonsteroidal anti. BMJ 1998. Hall GC. Hitoglou-Makedou A. Mann RD. Tashima CK.27:421-5. Lagnaoui R. Broekmans AW. 27. 21. Hamburger R. De Koning GHP. Arch Intern Med 1983. WHO Centre for International Drug Monitoring. Ottervanger JP. Analysis of the information in a central ADE database. Herchuelz A. Pierfitte C. Baker DE.17:166-80. Hartmann D. Begaud B. 32. Glasson P. Heerdink ER. Leufkens HG. Brater C. Postgrad Med J 1991. Ann Intern Med 1974. Staroukine M. Kaufman J.37:1072-6. Wa TC.5:123. Blackshear JL. Stillman MT. Herings RM. J Clin Pharmacol 1981. 30.140:1166-7.Drug-drug interaction between diuretics and NSAIDs 12.

MacDonald M. 1996:488-535.Chapter 2. 33. In: Dukes MNG (ed). Amsterdam: Elsevier Science. 106 .3 Lim PO. Antianginal and b-adrenoreceptor blocking drugs. Meyler's Side Effects of Drugs.

3.10:135-142 . TweeSteden Hospital and St. ACG Egberts2. the Netherlands Utrecht Institute for Pharmaceutical Sciences. FEVER AND URTICARIA: SYMPTOMS OR SYNDROME? EP van Puijenbroek1. the Netherlands 3 Hospital Pharmacy Midden-Brabant.4 ASSOCIATION BETWEEN TERBINAFINE AND ARTHRALGIA. RHB Meyboom1 and HGM Leufkens2 1 2 Netherlands Pharmacovigilance Foundation Lareb. the Netherlands PHARMACOEPIDEMIOLOGY AND DRUG SAFETY 2001.2. 's-Hertogenbosch. Utrecht. Tilburg. Elisabeth Hospital.

4 Summary Objective The antifungal agent terbinafine has been approved for marketing in the Netherlands since 1992. and analysed statistically in order to determine whether symptoms are interrelated. 108 . Both urticaria and arthralgia were statistically significantly associated with reports on terbinafine compared to all other reports in the database. two of these reports concerned urticaria. Since the introduction of terbinafine. the Netherlands Pharmacovigilance Foundation Lareb received eight reports of arthralgia during the use of this drug. These reports were described in more detail. In four reports the additional presence of skin reactions was mentioned. Possibly these symptoms have a shared aetiology. Adverse drug reactions (ADRs) may occur in about 10 % of the patients. Results Case series as well as the results of the statistical analysis show a clustering of symptoms among reports of patients using terbinafine. the majority being gastro-intestinal disorders and skin reactions. were included.Chapter 2. Methods All reports with known gender and a reporting date between 1 March 1992 and 1 January 1999. Conclusion The findings might point towards a clustering of these symptoms in patients using terbinafine. concerning patients older then 10 years. Two patients who reported arthralgia also had a fever. fever and arthralgia were interrelated was examined by logistic regression modelling. presumably an immunological reaction. The extent to which the symptoms urticaria.

when the drug is used by a large number of patients under different circumstances.11 In general. An example is the detection of taste disorders associated with terbinafine.8 erythema multiforme8-10 and Stevens-Johnson syndrome. The association between terbinafine and arthralgia is not yet mentioned in the Dutch Summary of Product Characteristics. In the Netherlands. Adverse drug reactions (ADRs) may occur in about 10 % of the patients treated with terbinafine. fever and urticaria.13. The drug interferes with the ergosterol biosynthesis via specific and selective inhibition of fungal squalene epoxidase.1 Terbinafine is effective in the treatment of onychomycosis and several other types of dermatomycotic infections. The reports suggest a clustering of arthralgia. Since the introduction of terbinafine.2. frequently occurring ADRs are usually detected in pre-marketing trials. The objective of this article is to describe the clinical characteristics of these reports and to analyse the clustering of these symptoms statistically in order to determine the strength of the association. the majority being gastro-intestinal disorders (5%). In these trials. In most countries these systems are maintained by so-called ‘national pharmacovigilance centres’. drugs are used by a well-defined group of a relative small number of patients for a limited time span. Eight reports concerned arthralgia. previously unknown ADRs are still being detected. and skin reactions ( 2-3%). after the international marketing of the drug. but will be in the future. the Pharmacovigilance Foundation Lareb is responsible for collecting and analysing these reports. which was described for the first time in 1992. Lareb received 294 reports of suspected adverse reactions to terbinafine. In four of these reports the reporting physician or pharmacist also mentioned the presence of skin reactions. In case symptoms are interrelated. including two reports of urticaria.7.12 After marketing. Two patients who reported arthralgia also had a fever. this might be an indication for the existence of a syndrome. fever and urticaria Introduction The orally and topically active allylamine antifungal agent terbinafine has been approved for marketing in the Netherlands since March 1992.14 Reporting of ADRs by pharmacists or physicians to ‘spontaneous reporting systems’ (SRS) play an important role in the detection of ADRs.3 Its spectrum includes a broad range of dermatophyte and some yeast species. 109 . terbinafine has been associated with severe drug eruptions such as toxic epidermal necrolysis.4-6 Rarely.Association between terbinafine and arthralgia.

lympadenopathy. Statistical 110 .Chapter 2. a copy of the results was asked for.15 These reports are considered to be a reflection of the ADRs that occur in daily practice. liver or renal function disorders or a family history of rheumatological disorders. A report to an SRS can be considered as a set of data considering a single patient. Since terbinafine is rarely used in children and in our database no reports concerning patients younger than 13 years were present. renal function and skin disorders. type and distribution of arthralgia or arthritis and accompanying symptoms such as fever. Reports were excluded where the gender of patients was not reported. additional questions were asked about the time of onset and course of the symptoms. taking into account the various degrees of underreporting that are an inherent attribute of spontaneous reporting. a history of allergic or rheumatological disorders. the reporting physician or pharmacist was contacted by telephone.16-19 After being received by Lareb. a reminder was sent.20 Case series For all reports of terbinafine coded by ‘arthralgia’. If there was no response after an additional period of 3 weeks. After 3 weeks. Analysis of clustering Reports with a reporting date between 1 March 1992 and 1 January 1999 were eligible for enrolment in the study. In the event an additional radiological examination or laboratory testing was carried out. only reports concerning patients older than 10 years of age were included. in combination with one ore more drugs and one or more ADRs. Special attention was paid to the symptoms involved. Furthermore.4 Methods Setting The Netherlands Pharmacovigilance Foundation Lareb is the National Centre for spontaneous reports of suspected ADRs originating from health care professionals in the Netherlands. Questionnaires were sent to the reporting physician or pharmacist together with a copy of the data already present in our database. additional information concerning the clinical details was retrieved from the reporter by means of a questionnaire. the reported possible adverse drug reactions are evaluated and coded by a qualified assessor according to the WHO Adverse Drug Reaction Terminology.

Polyarthralgia was present in all eight cases. urticaria and arthralgia. Additional information was received with regard to six out of the eight cases of arthralgia.22 To study a possible relationship between fever. Time of onset of the reaction ranged from 1 or 2 days until 18 days.Association between terbinafine and arthralgia.B. All patients were treated with terbinafine 250 mg once daily. Patients A-F and H were treated for onychomycosis. Only patient B had a history of backache.D-F and H had no history of joint disorders.21. which were adjusted for age and gender of the patients. source of the reports and year of reporting. Reporting Odds Ratios (RORs) were calculated. Clinical details of these patients (A-H) are presented in Table 1 and the addendum of this thesis. cases were defined as reports on which terbinafine (oral administration) was mentioned as the suspected medication. Patients A. The indication for use of patient G was not received. Seven physicians and one pharmacist submitted the reports. In the logistic regression analysis a forward stepwise approach was applied. fever and urticaria. The extent to which the covariates are interrelated can be examined by using statistical ‘interaction terms’ in a logistic regression model. Since we were interested in expressing the presence of terbinafine as a function of arthralgia. traumata. non cases were defined as all other reports. 111 . Results Case series Cases involved five men and three women. such as logistic regression analysis can be used to provide evidence for a possible underlying relationship between these different covariates. fever and urticaria methods.0 was used for all statistical assessments. One patient had moved so additional information could not be provided. the ADRs were considered being covariates and the presence of terbinafine as the suspected drug on the report form to be the dependent variable. In the SRS databases the strength of the association between a drug and ADR can be studied by calculating Reporting Odds Ratios. No clear pattern could be recognised in the type of joints involved and the distribution of the arthralgia. SPSS 8. liver or kidney function disorders. The median age was 42 years (range 34-66 years).

No signs of arthritis. Clinical details of eight reports concerning arthralgia during the use of terbinafine Patient. one week later followed by right knee. No additional laboratory examination. 41 Mycosis C.9 mmol/l. ADL impaired. 44 Onychomycosis Doxycycline 100 mg during 7 days (1 week after start of terbinafine) Ethinylestradiol/ desogestrel Ibuprofen 400 mg. 5 mm. Symmetrical affected. leukocyte count 5. Treated with nabumetone 1 g twice daily. F. streptococcal antibody titre or rheumatoid factors not increased. 112 . Beclomethasone inhaler Pain of hands and feet Arthralgia of shoulders.4 x 109/l. M. No signs of arthritis. Backache in history. 8.2x109/l. After 1 week re-examination of blood revealed no abnormalities. Arthralgia of right wrist. Recovered after stopping terbinafine leukocyte count 7. rash (pruritus. F. C-reactive protein levels. 11 mmol/l. GGT not increased. Recovered. ESR. 42 D. F. No additional information received.Table 1. ADL impaired 18 days after start Concomitant medication Description of suspected ADR Time for the ADR to occur Remarks Fully recovered after stopping terbinafine. ALT. Pain on flexion and extension of wrists and knees. Treated by ibuprofen 400 mg three times daily. ‘Joint complaints’. erythema) 1-2 days after start 2 weeks after start 10 days after start Recovered after stopping terbinafine Lab: ESR. In blood smear. some atypical lymphocytes. Indication age (years) for use A. sex. B. Hb. 37 Onychomycosis Onychomycosis Onychomycosis Arthralgia. M. hips 10 days after start and elbows. miconazole cream. 41 E. GGT. malaise.

B. Indication for use of patient G was not known. prednisolone and promethazine. stomatitis and peeling of the skin developed. LDH. All patients were treated with terbinafine 250 mg once daily. ALT. D-F and H had no history of rheumatological or allergic disorders.8 x109/l. anorexia. leukocytes and blood smear within normal limits.M. Treated with terfenadine. 869 U/l. Treated with diclofenac and domperidone. These data could not be retrieved for patients C and G. vomiting. 50 Urticarial rash. 34 H. abdominal pain. No joints specified Treated with terfenadine and prednisolone. GGT. vomiting. Also rash. Lab: Hb. Generalized urticaria and arthralgia.3 mmol/l. No joints specified. traumata. 202 U/l. Fever of 39ºC leukocyte count 4. 5 mm. ESR. Miconazole cream 10 days after start Echinacea containing drug 2 weeks after start G. smell and taste disorders. headache. 1-2 One week after the start of terbinafine. liver or kidney function disorders. 9. 113 .F. Patients A. M. diffuse arthralgia. 1672 U/l. kidney function normal. ESR. Recovered. M. fever. 66 Onychomycosis Unknown Onychomycosis Budesonide inhaler Pain in almost every joint and Couple of days myalgia. Hb. No signs of arthritis. No rechallenge was carried out in either of the patients.

35-2. Of 98 patients sex or age was not known and 649 reports were excluded because the patients were younger than 10 years of age. Analysis of clustering Between 1 March 1992 and 1 January 1999.52-6. These data could not be retrieved for patients C and G.8%) 2 (0.01%) 15.3%) 1 (0.1%) 2 (0.7%) 17 (5. fever and arthralgia reported No arthralgia. arthralgia and fever on terbinafine versus all other drugs Reports on terbinafine Reports on other n (%) drugs n (%) Only arthralgia reported Only urticaria reported Only fever reported Both arthralgia and urticaria reported Both arthralgia and fever reported Both fever and urticaria reported Urticaria. Distribution of ADRs: urticaria. The distribution of the ADRs urticaria. of which 294 reports mentioned terbinafine as the suspected medication.14 (95% CI 1. A total of 16. of which eight were associated with terbinafine. 114 .9%) 15. urticaria of fever reported Total 5 (1.01%) 2 (0.01%) 2 (0. Joint complaints were reported 154 times.9%) 345 (2.4 rheumatological or allergic disorders.776 reports were received by the Netherlands Pharmacovigilance Foundation. There were no joint disorders in the family history of any of the patients concerned. fever and arthralgia on reports concerning terbinafine and other reports are presented in Table 2.179 (95.18)) and arthralgia (adjusted ROR 3.7%) 1 (0.47)) were significantly associated with reports on terbinafine (Table 3).72 (95% CI 1.3%) 1 (0. In none of the patients a rechallenge was carried out.833 Both urticaria (adjusted ROR 1.8%) 294 142 (0. Table 2.2%) 161 (1. a total number of 16.127 reports were included in the analysis.Chapter 2.3%) 267 (90.

The odds ratios and corresponding 95% confidence intervals are shown in Table 4. In general.32-4.22) 3.18) Adjusted odds ratio (95% CI)* 1. arthralgia and urticaria in association with the oral antifungal drug terbinafine. urticaria and terbinafine Unadjusted odds ratio (95% CI) Fever Urticaria Arthralgia 1.36) 1.Association between terbinafine and arthralgia. fever and urticaria Table 3.83-1.5 Nevertheless more serious reactions may occur. ADRs to terbinafine are often mild.83-1.33 (95% CI 1.46-6.17)) and arthralgia*urticaria (adjusted ROR 3.35 (95% CI 1.66 (1.00 (1.06 (0.03-10.73)).72 (1.23. Results of logistic regression (forward stepwise approach) .37) 1.66 (95% CI 1.14)) as well as the interaction terms arthralgia*fever (adjusted ROR 2. Association between terbinafine and arthralgia.32-4. Table 4.5%.38-2.35 (1.9% were considered to be related to be ‘probably’ or ‘possibly’ related to terbinafine.29-2.33 (1.03-10. For 115 .14 (1. source of the reports.884 patients was 10. Association between arthralgia.29-2. fever and urticaria. The reported overall incidence of ADRs in a large post marketing study involving 25. The majority involved the gastrointestinal system and the skin.17) 3. Liver and kidney function disorders may decrease the elimination of terbinafine. age and gender of the patient The covariates being the best predictors of the dependent variable were urticaria (adjusted ROR 1.24 These patients are more likely to experience ADRs due to an increase in plasma levels of terbinafine.75 (1.14) Discussion The case series as well as the statistical analysis suggest a clustering of fever. fever.18) 3.73) 1.47) *Odds ratio adjusted for year of reporting. Of all events 55.06 (0. Arthralgia * fever Arthralgia * urticaria Urticaria Odds ratio (95% CI) 2.35-2.52-6.

reported to Lareb. fever. In this study.Chapter 2.2% were thought to be related to terbinafine treatment.26 True serum sickness is caused by an immune complex-mediated reaction to a foreign serum protein occurring 1-3 weeks after exposure. Since this is not a parameter commonly tested by general practitioners. For instance. For the majority of the reactions. skin reaction or fever may occur with many drugs. and were considered as being possibly related to the use of terbinafine.3% of the patients. this information was not available.2%. but they were not observed. Serum sickness-like reactions. terbinafine has been associated with the clinical diagnosis of a serum sickness-like reaction in one case report. requires information about complement factors C3 and C4.28. Patient A also used doxycycline. ankles and wrists) in a symmetrical distribution. Arthralgia occurred in 0.8% in the postmarketing study mentioned.24 In another post marketing study among 10. myalgia and arthralgia 6 weeks after the start of terbinafine therapy. In the literature anecdotal case reports have appeared.4%).3%). followed or accompanied by pain and swelling of the joints (knees. the time of onset is compatible with a serum sickness-like reaction.4% of the patients. This concerned a well documented case of an-81-year-old male. of which only 25. It is not yet clear if these reports represent distinct or related symptoms. who developed exanthematous rash. of which urticaria was reported 47 times (0. in the latter case they might share a common aetiology. 10. which were probably part of the previous mentioned postmarketing study.361 patients. however. musculo-skeletal events were reported by 1. with arthralgia or arthritis. Dermatological events were reported by 341 patients (3.27-29 The diagnosis of a true serum sicknesslike reaction. describing symptoms similar to those in our study. which was incidentally associated with a serum sickness-like reaction. Usually it presents with an urticarial rash.25 Both studies had a descriptive character and no control groups were used.4 this reason we asked for liver and kidney function disorders in our questionnaire. None of the musculo-skeletal events was serious and there was no evidence that terbinafine exacerbated pre-existing muskulo-skeletal symptoms.30 Involvement of doxycycline in this patient cannot be 116 . Complaints of the musculo-skeletal system were reported in 0. myalgia in 0. but penicillin is the most common cause.9% of the patients had pre-existing musculoskeletal system disorders. Details on the simultaneous occurrence of different symptoms were not provided. Half of the cases of arthralgia and myalgia occurred within the first 2 weeks of treatment.

41 The description of the symptoms of patient H might resemble a Stevens-Johnson syndrome or erythema multiforme.Association between terbinafine and arthralgia. Circulating antibodyantigen complexes and complement activation have been demonstrated in the blood and skin lesion of patients.7. terbinafine has been associated with a hypersensitivity syndrome consisting of fever.44 As far as we know. but this could not be ruled out. and a skin reaction with erythema and peeling. since no antinuclear antibodies were checked. joint pain and sometimes swelling of the joint can occur together as 'urticarial arthralgia'. fever and urticaria fully excluded.42 Not only terbinafine. but also other antifungal agents have also been associated with joint disorders.27 Itraconazole was also associated with a serum sickness-like reaction in a 53-year-old female. there are no signs from mycotic infections that resemble the symptoms presented in our patients. Terbinafine has in incidental case reports also been associated with severe drug eruptions such as toxic epidermal necrolysis.38 and the exacerbation of a systemic lupus erythematosus39 were reported. lymphadenopathy and hepatitis 10 days after the start of oral terbinafine.35 Also druginduced psoriasis. lymfadenopathy and hepatic dysfunction after 4. a Stevens-Johnson syndrome may be accompanied by extracutaneous manifestations. like arthralgia.42.8 erythema multiforme. fever.33 Also our patient H had abnormal liver function tests. although some food-allergic reactions also might be immune complex-mediated. Other rare skin reactions. an immune mechanism might be involved in the process. Another hypersensitivity syndrome was described in a 66-year-old male who developed a cutaneous eruption.36 cutaneous lupus erythematosus37.5 weeks of therapy. The antifungal antibiotic amphotericin B may cause severe pains in muscles and joints along with many other side-effects. In our patients.8-10 Stevens-Johnson Syndrome11 and acute generalised exanthematous pustulosis.31 In another case history in the literature. the symptoms were not suggestive of a SLE-like syndrome.32 No arthralgia was mentioned in this patient.4. Confounding by indication is therefore unlikely.34. urticaria. liver and kidney disorders and fever. associated with the use of terbinafine are a fixed drug eruption40 and an erythema anulare centrifugum-like psoriatic drug eruption. in which attacks of severe urticaria and joint pain occur coincidentally due to an urticarial reaction of the synovia.43 In a Stevens-Johnson syndrome. Next to mucosal lesions. Usually it involves a type I hypersensitivity reaction. 117 . pruritic eruption. Alternatively.

fever and urticaria.4 In an additional analysis the clustering of the separate symptoms was analysed in more detail. Also in this situation the variables urticaria and the interaction term arthralgia*fever were statistically significant. if they are related (part of a syndrome) or coincidental. it is of major interest. Case series as well as the results of 118 . we also included the WHO term ‘arthritis’ in a separate analysis. but cannot be considered as conclusive evidence. next to urticaria two interaction terms were found to be the strongest predictor of the dependent variable. In logistic analysis. In general. no report was present in which all three symptoms were mentioned together. For methodological reasons. however. We propose that a similar approach may be more widely used in identifying drug-induced syndromes in spontaneous reporting databases. the corresponding interaction term indicating the clustering of all three symptoms could not be taken into account in the analysis. The findings of the statistical analysis are in support of a clustering of the symptoms involved. These signals can be detected by means of the analysis of individual reports.Chapter 2. Because the distinction between arthralgia and arthritis is sometimes difficult to make. All eight patients in our analysis had ‘arthralgia’.46 When two or more adverse events occur in a patient. In the control group. urticaria and fever are likely to be related or separate events.e. but often difficult to determine whether they both have been caused by the same or by a different pharmacological or pathological mechanism. This indicates a clustering of arthralgia and fever as well as arthralgia and urticaria in reports on terbinafine. Also more complex relationships can be studied in SRS databases. i. In this study we have used statistical reasoning to address the question if terbinafine-associated arthralgia. suspicions of new or unexpected ADRs originating from reporting systems only have a signal generating value. Conclusion Reports to the Netherlands Pharmacovigilance Foundation Lareb are suggestive of an association between the use of the antifungal agent terbinafine and the occurrence of arthralgia. For this reason. Spontaneous reporting systems are primarily designed for the detection of individual symptoms in relation with drug use.45. like signalling possible drug-drug interactions. further epidemiological studies are necessary to confirm a possible causal relationship between a suspected drug and suspected ADRs.

Toxic epidermal necrolysis and erythema multiforme following therapy with terbinafine. Considering the possibility of an immunological reaction. 5. Treatment of onychomycosis with terbinafine. Lynde CW. 9. Carstens J.43:259-84. presumably an immunological reaction. and the Netherlands Organisation for Applied Scientific Research (TNO). fever and urticaria the statistical analysis show a clustering of symptoms among reports of patients using terbinafine. McGregor JM.36:858-62. 119 .Association between terbinafine and arthralgia. Sogaard H. White SI.14:124-7. Arch Dermatol 1997. Br J Dermatol 1996.134:188-9. Faulds D. Terbinafine therapy may be associated with the development of psoriasis de novo or its exacerbation: four case reports and a review of drug-induced psoriasis.126(Suppl 39):40-6. 6. Baudraz-Rosselet F. 4. Rakosi T. Nahata MC. and therapeutic potential in superficial mycoses. Oral terbinafine: a new antifungal agent. Drugs 1992. Rustin MH. Clin Exp Dermatol 1989. 3. Terbinafine and erythema multiforme. Ann Pharmacother 1997. Shear NH. 7.133:1213-9. These findings might point towards a shared aetiology of these symptoms. Sibbald RG. Balfour JA. Acknowledgements We acknowledge physicians and pharmacist of the patients concerned for their help. Monka C.31:445-56. Kenzelmann R. Krupp P. References 1.74:3912. Thestrup-Pedersen K. Clinical efficacy and tolerability of terbinafine (Lamisil)-a new topical and systemic fungicidal drug for treatment of dermatomycoses. Wendelboe P. O'Sullivan D. it is important to inquire about involvement of other organ systems and to have additional laboratory examination done. Jones TC. 8. Wili PB. Toxic epidermal necrolysis induced by terbinafine in a patient on long-term anti-epileptics. Br J Dermatol 1994. J Am Acad Dermatol 1997. 2. Br J Dermatol 1992. it is advisable that the use of terbinafine in these patients should be stopped. Villars V. Bowen-Jones D. for partly funding this study (project ZON 282632). Acta Derm Venereol 1994.884 patients. Abdel-Rahman SM. Safety of oral terbinafine: results of a postmarketing surveillance study in 25. Terbinafine. Knowles SR. subdivision Prevention and Health. A review of its pharmacodynamic and pharmacokinetic properties.131:587-8. Gupta AK. Hall M. If arthralgia develops.

Vidal PC. Braunwald E.140:1166-7.340:728. 32. Hampshire: MacMillan Publishers Ltd.42:55965. 15. Gupta AK. De Koning FH. Rzany B.J Clin Pharmacol 1996. Br. Stevens-Johnson syndrome after terbinafine therapy. Gupta AK. 25. Clinical pharmacokinetics of terbinafine (Lamisil). Kapoor KV. 1991:689-713. Clin Exp Dermatol 1989. Gonzalez QA. 26. Bangs A. J Am Acad Dermatol 1994. Textbook of Adverse Drug Reactions. Loss of taste and terbinafine.30:509. 31. 120 . 18. Halpern S. Ned Tijdschr Geneeskd 1996. Fauci AS. New York: McGraw-Hill. 12. Lekkerkerker JFF. The under-reporting of adverse drug reactions seen in general practice. Hall GC.28:347-54. Pharm Med 1986.67:313-4. Broekmans AW. WHO Centre for International Drug Monitoring. 29.339:1483. Isselbacher KJ. Balter MS. JGP Serum sickness-like reactions to cefaclor. 13.4 10. Atkin K. Drug-induced arthritis and arthralgia. Comparing safety of drugs. Is reporting rate a good predictor of risks associated with drugs? Br J Clin Pharmacol 1999. O'Sullivan DP. 27. BMJ 1998. Terbinafine: an update. Postmarketing surveillance of oral terbinafine in the UK: report of a large cohort study. In: Davies DM (eds). Shear NH. 23. Tubert-Bitter P. Gehring W. Kruczynski K. Drug allergy and tests for its detection.317:119-20. Can J Clin Pharmacol 1995. Lancet 1992. 21. Drugs 1984. Mockenhaupt M. Lumley CE. Underreporting of suspected adverse drug reactions to newly marketed ("black triangle") drugs in general practice: observational study. Wilson JD. Shear NH. 1994:1638-43. Is there an allergic synovitis? J R Soc Med 1990. Kopstein JB.37:979-88. Needham CA. Munro DD. 17. Postgrad Med J 1991. Vree PH. Assem E-SK. Hypersensitivity reaction to terbinafine.Chapter 2.7:119-37.83:312-4. Martin RM. Juhlin L. 11. In: Frank MM. Post-marketing Surveillance In: Stephens MDB (ed). Leufkens HG. J Clin Epidemiol 1992. J Am Acad Dermatol 1997.47:329-31. 1995. Moore ND.2:129-30. Oxford: Oxford University Press. Anonymous. 1992:183-5. Bakker A. WHO Adverse Drug Reaction Dictionary. Oral terbinafine and erythema multiforme. Martin JB. 24. Kendall FD. Loss of taste and terbinafine. Golding DN. Detection of New Adverse Drug Reactions. Kasper DL (eds).14:110-3. Jensen JC. Lancet 1992. Ottervanger JP. Br J Clin Pharmacol 1997. Bégaud B. Wilton LV.44:277-281. J Am Acad Dermatol 1997. Hart FD. Todd P. Begaud B. Nieuwe regels voor het melden van bijwerkingen in Nederland na 1995. Non-puerperal lactation associated with antidepressant drug use. Mann RD . De Koning GHP. 28. 14. Walker SR. 19. Serum sickness-like reaction associated with oral terbinafine therapy. 16. Egberts AC. Doxycycline-induced parotitis. Tijssen.45:1177-84. Meyboom RH. Immune-complex diseases. Schopf E. Lagnaoui R.20:247-8. 20. Uppsala. Lawley TJ.36:1018-9. Principles of Internal Medicine. 30. 22.1:205-12. Stricker BH. Clin Exp Dermatol 1995. Stricker BHCh. Post Marketing Surveillance 1993. Pierfitte C.

Association between terbinafine and arthralgia, fever and urticaria
33. 34. Gupta AK, Porges AJ. Hypersensitivity syndrome reaction to oral terbinafine. Austral J Dermatol 1998;39:171-2. Dupin N, Gorin I, Djien V, Helal H, Zylberberg L, Leibowitch M, Escande JP. Acute generalized exanthematous pustulosis induced by terbinafine. Arch Dermatol 1996;132:1253-4. Condon CA, Downs AM, Archer CB. Terbinafine-induced acute generalized exanthematous pustulosis. Br J Dermatol 1998;138:709-10. Wilson NJ, Evans S. Severe pustular psoriasis provoked by oral terbinafine. Br J Dermatol 1998;139:168. Murphy M, Barnes L. Terbinafine-induced lupus erythematosus. Br J Dermatol 1998;138:708-9. Brooke R, Coulson IH, al-Dawoud A. Terbinafine-induced subacute cutaneous lupus erythematosus. Br J Dermatol 1998;139:1132-3. Holmes S, Kemmett D. Exacerbation of systemic lupus erythematosus induced by terbinafine. Br J Dermatol 1998;139:1133. Munn SE, Russell J. Terbinafine and fixed drug eruption. Br J Dermatol 1995;133:815-6. Wach F, Stolz W, Hein R, Landthaler M. Severe erythema anulare centrifugum-like psoriatic drug eruption induced by terbinafine. Arch Dermatol 1995;131:960-1. Nethercott JR and Choi BC. Erythema multiforme (Stevens-Johnson syndrome)- chart review of 123 hospitalized patients. Dermatologica 1985;171:383-96. Denman AM. Stevens-Johnson syndrome followed by persistent recurrent severe arthralgia. Br J Reumatology 1990; Br J Rheumatol 1990; 29:214. Park H, Knowles S, Shear NH. Serum sickness-like reaction to itraconazole. Ann Pharmacother 1998;32:1249. Van Puijenbroek EP, Egberts ACG, Meyboom RHB, Leufkens HGM. Signalling possible drug-drug interactions in a spontaneous reporting system: Delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole. Br J Clin Pharmacol 1999;47:689-93. Van Puijenbroek EP, Egberts ACG, Leufkens HGM. Detecting drug-drug interactions using a database for spontaneous reports of adverse drug reactions: an example with diuretics and non-steroidal anti-inflammatory drugs. Eur J Clin Pharmacol 2000;56:733-9.

35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45.

46.

121

Part 3

IMPLEMENTATION OF THE QUANTITATIVE APPROACH

In the third part of this thesis the implementation of quantitative signal detection at the Netherlands Pharmacovigilance Foundation Lareb is discussed. Detection of new adverse drug reactions (ADRs) after marketing is often based on a manual review of reports sent to a Spontaneous Reporting System (SRS). Among the many potential signals that are identified, only a limited number are important enough to require further attention. Goal of the study of Chapter 3.1 is to gain insight into factors contributing to the selection and dissemination of possible signals from the Netherlands Pharmacovigilance Foundation Lareb to the Dutch Medicines Evaluation Board and the contribution of quantitative signal detection in this process. Although in various centres quantitative techniques are being developed, they are generally not routinely implemented. The practical aspects of the implementation of quantitative signal detection are described in Chapter 3.2. At the Netherlands Pharmacovigilance Foundation quantitative signal detection has been used in various ways for a number of years. In this chapter, the role and position of quantitative signal detection and the way it is applied is described.

.

HGM Leufkens2. WL Diemont1. TweeSteden Hospital and St. Utrecht Institute for Pharmaceutical Sciences. Elisabeth Hospital. the Netherlands 3 Hospital Pharmacy Midden-Brabant.3 1 2 Netherlands Pharmacovigilance Foundation Lareb. the Netherlands BRITISH JOURNAL OF CLINICAL PHARMACOLOGY (IN PRESS) . 's-Hertogenbosch. Utrecht. and ACG Egberts2. AC van Grootheest1.3.1 DETERMINANTS OF SIGNAL SELECTION IN A SPONTANEOUS REPORTING SYSTEM FOR ADVERSE DRUG REACTIONS EP van Puijenbroek1. Tilburg. the Netherlands Department of Pharmacoepidemiology and Pharmacotherapy.

Knowledge of these factors may improve the efficiency of the underlying signal selection process. 95% CI 1. which could be expressed as a combination of a single ATC code and a single WHO preferred term. Among the many potential signals that are identified.7. For each case. Results Multivariate analysis showed that the presence of a ‘serious report’ (Odds Ratio 3. The number of reports and the time after marketing of the drug had no influence. Conclusion This study showed that selection of signals is based on both qualitative and quantitative aspects.8. such as the fact whether the ADR or drug is new. Logistic regression analysis was used to investigate the influence of various factors. 95% CI 2.Chapter 3. the strength of the association. 95% CI 1.4) were all independently associated with signal selection. 95% CI 1. Goal of this study is to gain insight into factors contributing to the selection and dissemination of possible signals originating from the SRS maintained by the Netherlands Pharmacovigilance Foundation.3-16) and the presence of a disproportionate association (OR 3.8-13). all signals (n=42) disseminated to the Medicines Evaluation Board from the second quarter of 1997 until the third quarter of 2000.3-11). the ADR being unlabelled (OR 6. were included. four controls were matched in time. 126 .1. the seriousness of the reaction and the documentation of the reports.5. Methods In a case control design. only a limited number are important enough to require further attention. a WHO ‘critical term’ (OR 4.4-8.1 Summary Objective Detection of new adverse drug reactions (ADR) after marketing is often based on a manual review of reports sent to a Spontaneous Reporting System (SRS).

do not systematically apply such criteria. the selection of the patients involved and the limited duration of the trials. On a three-monthly basis the most relevant signals are published in a so-called quarterly report to the MEB. the relation being previously unknown or incompletely documented. However. whether or not the issue is New. relatively few of them are important enough to require further attention. this agency is informed periodically about generated signals. which are for instance represented by the acronym ‘SNIP’. Although in a later stage the 127 . signal selection is carried out by a review of every incoming report in a weekly meeting between trained assessors.1 For this reason. Several factors may help to determine whether or not a particular signal is worth further investigation. These categories refer to the Strength of a signal. the signals described have a premature character. Working on behalf of the Dutch Medicines Evaluation Board (MEB). used by the Medicines Control Agency. however. Spontaneous reporting systems (SRSs) still play an important role in providing early signals concerning suspected ADRs. At the Netherlands Pharmacovigilance Foundation Lareb.2 The detection of signals from SRSs generally results from a systematic manual review of every incoming report sent to pharmacovigilance centres. criteria need to be formulated to help to determine if there is a need for further attention concerning a particular signal. four categories may be identified. For this reason. the clinical Importance as judged by the seriousness of the reaction and severity of the cases and finally the potential for Preventive measures by the regulatory authorities. close monitoring for unexpected adverse drug reactions (ADRs) remains necessary due to the limited size of pre-marketing trials. Criteria for selecting interesting signals about which the MEB should be informed. In general. post marketing surveillance aims at a timely detection of either new ADRs or an increase of the frequency of ADRs which are already known to be associated with the drugs involved. are not predefined and depend on the knowledge and experience of the assessors involved. maintaining the SRS in the Netherlands. Since the goal of this report is to give an early warning to the authorities. A signal can be defined as reported information on a possible causal relation between an adverse event and a drug.Determinants of signal selection Introduction After marketing of a drug. however.3 Most pharmacovigilance centres. In the present way of signal detection many potential signals are identified.

possible ADRs are assigned to a so-called ‘preferred term’. All ADRs are coded according the WHO adverse drug reaction terminology. these quarterly reports give a unique chance for studying the early stages of the process of signal selection and dissemination of information from SRSs. Cases were defined as associations that could be expressed as a combination of a single ATC code and a single WHO preferred term. The following exclusion criteria were applied: 1) associations concerning a group of related substances or related ADRs. The reports are evaluated. Knowledge of these factors may improve the efficiency of the underlying signal selection process. The suspected drugs as well as concomitantly used drugs are coded according to the ATC classification. factors associated with the selection and dissemination of signals to be reported to the MEB were evaluated.1 signals described may be superseded. drug-drug interactions but also reviews of drugs that were marketed recently in the Netherlands. the preferred terms ’anxiety’ and ‘nervousness’ share the same high-level term ‘anxiety’. Methods Setting The Netherlands Pharmacovigilance Foundation Lareb collects and analyses reports of suspected ADRs reported by Dutch physicians and pharmacists. which gives a detailed description of the clinical event. Preferred terms are linked to ‘high-level terms’. coded and filed in a database. The goal of this study is to gain insight in the various factors that play a role in the subjective process of signal selection. All signals described in the quarterly reports from the first publication in the second quarter of 1997 until the third quarter of 2000 were reviewed. which provides a code for qualitatively similar conditions. To the reporting health professional a letter of confirmation is sent together with information regarding the reported association. As an example. In a case control design. A report may concern one or more suspected ADRs and one or more suspected drugs. like ‘neuropsychiatric ADRs’ or ‘allergic 128 . Design Associations described in the quarterly reports to the MEB may represent possible new signals of possible ADRs. In this way clustering of ADRs for analysing purposes is possible.4 In this respect.Chapter 3.

Determinants of signal selection reactions’. Factors Factors possibly related to the selection of signals were distinguished in four different types. 3-4 reports and more than 4 reports. For each case. the ADR was considered labelled. being annually updated standard works often consulted by Dutch physicians or pharmacists in the Netherlands. factors related to the strength of the association. Factors related to strength of the association -The total number of reports received on this particular association at the time the index or control report was received. Three categories were defined: 1-2 reports. not being published in the quarterly reports. Controls consisted of those associations. In this way cases and controls were matched in calendar time. 129 . Similar to the ‘index report’ the reports that were used to select the control associations are called ‘control reports’. four controls were selected.6 -Period of marketing of the drug involved.5. for example gender or a special circumstance. otherwise unlabelled. 2) ‘review’ articles 3) drug-drug interactions. and 4) associations concerning a single ATC code and single preferred term but associated with an additional factor. A distinction was made between the drugs that have been marketed for less than five years on the Dutch market. An example of the latter one is for instance a possible withdrawal syndrome during the use of labetolol in a newborn infant. In case a physician or a pharmacist described two or more associations on the reporting form. the first association that was mentioned was chosen. The specific report that triggered the particular association to be selected as a signal was called the ‘index report’. namely the fact whether the association or drug is new. which were described in reports received by Lareb. The following factors were studied: The fact whether the association or drug is new -In the event the suspected ADR is mentioned in the Dutch text books ‘Farmacotherapeutisch Kompas’ or the ‘Informatorium Medicamentorum’. 10 and 20 reports previously and 10 and 20 reports later than the index report. factors related to the seriousness of the reaction involved and factors related to the documentation of the reports to the SRS.

In the event the denominator of the fraction is zero. 130 . In this situation. the ROR can not be calculated. disability. being indicative of serious disease states. 2x2 contingency table used for the calculation of ADR Reporting Odds Ratios An association was considered to be disproportionate in case that the lower limit of the 95% confidence interval was greater than 1. Factors related to documentation of the reports. Reported suspected ADR Suspected ADR of the Other suspected association involved ADRs Drug Suspected drug of the association involved Other suspected drugs a c b d Figure 1. the association involved was considered to be reported more frequently than its background frequency and having the same impact on the assessors as a statistically significant ROR. the ADR disappeared after cessation of the drug. the corresponding Reporting Odds Ratio (ROR) was calculated. -The presence of a so-called ‘critical term’. For calculating the number of reports concerning the suspected drug. (prolonged) hospital admission. Critical terms are a subset of the WHO preferred terms. the preferred term was used. disproportionality was calculated. critical terms may be of particular interest for signal generation. or congenital malformations. for calculating the number of ADRs. i. a life threatening situation. As a measure of disproportionality.7 This was done by creating a 2x2 contingency table for every association involved (Figure1). -The dechallenge of the index or control report is positive. This may occur in the event of a rare ADR.11 For this reason. the full ATC code was used.1 -The presence of a disproportionate association. which can be regarded as important to follow up. For every association. Factors related to the seriousness of the reaction involved -The index or control was considered serious in the event of death.e.8-10 The ROR can be calculated as a*d/b*c (see Figure 1) and expressed as a point estimate with the corresponding 95% confidence interval.Chapter 3.

the independent statistically significant factors resulting from the multivariate analysis were subsequently analysed in more detail. With respect to the selection of signals. The fact that the reports was sent by the physician (in attendance) of the patient. positive and negative predictive value of both the separate factors as well as the combination of these factors.e. for calculating the parameters of performance Microsoft Excel 97 was used. In the first part of the study a univariate analysis was carried out. For logistic regression analysis the statistical package SPSS 10. another 7 associations were excluded because they concerned possible drug-drug interactions. Odds ratios were expressed as point estimates with corresponding 95% confidence intervals. the performance expressed as sensitivity. An overview of the selected signals is shown in Table 1. Sixteen associations were excluded.0 was used. 7 associations were excluded because they were published as a ‘review’ article.Determinants of signal selection -The rechallenge of the index or control report is positive. 131 . i. specificity. Analysis Logistic regression analysis was used to analyse the influence of the various factors. Results Between the second quarter of 1997 and the third quarter of 2000 a total number of 76 associations were published in the quarterly reports to the MEB. and finally 4 associations were excluded because an additional factor was involved. because they concerned a group of related substances or ADRs. Factors that were statistically significantly associated with the selection of signals were analysed in the second part of the study in a multivariate analysis. A total number of 42 signals were included in the analysis and were matched with 168 controls. Finally. was calculated in respect to the manual selection of the associations in the signal selection process as the gold standard. the ADR reappeared after the drug was used again.

Raynauds syndrome alendronate .1 Table 1.decrease in libido miconazol .sialoadenitis valproic acid .parkinsonism metronidazol .dyspnoea diclofenac .angiooedema miconazol oral gel .death itraconazol .leucopenia nefazodone .trombopenia vigabatrin .anaphylactic reaction atorvastatin .hallucination losartan .urinary retention co-trimoxazole .trombotic trombocytopenic purpura rofecoxib .influence op protrombin time tramadol .angiooedema cisapride .convulsions paroxetin .Stevens-Johson syndrome minocycline .leucopenia diclofenac .death terbinafin .Chapter 3.chocking rulizole .taste disorder mefloquin .pulmonary fibrosis Date of publication 2nd quarter 1997 2nd quarter 1997 2nd quarter 1997 3nd quarter 1997 3nd quarter 1997 4nd quarter 1997 4nd quarter 1997 4nd quarter 1997 1nd quarter 1998 1nd quarter 1998 1nd quarter 1998 2nd quarter 1998 2nd quarter 1998 2nd quarter 1998 2nd quarter 1998 3nd quarter 1998 3nd quarter 1998 4nd quarter 1998 4nd quarter 1998 4nd quarter 1998 4nd quarter 1998 1nd quarter 1999 1nd quarter 1999 1nd quarter 1999 2nd quarter 1999 2nd quarter 1999 2nd quarter 1999 2nd quarter 1999 3nd quarter 1999 3nd quarter 1999 3nd quarter 1999 4nd quarter 1999 4nd quarter 1999 4nd quarter 1999 1nd quarter 2000 1nd quarter 2000 2nd quarter 2000 2nd quarter 2000 2nd quarter 2000 3nd quarter 2000 3nd quarter 2000 3nd quarter 2000 132 .eczema loperamide .interstitial pneumonia clopidrogel .rhabdomyolysis interferon alfa 2B .micturition disorder irberstartan .QT prolongation lamotrigin .hepatitis valproic acid .polycystic ovary syndrome acitretin – tast loss simvastatin .QT prolongation sildenafil .restless legs syndrome losartan .death pergolide . Selected signals from the quarterly reports to the MEB from the 2nd quarter of 1997 until the 3rd quarter of 2000 Association norfloxacin-fixed drug eruption oxybutinin .alopecia lamotrigin .death fexofenadin .anaphylactic reaction quetiapine .arthralgia lithium .tremor lamotrigin .haemolytic anaemia oral budesonide .visual field defect tolcapone .priapism olanzapin .

0 (0.2) 3. category) 3 or 4 reports more than 4 reports ROR full ATC code / preferred term statistically significant (lower limit 95% CI >1) Factors related to the seriousness of the reaction involved presence of a ‘serious’ ADR WHO critical term present Factors related to documentation of the reports to the SRS dechallenge positive rechallenge positive Source of reports: number of reports by physicians 17 (41) 4 (9.4) 27 (64) 17 (41) 67 (40) 39 (23) 2.6-16) 22 (52) 8 (19) 12 (29) 23 (55) 69 (41) 28 (17) 71 (42) 46 (27) 1 (ref) 0.5-16) 7.4) 21 (50) 26 (62) 20 (12) 30 (18) 7.2) 1. Differences between cases and controls and the results of the univariate analysis Cases n (%) n=42 How new is the association or the drug? ADR unlabelled Suspected drug shorter than 5 years marketed Factors related to strength of the association Absolute number of reports 1 or 2 reports (ref.1-4.5 (0.6-6.5) 1. the proportion of associations with a statistically significant ROR.6) Controls n (%) n=168 Odds ratio (95% CI) 133 .8-3.2 (1.4 (3.2-1.6 (0.5 (3.3) 0.2 (1. The absolute number of reports concerning the association does not contribute in the selection of signals. However.2 (0.9 (0.6-2.7 (1.4-2.5) 2.Determinants of signal selection Univariate analysis Table 2 shows the differences between cases and controls and the results of the univariate analysis.3-5.3-3. is higher among the cases (Odds Table 2.5) 107 (64) 1.5) 31 (74) 60 (36) 16 (9.

Table 3.5 (1.1 Ratio 3. or the source of the reports does not make a statistically significant contribution to the selection of possible signals. the ADR being unlabelled (OR 6.3-11) 4. 95% CI 1. The impact of a ‘critical’ term (OR 7.3-5.or rechallenge.8-13) 6. but this factor also has the lowest 134 . a WHO ‘critical term’ (OR 4. 95% CI 3. 95% CI 1. Also the proportion of unlabelled ADRs is statistically significant higher among the cases (OR 2.1 (2. Information regarding the quality of the documentation of the reports like the presence of a positive de. as well as the proportion of drugs that are shortly marketed than five years (OR 2.1-4.6-16) present among the cases is comparable with the presence of a ‘serious’ (OR 7.4.4-8.4) were all independently associated with signal selection. positive and negative predictive values for the factors contributing significantly to the selection of cases.5.3-16) 1. 95% CI 1.1. The presence of a ‘serious report’ (OR 3.5). the influence of factors that were positively associated with signal selection in the first analysis was analysed in a multivariate logistic regression analysis.5-16) report. 95% CI 3. 95% CI 1.7.Chapter 3. 95% CI 2.4).64). specificity.2.3-16) and the presence of a disproportionate association (OR 3.6-6.5-3.7.5) Performance of separate factors Table 4 shows sensitivity.4-8.5.4) 3.8.3-11). Results of the multivariate logistic regression analysis Odds ratio (95% confidence interval) ROR full ATC code / preferred term statistically significant Index or control report concerns a ‘serious’ ADR Critical term present ADR unlabelled Suspected drug shorter than 5 years marketed 3. Sensitivity is highest for the ADR being unlabelled (0. 95% CI 1.4 (0. 95% CI 1.2. The results are shown in Table 3.7 (1. The time since marketing of the suspected drug apparently did not make an additional contribution to the selection of signals.8-13).8 (1.6). Multivariate analysis In the second part of the study.

28 0.6 1 0. positive predictive value and negative predictive value of independent factors contributing to the signal selection process Sensitivity Specificity Positive predictive value Negative predictive value Individual factors ROR full ATC code / preferred term statistically significant Index or control report concerns a ‘serious’ ADR Critical term present ADR unlabelled Combined factors One or more factors present Two or more factors present Three or more factors present All four factors present 1 0. Factors responsible for the selection of signals were divided in four categories.52 0.26 0.48 0.33 0.81 0.73 0. all types seem to have influence on the selection of these signals for dissemination to the 135 .72 0. Except for the latter one.88 0. With respect to the negative predictive value all factors are comparable. a disproportionate number of associations in the database and the fact that the suspected ADR is unlabelled. all play a role in the signal selection process.87 Discussion Our study showed that the seriousness of the reaction.88 0.60 0. specificity. factors related to the strength of the association. being the fact if the association or the drug involved was new.80 0.95 0.92 0. Table 4. Sensitivity.29 0.99 0. all four independent factors resulting from this analysis are more or less comparable with respect to the performance.07 0.87 0.46 0. the seriousness of the association and factors related to the documentation of the reports.53 0. the presence of a critical term.55 0.89 0.62 0.28).Determinants of signal selection specificity (0.60) and lowest positive predictive value (0. Although slight differences exist.64 0.51 0.90 0.82 0.50 0.

further research will be necessary to gain insight in the contribution of the level of documentation of the reports in the selection of possible signals.17.18 Also techniques for the identification of possible drug-drug interactions and syndromes 136 . and the presence of a positive dechallenge or rechallenge in any other report than the index or control report. however. however. had an additional contributed to the selection of signals compared with the information present in the index of control report. were not statistically significant.1 MEB. Since the amount of data is increasing there is a growing need to additional quantitative signal detection techniques.3 A retrospective analysis of pharmacovigilance topics published between 1973 and 1990 in the Netherlands revealed that 46% of the topics was new and 57% referred to established products.9. but also other reports that were received previously may have contributed to the selection. Nevertheless. 64% of the ADRs were unlabelled and 60% referred to established products. Spontaneous Reporting Systems for ADRs have been used over the past 40 years. the aforementioned categories are also part of the ‘SNIP’ criteria. we analysed the contribution of information concerning the fact whether a report was ‘serious’.Chapter 3. The way the reports are presented seems not to be decisive in the selection of signals. like the completeness of the medication history or the absence or presence of detailed clinical information.16 Recently new approaches have been developed like the Bayesian Confidence Propagation Neural Network.13 The optical impact focussed the attention to specific associations. Not only the information present in the index report may have contributed to the selection of an association. is still increasing. Napke used a so-called pigeonhole system in which every incoming report was colour coded. Therefore. The quality of the reports is essential for an optimal assessment and for maintaining a high quality of the database. the level of documentation of a report also involves other aspects.14. For this reason. None of these factors. Except for the factors related to the documentation. Factors indicating the presence or absence of information about a possible dechallenge or rechallenge in the index or control report and the source of the reports.15 The development and implementation of these techniques. which are mainly based on searching for a disproportionate number of associations. the presence of a critical term. In the mid 70’s the first ideas for computer programmes were developed. the results of our study are in accordance with these findings. In our study.12 Although the definition of the various factors slightly differ.

in the event the ROR is calculated based on the first five positions of the ATC code (ATC5) and the preferred term. in the event a ROR could not be calculated. however. For this reason. which provides information concerning the occurrence of the related ADRs in chemically related substances. a ROR can not be calculated. the Reporting Odds Ratio concerning the first five positions of the ATC code (ATC5) and the high level term can be calculated. but since the assessors are familiar with the interpretation of the ROR. disproportionality is a predictive factor for selecting possible signals. cannot replace the traditional case-by-case approach. the source of the reports. For instance. Although information technology may be helpful in identifying possible signals. but the ROR can be calculated in various other ways. information is provided concerning the occurrence of the suspected ADR in chemically related substances in the database. concise information concerning the reports is presented to the assessors on an overview form. different information can be obtained. the Reporting Odds Ratio can be calculated concerning the full ATC code (ATC7) and the high level term. The ROR could not be calculated for two index cases and one control case. In all three cases the poison probability was calculated. indicating that these other approaches did not provide any additional information on top of the normal approach. time of onset of the ADR and the fact 137 .Determinants of signal selection are being developed. for which p<0. but serve as an additional tool in signal analysis. possibilities for including clinical information in this decision making process are not yet available. This is a drawback for the use of this measure. the association was considered to be disproportionate present in the database. in the event the numerator is zero. By choosing another level of aggregation with respect to the suspected drugs or ADRs. In preparation for the weekly assessment meeting where the selection of possible signals take place. coding of the suspected drug and suspected ADR. yielded similar results as the calculation of the ROR based on the full ATC code and the preferred term. it was also used in our analysis. The results of our study showed that. the ROR was calculated with respect to the full ATC code and the preferred term.19 Quantitative techniques. In the present study. Similarly. However. All these various approaches.001. a description of the event. which provides information concerning the occurrence of the related ADRs associated with the suspected drug in the database. This concerns information about the gender and age of the patient. Finally. however. in contrast to the absolute number of reports sent to the SRS.

Although slight differences with the official SPC occasionally may exist. which may cause reporting bias.Chapter 3. two standard works that are frequently used in daily practice by physicians and pharmacist were used. On other datasets differences of the performance may exist. the fact if the ADR is labelled and the seriousness of the report is available. since this may lead to an early warning of the association under concern. Historically. Signals published in the quarterly reports have a preliminary character. the performance of these factors is based on calculations on the dataset of the Netherlands Pharmacovigilance Foundation. The results of this study may be used to improve the signal selection process. The assessors may refer to the database for additional information like a more extensive description of the clinical event and the concomitant medication the patient used. selection bias may be present. Selected cases serve as a point of attention for further research or attention for other cases that might be reported. This implies that the causality of the signals mentioned in Table 1 does not necessarily have to be proven. Since 1999 also information concerning the extent of disproportionality in the Lareb database is presented as a ROR with corresponding 95% confidence interval. the number of reports on the high-level term and the standard residual value as a measure of disproportionality. Data concerning the presence of the association in the WHO database are available but are not presented in advance. After all. Although for instance information about the extent of disproportionality. coding of the reports at Lareb always required the use of these two standard works and not the official SPC text. However. but this reflects the present procedure of selection and dissemination of signals. the goal is to inform the MEB about possible new signals in a rather early stage. by making a pre-selection of associations based on the presence of one of these factors. this coding was also used in this study. Furthermore. the number of reports on the ATC code. this does not necessarily have to be unfavourable.1 whether the suspected ADR is labelled or unlabelled. Media attention of previous publication of a case report might raise the number of reports on an association. quantitative information is provided like the number of associations reported. To make a distinction between ‘unlabelled’ and ‘labelled’ associations. 138 . Although this can not completely be ruled out in individual reports.

2000:109-24. J Clin Epidemiol 1992. In: Inman WHW (ed). Farmacotherapeutisch Kompas 2000/2001. Leufkens HGM.A revised approach to international signal analysis. Edwards IR.H. Andrejak M. Lindquist M.45:1177-84. Pharmacoepidemiology and Drug Safety 1999.44:277-81. Lancaster: MTP press. Lindquist M. A view from regulatory agencies. Bakker A.44:513-8.P. Moore N. New York: John Wiley & Sons ltd. Pharmacoepidemiology. 9. Uppsala. and management. seem to have a comparable impact in the process of the selection signals to be disseminated. Fucik H.8:S15-S25. 1995. Egberts ACG. 2000. Informatorium Medicamentorum 2000 Wetenschappelijk Instituut Nederlandse Apothekers. 11. 's Gravenhage. WHO Adverse Drug Reaction Dictionary.356:1255-9. the presence of a ‘critical term’ and the fact whether or not the association under concern is labelled. References 1. Br J Clin Pharmacol 1997. Commissie Farmaceutische Hulp van het College voor Zorgverzekeringen. Non-puerperal lactation associated with antidepressant drug use. A better understanding of these factors may eventually improve the efficiency of the signal selection process. Aronson JK. 2. Edwards IR. Nunes HM. Leufkens HG. van der Kuy A (ed). (submitted for publication) Stricker B. Anonymous. Bate A. 5. Amstelveen. Van Puijenbroek EP. 6. Ståhl M. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for Adverse Drug Reactions. In: Graham DJ. Orre R. 1986. Br J Clin Pharmacol 1997. 10. Kreft-Jais C. Meyboom RH. Reports of hypoglycaemia associated with the use of ACE inhibitors and other drugs: a case/non-case study in the French pharmacovigilance system database. 139 . Begaud B. Serum sickness-like reactions to cefaclor. 3. Lancet 2000. Kurz X. Both the extent of disproportionality as well as the seriousness of the reports. WHO Centre for International Drug Monitoring. 4. 8. The results of this study revealed which factors are primarily responsible for signal selection and dissemination. Noblet C. Monitoring for Drug Safety.Determinants of signal selection Conclusion This study showed that selection of signals generated by the subjective review of data sent to a spontaneous reporting system for adverse drug reactions is based on both qualitative and quantitative aspects. Egberts AC.Ch and Tijssen J. De Koning FH. Waller P. Strom BL (eds). 7.G. Adverse drug reactions: definitions. 2000. Ollagnier M. Haramburu F. diagnosis. From association to alert .

Meyboom RHB. et al. Bate A. Methods Inf Med 1974. Methods Inf Med 1971. Characteristics of topics in pharmacovigilance in the Netherlands.47:689-93.34:47393. Hekster YA. Finney DJ.12:207-19. 13. De Freitas RM. Napke E.10:237-45. Systemic signalling of adverse reactions to drugs. Pharmacoepidemiologic approaches to the evaluation of antidepressant drugs (Thesis) Utrecht University. Transformation of a database of spontaneously reported suspected adverse drug reactions and its use as a tool in signal detection. and Leufkens HGM. 140 . In: Inman WHW (eds). Clin Drug Invest 1996. Bate A. 15. 1986:65-6. Monitoring for Drug Safety. Br J Clin Pharmacol 1999. Lansner A. Egberts ACG. A Bayesian neural network method for adverse drug reaction signal generation. Statisical logic in the monitoring of reactions to therapeutic drugs. Canada.1 12. Lansner A. and Lindquist M. de Koning GHP. Finney DJ. Bayesian neural networks with confindence estimations applied to data mining. 17. Orre R. 14. 16.Chapter 3. Edwards IR. Orre R. Olsson S. Van der Hofstede JW. In: Egberts ACG. Signaling possible drug-drug interactions in a spontaneous reporting system: Delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole. the Netherlands 1997:111-24. Meyboom RHB. Gribnau FWJ. Computational Statistics and Data Analysis 2000. Egberts ACG. Van Puijenbroek EP. Eur J Clin Pharmacol 1998. 19. Utrecht. 18. 54:315-21. Egberts ACG. Lindquist M.13:110. Meyboom RHB. Lancaster: MTP press.

3.2 PRACTICAL APPLICATION OF QUANTITATIVE SIGNAL DETECTION IN THE SPONTANEOUS REPORTING SYSTEM FOR ADVERSE DRUG REACTIONS IN THE NETHERLANDS .

2 Introduction Despite extensive research and large pre-marketing trials. i. Several approaches were introduced. for instance the signalling of changes in trend or the ‘reaction proportion signalling’ introduced by Finney.1 Health professionals (physicians and pharmacists) report suspected associations between ADRs and drugs to an SRS on a voluntary basis. which are used to monitor the safety of drugs after marketing. they are as yet not routinely applied. only limited information concerning possible adverse drug reactions (ADRs) of a drug is available at the time of marketing. In this chapter. For this reason. Methods have been developed to support the process.e.6-8 Currently. there is a clear need for a continuous and systematic surveillance of drugs for any unexpected ADRs. like the statistical quantitative analysis of the data set of an SRS. the role and position of quantitative signal detection and the way it is applied at Lareb will be described.Chapter 3. involving the comparison of records of a particular drug and a specific suspected ADR with those of a larger set of drugs and reactions. modifications of the latter approach have been developed and implemented. the reported ADR(s) and suspected drug(s) are difficult to establish. 142 . the search for unexpected associations.2-4 Over the years. such as the Proportional ADR Reporting Ratio used by the Medicines Control Agency5 and the Reporting Odds Ratio. In the last few decades spontaneous reporting systems (SRSs). At the Netherlands Pharmacovigilance Foundation Lareb quantitative signal detection and subsequent evaluation of signals has been used in various ways for a number of years.8-11 Despite the increasing popularity of these new approaches.5. Such quantitative analyses can provide additional information concerning a possible relationship between a suspected ADR and a drug. have earned a reputation for offering a fast and reasonably efficient (and cost-effective) way of detecting ADRs. is usually carried out by a systematic manual review of all reports sent to an SRS. This procedure is time consuming and more complex associations between patient characteristics. several such (modified) quantitative approaches are being used by a number of SRS centres as well as the WHO Collaborating Centre for International Drug Monitoring (the Uppsala Monitoring Centre). The process of signal detection.

6-8 Based on the classical 2x2 contingency table. Here. For the analysis of disproportionality a case/non-case design is used. If the suspected ADR is mentioned in the Dutch textbooks ‘Farmacotherapeutisch Kompas’ and/or the ‘Informatorium Medicamentorum’ the ADR is considered labelled. the following values are calculated (see Figure 1). and subsequently filed in a database.14 All new reports are discussed on a case-by-case basis in weekly assessment meetings. along with patient characteristics. Cell b represents the number of reports concerning the suspected drug but now in association with other possible ADRs and cell c the number of reports concerning the suspected ADR 143 . and if this is not the case it is classified as unlabelled. The 2x2 contingency table for calculation of disproportionality. potentially relevant associations (signals) are selected for dissemination to the Medicines Evaluation Board and/or publication in national and international journals. Cell a represents the number of reports in which the same combination of suspected drug and suspected ADR was mentioned. is subjected to review by qualified assessors. The extent to which the association between ADR and suspected drug stands out with respect to its background frequency in the database a Reporting Odds Ratio (ROR) is calculated. Data concerning the suspected ADR(s) and drug(s) are coded using the WHO Adverse Drug Reaction terminology12 and the Anatomical Therapeutic Chemical (ATC) classification system. respectively. Every incoming report represents an association between one or more suspected ADRs and one or more suspected drugs.Application of quantitative signal detection Quantitative signal detection as applied by the Netherlands Pharmacovigilance Foundation Lareb Each report received by the Netherlands Pharmacovigilance Foundation Lareb. Reports with the suspected ADR Reports with the suspected drug All other reports a c Reports without the suspected ADR b d Figure 1. for each association of drug and ADR.13. which institute manages and maintains the national SRS on behalf of the Dutch Medicines Evaluation Board.

15 Although several measures of disproportionality can be used. like drug-drug interactions and drug-related syndromes. for datamining purposes. all having their own specific applicability. Finally.11 The purpose of the quantitative signal detection as applied at the Netherlands Pharmacovigilance Foundation is threefold. For instance the WHO Centre for International Drug Monitoring (Uppsala Monitoring Centre (UMC) uses a so-called Bayesian Confidence Propagation Neural Network (BCPNN) analysis to isolate those associations that stand out as being different to the generality of the database.16 validation studies are few and far between. 144 . cell d reflects the number of reports concerning other drugs associated with other ADRs. the results of the statistical analysis are used as an aid in the traditional case-by-case review. Finally.2 associated with other drugs. quantitative signal detection is used for evaluation purposes to study signals in more detail. First. if more than three reports are included (Chapter 1. Recently a validation study had been carried out concerning the aforementioned BCPNN analysis of the WHO. The 95% confidence interval is calculated by: 95 % CI = e ln( ROR ) ± 1 . for example by adjusting for covariates or analysing more complex relationships. Secondly.1). on an ad hoc basis. a list of associations of ADRs and suspected drugs that are disproportionately present in the database is generated periodically. when applied to the Lareb dataset. 96 æ 1 1 1 1 ö ç + + + ÷ è a b c d ø At the Netherlands Pharmacovigilance Foundation the ROR is used to analyse the extent of disproportionality. The ROR is defined as the product of the exposure odds among the cases with respect to the exposure odds among the non-cases and is calculated by (a*d)/(b*c). this approach yielded a positive predictive value of 44% and a negative predictive value of 85% in the detection of signals as compared to reference literature sources. When used for the analysis of the WHO database. Since there is no golden standard concerning the various measures of disproportionality. Several approaches for calculation of the extent of disproportionality are currently available.Chapter 3. the results of the various approaches do not essentially differ.

an ROR is calculated on the basis of the number of reports with the generic compound (expressed as the entire ATC code) and the suspected ADR. The second ROR refers to the association between angioedema and ACE inhibitors in general and is also statistically significant (12.0-17. To allow a more in-depth discussion of the reports. suspected ADR and the source of the report (either physician or pharmacist). In addition.0. implying that ACE inhibitors are indeed generally associated with this ADR. additional information regarding the association between drugs and ADR(s) reported in the literature and details with respect to concomitant medication used by the patient. the suspected drug. representing the extent of disproportionality for chemically-related substances. In the overview form depicted in Figure 2 the results of the quantitative signal detection of the Lareb database are shown. Firstly. implying that angioedema is significantly associated with enalapril in reference to other reports in the database.Application of quantitative signal detection Quantitative signal detection as an aid in the traditional case-by-case review In the case-by-case review the results of the quantitative approach serve as an additional source of information. In addition to information regarding the patient’s age and gender. a second ROR is calculated based on the first five positions of the ATC code. 95% CI 8.2). Figure 2 shows an example of such an overview form.2). for instance groups of beta-blocking agents or ACE inhibitors. The first ROR shows the extent to which ‘enalapril’ is associated with reports of angioedema.0. additional information may be generated if the information on the overview forms is not sufficient to enable an optimal assessment. 145 . the results of the quantitative analysis are presented. two different RORs are calculated for this quantitative signal detection. Routinely. This ROR is statistically significant (11. 95% CI 7. For the weekly assessment meetings a first impression of the drug-ADR associations is compiled using overview forms containing concise information on those reports that have been received in a particular period.3-17. This information generally concerns more detailed data derived from the original reports sent in by the physicians or pharmacists.

b and c) of the contingency table are shown.17 The UMC database contains summarised information regarding all reports on combinations of drugs and ADRs.000). two Reporting Odds Ratios with the corresponding numbers of the cells (a. Among other data. respectively.15) 114 / 1476 / 281 11.99 (7.000. quantitative signal detection can also be used more actively as a datamining tool.17) Figure 2. overviews with disproportionate associations can be generated. Quantitative signal detection as a datamining tool Apart from having a function in case-by-case analyses. By screening the database.34-17.Event 1: Labelled: yes (a/b/c) ATC 7-WHO preferred term: ATC 5-WHO preferred term: ATC code: C09AA02ENALAPRIL/ENALAPRILATE WHO preferred term: ANGIOEDEMA ROR (95% confidence interval) 59 / 708 / 336 10. expressed as the ATC code and the ‘preferred term’ of the WHO ADR terminology.04-17. For this reason. the total number of Lareb reports in the WHO-UMC database is relatively small (approximately 30. However.000) compared to the total number of reports in the WHO database (approximately 2. In this way the Lareb database is supplemented with quantitative data from a second source. which thus provides additional insight into the existence of possible signals. Besides the results of the calculations based on the data set of the Netherlands Pharmacovigilance Foundation.2 Report number: 30406 Date of entry: Jan 25th 2001 Source: General Practitioner Gender: F Region: Central districts of the Netherlands Age: 51 Seriousness: not reported Drug 1: RENITEC TABLET 10MG Event 1: swelling of cheek and lips: angioedema Drug 1 . so the reports are filed in both databases. On a quarterly basis. The reports from Lareb are in turn sent to the Uppsala Monitoring Centre.Chapter 3.97 (8. Example of an overview form used in the assessment meetings at Lareb. it is unlikely that the Dutch reports will have a great influence on the calculations based on the data set of the UMC. the quantitative data of each drug-ADR combination received by the UMC are sent to the national pharmacovigilance centres. which implies that the two sources of information can be regarded as independent. the outcome of the quantitative signal detection on the UMC data set can be utilised. A disproportionate association may represent a signal and should consequently be 146 .

As an illustration (see Table 1).Application of quantitative signal detection analysed in more detail. An example is the clustering between drugs in the search of possible drug-drug interactions. Other filters may. for instance. Lareb has filtered out all reports received between January 1st 1990 and November 1st 2000 in which a suspected association with angioedema was mentioned. With the datamining technique one or more additional filters can be applied.19 Another example is the analysis of syndromes.20 These approaches have been described in more detail in Chapters 2. Such statistical analyses can not only be used for the evaluation of signals. but can also be used for the routine screening of the database for drug-drug interactions and syndromes. one or more (suspected) drugs and one or more suspected ADRs. which may represent a possible drug-drug interaction.18. as well as the fact whether the association is labelled or unlabelled. angioedema was found to be mainly associated with ACE inhibitors and angiotensin II inhibitors. in descending order.4. 147 . the clustering of suspected ADRs is analysed. in which.15 This approach can be used to identify disproportionate associations that may have been missed by the traditional case-by-case approach. similar to the detection of drug-drug interactions. The analysis of more complex relations such as drug-drug interactions and drug-related syndromes A report sent to an SRS contains information about one patient. be used to trace a specific drug or a suspected ADR. all these factors may be interrelated and since these complex relationships may not always be transparent statistical analyses are often indispensable. Logistic regression analyses can be used to analyse this clustering by evaluating interaction terms in the logistic model. As expected.3. For each combination of two drugs it can be calculated whether the number of observed ADRs exceeds the number of expected ADRs (Figure 3A).2 to 2. However. For example. The associations are sorted by the lower limit of the confidence interval of the ROR. In addition to the drug and suspected ADR the number of reports in the cells of the aforementioned contingency table are shown. filtering the unlabelled cases allows a selection to be made of associations that need to be monitored more closely because they may represent unknown ADRs.

7 1.9 13.5 25.1 6.2 18.4 5.3 4.7 14.7 8.4 52.2 9.6 4.3 5.2 24.2 11.2 14.2 49.0 1.5 5.6 12.4 10.9 11.0 3.4 8 7 50 17 3 5 14 15 3 4 3 4 3 19 24 346 159 16 38 165 246 31 56 41 71 55 Labelled / unlabelled ROR Lower limit Upper limit 95% CI 95% CI a b c 307 308 265 298 312 310 301 300 312 311 312 311 312 The table shows disproportionality expressed as the ROR with corresponding 95% confidence interval and the number of reports in cells a.8 2.9 17.5 14.1 4.6 1. Example of the datamining approach Drug fosinopril enalapril with diuretics enalapril/enalaprilate lisinopril losartan with diuretics quinapril/quinaprilate captopril losartan irbesartan paracetamol isosorbide dinitrate rofecoxib acetylsalicylic acid labelled labelled labelled labelled labelled labelled labelled labelled unlabelled labelled unlabelled labelled labelled 32.3 75.4 15. In this example.Table 1. b and c of the contingency table of various combinations of angioedema and reported drugs.8 7.3 2.9 8.8 2. 148 .2 4.7 9. the associations are sorted by the lower limit of the confidence interval in descending order.7 4.8 22.

The interface displays the cross-section of the cubes represented in Figure 3. however. ADR Drug A Drug ADR A Drug B ADR B A B Figure 3. a graphical interface has been developed that facilitates a fast and efficient screening of the database for possible drug-drug interactions or drug-related syndromes. 149 . Datamining approach for (A) the detection of drug-drug interactions and (B) drug-related syndromes In collaboration with the Prevention and Health department of the Netherlands Organisation for Applied Scientific Research (TNO). It is important. An example of output generated by this interface is shown in Figure 4. the approach can be applied to screen the database for drug-related syndromes. These calculations can be carried out for the various ADRs in the database. in the analysis of drug-drug interactions all drugs that are used at the moment of occurrence of the event are included in the calculations.Application of quantitative signal detection Although the reporting health professionals usually will make a distinction between suspected and concomitant medication. Similar to the detection of drug-drug interactions. represented by the cross-sections in Figure 3A. to choose an ADR that is frequently associated with both types of drugs. In this analysis the calculations in the cross-sections concern the combination of two ADRs in association with one particular drug that is more frequently reported than usual.

In addition.2. If an apparent drug-drug interaction is in fact caused by an increased susceptibility for developing an ADR in certain patients.2 Combination of itraconazole and oral contraceptives Itraconazole only Figure 4. For various combinations of two types of drugs. other drugs will possibly show the same type of drug-drug interaction in this category of patients since the same confounding factor. The intensity of the colour in the cells represents the level of statistical magnitude of the interaction. Example of the outcome of a drug-drug interaction analysis performed with the datamining interface. patients with a similar condition. The strength of the interaction between the two types of drugs is represented by the intensity of the colour. The concept of the detection of drug-drug interactions may also be used to determine possible risk factors for developing certain ADRs. i. The details of the possible itraconazolecontraceptive interaction in association with delayed withdrawal bleeding are described in more detail in Chapter 3. The intensity of the colour in the marginal cells reflects the level of statistical significance for each drug associated with the ADR separately. the chance of a delayed withdrawal bleeding was examined. 150 . is involved in both cases. Distinguishing between a drug-drug interaction and the increased risk of developing an ADR due to the underlying disease may be rather difficult.e. information concerning the various calculations is displayed.Chapter 3.

Application of quantitative signal detection Tracheal sympaticomimetc drugs Other tracheal drugs Non-steroidal antiinflammatory drugs Figure 5. and NSAIDs will in all probability increase the number of reported ADRs involving the respiratory system. that an actual drug-drug interaction between NSAIDs and both groups of unrelated drugs does exist. Example of an interaction analysis of possible risk factors. the analysis described above constitutes a valuable contribution to the detection of the 151 . however. if the analysis of several chemically or pharmacologically unrelated drugs show an apparent ‘drug-drug interaction’ in relation to one or more ADRs. The strength of the interaction between the drugs is indicated by the intensity of the colour. Thus. it is more likely that a specific type of patient has an increased risk of developing one or more of these ADRs. as well as other inhaled medication. It is unlikely. The concomitant use of inhalation sympaticomimetic drugs. An example is the increased risk of developing respiratory ADRs in patients with asthma taking non-steroidal anti-inflammatory drugs (NSAIDs). using the datamining interface.21 The results of such an analysis produced by means of the above-mentioned computer interface are shown in Figure 5. However.

Human analysis and interpretation of the data still is the decisive factor in the signal detection process. such as clinical information and findings reported in the literature (see Figure 6). 152 . it can be stated that an association that is not disproportionately present in the database can still represent a true signal while a disproportionately present association on the other hand may not indicate a true signal. thus providing information from different perspectives. possibly in the near future. the results of the dataset analyses should be regarded as additional sources of information and complementary to the traditional analysis techniques. is currently gaining ground. Again. The results of a quantitative approach should be considered as additional information that is to be interpreted in combination with other sources. genotype. an intelligent interpretation of the various combined aspects is required before a conclusion about a possible relationship between suspected drug and ADR can be drawn. gender and. such as age. Integrating information Finney stated that the essence in ADR monitoring is to collect facts that individually tell little but that collectively form a fabric of clues to drug dangers. Integration with clinical and pharmacological assessments still remains an important principle in signal detection. Following this line of reasoning.Chapter 3. Bringing together information from different sources is of great benefit to the analysing process. By choosing different reference groups comparisons with similar drugs or related ADRs can easily be made. The aforementioned concept can be extended to the identification of other risk factors. The method of statistical analysis of SRS data sets.2 increased danger of respiratory ADRs occurring in patients suffering from asthma or Chronic Obstructive Pulmonary Diseases (COPD) when using NSAIDs.2 Subsequently. At the Netherlands Pharmacovigilance Foundation a quantitative approach is routinely applied in various ways. or quantitative signal detection.

Br J Clin Pharmacol 1997. 2. The detection of adverse reactions to therapeutic drugs.45:1177-84. Meyboom RHB. Anello C. Finney DJ.Application of quantitative signal detection Literature Health professional Reports sent to SRS Medicines Evaluation Board Overview forms and detailed reports Possible signals Science Signal detection Quantitative signal detection WHO combinations database Figure 6.19:3199209. Statistical logic in the monitoring of reactions to therapeutic drugs. and signal dissemination routes at Lareb. O'Neill RT. Egberts ACG. Discovery of adverse drug reactions. Overview of information sources used for signal detection. 8.10:237-45. Reports of hypoglycaemia associated with the use of ACE inhibitors and other drugs: a case/non-case study in the French pharmacovigilance system database. Systemic signalling of adverse reactions to drugs. Leufkens HGM. Methods Inf Med 1971. A comparison of selected phase IV studies with spontaneous reporting methods. Stricker BHC.13:110.1:153-61. 7. 6. Mendelis PS. 5.44:513-8. Haramburu F. Bakker A. Ollagnier M. Br J Clin Pharmacol 1997.44:277-81. 4. Graham CF. Stat Med 1982. Evans S. Moore N. Noblet C. Pharmacovigilance: a science or fielding emergencies? Stat Med 2000. Tijssen JGP Serum sickness-like reactions to cefaclor. Stanley GR. Knapp DE. Finney DJ. Finney DJ. 153 . Begaud B. J Clin Epidemiol 1992. Andrejak M. 3.249:2226-8. Non-puerperal lactation associated with antidepressant drug use. Rossi AC. Methods Inf Med 1974. de Koning GHP. References 1. Kreft-Jais C. JAMA 1983.

56:733. 10. Leufkens HG. Roberts LJ. Knapp HR. Fucik H.23:533-42. (submitted). Egberts AC. Clinical implications of prostaglandin and thromboxane A2 formation (1). 21. van der Kuy A (ed). Transformation of a database of spontaneously reported suspected adverse drug reactions and its use as a tool in signal detection In: Egberts. Lindquist M. A retrospective evaluation of a data mining approach to aid finding new adverse drug reaction signals in the WHO international database. 11. 's Gravenhage. A Bayesian neural network method for adverse drug reaction signal generation. WHO Adverse Drug Reaction Dictionary. Association between terbinafine and arthralgia. 2000.2 Egberts ACG. Br J Clin Pharmacol 1999. Bate A. Uppsala. 12.-8. Farmacotherapeutisch Kompas 2000/2001.47:68993. Commissie Farmaceutische Hulp van het College voor Zorgverzekeringen. Egberts ACG. 9. Meyboom RHB. Nunes HM. Egberts AC. Signalling possible drugdrug interactions in a spontaneous reporting system: delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole. 17. Informatorium Medicamentorum 2000. Orre R. van Puijenbroek EP.319:689-98. Lindquist M. Heerdink ER. Wetenschappelijk Instituut Nederlandse Apothekers. fever and urticaria: Symptoms or syndrome? Pharmacoepidemiol Drug Saf 2001. Edwards IR. Van der Hofstede JW. Bayesian data mining in large frequency tables. 14. Drug Saf 2000. WHO Centre for International Drug Monitoring. with an application to the FDA spontaneous reprting system.53:177-89. Bate A. Jackson EK. From association to alert . Edwards IR. 18. Eur J Clin Pharmacol 1998. Edwards IR. Anonymous. 13. N Engl J Med 1988. Eur J Clin Pharmacol 2000.10:135-142. the Netherlands.Chapter 3. Amstelveen. Meyboom RHB. Pharmacoepidemiologic approaches to the evaluation of antidepressant drugs (Thesis) Utrecht University. 154 . 19. Lindquist M. Utrecht. van Puijenbroek EP. The American Statistician 1999. van Puijenbroek EP. 2000. DuMouchel W.54:315-21. Bate A.1997:111-24. Oates JA. Leufkens HG.A revised approach to international signal analysis.8:S15S25. Ståhl M. Egberts ACG. Meyboom RH. 16. FitzGerald GA. Orre R. 1995. Detecting drug-drug interactions using a database for spontaneous adverse drug reactions: an example with diuretics and non-steroidal anti-inflammatory drugs. Lansner A. Lindquist M. Branch RA. Meyboom RH. Leufkens HGM. Olsson S. Pharmacoepidemiol Drug Saf 1999. A comparison of measures of disproportionality for signal detection in spontaneous reporting systems for adverse drug reactions. Ståhl M. ACG. 15. Leufkens HGM. van Puijenbroek EP. De Freitas RM. 20.

GENERAL DISCUSSION .

which necessitated adjustments for covariates such as age. showed that differences in upper gastrointestinal complaints associated with the use of various NSAIDs reported to the Food and Drug Administration were considerably reduced after adjustment for the underlying reporting rates. a limited number of SRSs and the WHO Monitoring Centre for International Drug Monitoring have acquired experience with quantitative or statistical approaches. Lareb runs and maintains a regionalised spontaneous reporting system. but the first steps towards its establishment had already been taken as early as 1984.2 As a result of new European regulations. Lareb. the Lareb methodology serves as a tool facilitating the improvement of pharmacotherapy by providing the reporting physicians and pharmacists with customised feedback. the source of the reports or the number of prescriptions dispensed. in some form or another. always played a role in pharmacovigilance. in their 1987 study. although this has mainly been confined to the association of a single drug and one particular adverse drug reaction (ADR). The initial concepts and procedures of Lareb have been described in more detail by De Koning. it was not until recently that developments in information technology enabled the routine application of these techniques. information from spontaneous reporting systems (SRSs) concerning associations between drug(s) and unwanted clinical events was expressed in terms of the absolute number of reports or crude estimates. The Netherlands Pharmacovigilance Foundation Lareb was officially founded in 1991. In the mid 1990s Egberts had already laid a firm basis for a quantitative approach at Lareb in which period he also reported his findings on studies into the possibilities for the use of additional sources of information in the signal detection 156 .1 Although the first theoretical concepts of the quantitative approach to SRS data were already described in the mid 1960s. was appointed as the national spontaneous reporting centre for ADRs in the Netherlands in 1995. gender. the need for a more considered interpretation of the data became more urgent. In the past two decades. Besides its function as an ‘early warning’ system. who. One of the characteristics of Lareb is the short communication lines it maintains with the reporting health professionals. Gradually. acting on behalf of the Medicines Evaluation Board (MEB). In the dawning of pharmacovigilance.General discussion Introduction Quantitative signal detection has. This was illustrated by Rossi et al.

syndromes and covariate-drug interactions. additional quantitative techniques have been developed and applied to supplement the classical case-by-case approach in order to increase the sensitivity of the system in picking up new and relevant signals.General discussion process. The selection of signals generated by the subjective review of data in a case-by-case analysis appeared to be based on both qualitative and quantitative aspects.4 Due to their combined efforts the Netherlands Pharmacovigilance Foundation now had a solid basis for further enhancements of their signal detection process. It does not only facilitate the detection and analysis of unknown associations between a single drug and a single event. these quantitative approaches can be more sensitive in detecting complex patterns within large sets of reported suspected drugs and ADRs than people whose mental capacity is too limited to recognise these hidden relationships. The results of the quantitative approaches applied in various centres appeared to be broadly comparable when more than three cases per combination of suspected drug and ADR were collected. More recently. The extent of disproportionality. proved to be more important than the number of reports received per combination of suspected drug and ADR. suggestions for further research are made. 157 . the detection of drug-drug interactions and drugrelated syndromes. in Part Three the main determinants of signal detection were analysed and the application of quantitative signal detection at the Netherlands Pharmacovigilance Foundation Lareb was described. however. The studies presented in this thesis have shown that the quantitative approach is an essential part of the analysis procedures applied by the Netherlands Pharmacovigilance Foundation. In this fourth and final part. Furthermore. it also offers possibilities for the analysis of more complex relations such as drug-drug interactions and drugrelated syndromes. The first part of this thesis focused on methodological and statistical aspects of quantitative signal detection.3 The qualitative theoretical concepts of pharmacovigilance were subsequently elaborated by Meyboom. Finally. In Part Two examples were given concerning the analysis of the relationship between a single drug and a suspected ADR. Additionally. additional methodological considerations and practical implications of the use of quantitative signal detection in pharmacovigilance will be discussed. Additionally. we showed that non-selective underreporting does not influence the analysis of drugdrug interactions.

it is exactly this dualistic character that is responsible for both their strengths and their weaknesses. However. 158 . Meyboom et al. current information technology allows calculations to be made in a short time span and additionally permits concurrent adjustments for various covariates like age. but also to quantitative signal detection.5 This not only applies to the classical case-by-case analysis. thoughtful interpretation will continue to be of vital importance. as the next step in the detection and analysis of signals.General discussion The added value of quantitative signal detection Quantitative approaches are becoming increasingly important in signal detection. which method has been an important tool in the analysis of SRSs for many years. stated that each method applied in pharmacovigilance (e. On the other hand. since calculations are not hampered by subjective information. In other words. a quantitative approach should be viewed as an additional source of information. not as a substitute for the classical case-by-case analysis. it should be considered as a filtering mechanism to focus attention on those associations that are most likely to represent true signals. Moreover. spontaneous reporting. This is not to say that they detract from case-by-case analyses. However. they are limited in their ability to provide quantitative information. Case reports or case series resulting from the former approach are highly sensitive in picking up qualitative signals. As often with the introduction of new approaches. prescription event monitoring or case control surveillance) follows a more or less individual approach to signal detection. there is the risk that they are regarded as a panacea for the limitations inherent to the prevailing approach. When used as a datamining tool.g. Quantitative approaches in signal detection unify the qualitative and quantitative aspects of detection. For the analysis of possible signals resulting from a caseby-case analysis it merely provides additional information. gender and concomitant medication. in which each report is individually assessed as to whether or not the reported association represents a signal. A quantitative approach is objective. The main disadvantage of a quantitative approach is the fact that clinical information can only be taken into account to a limited extent.

7.000 patients the certainty of finding at least one case of an ADR with a true incidence rate of 1 in 1. For example. and covariates are generally difficult to control for.8 The likelihood of an ADR being detected depends on the frequency of the ADR.6 This is. specific doses of the drug under investigation are prescribed for limited periods of time. Additionally. clinical trials are limited in their power to detect ADRs. the number of patients included in the trial and the background incidence of the ADR. 159 .General discussion Methodological considerations As stated previously. these systems have proven to be relatively efficient in the surveillance of drugs for possible harmful effects. three different populations can be distinguished (see Figure 1). i. among other reasons.g.000 would be more than 99%.9 Although spontaneous reporting systems suffer from certain methodological problems.e. with a sample size of 5. the sample size needed to detect this ADR would increase dramatically. In the event the ADR cannot be clearly differentiated from the natural background incidence. makes them unsuited for the detection of ADRs with a long latency time. the elderly and those with significant concurrent diseases. Validity With respect to the interpretation of the outcome of quantitative analyses of SRS databases by means of measures of disproportionality. the structured nature of trials. however. caused by the small number of patients that are studied before the drug is marketed and by the frequent exclusion of patients that may be more susceptible to developing ADR(s). e.

The second population consists of a smaller proportion of this group of patients. namely those who experience an ADR (X'). however. reporting of the drug is represented by k2. In quantitative analysis. for the practising health professional it may be important to have information about the occurrence of an ADR among patients in the drug using population (X) at his or her disposal. Still. an estimation of the level of disproportionality in the ADR experiencing population (X') can be made. Based on these calculations. Finally. The SRS. the third and by far smallest population is constituted by the number of patients of the latter group of whom the suspected ADR is actually reported (X''). the measure of disproportionality expresses the relationship between drugs and suspected ADRs within the ADR reported population (X'').General discussion Drug using population X ADR experiencing population X' ADR reported population X'' k1 incidence ADR k2 reporting ratio k1 k2 Figure 1. The different populations relevant for the interpretation of spontaneous reporting The largest population represents patients that actually use the suspected drug (X). The incidence of the drug is represented in Figure 1 by k1. is primarily designed to generate hypotheses of these possible 160 .

This information should not be considered as having an exact meaning. and may 161 . If. calculations based on measures of disproportionality would yield no differences. being part of the reference group. the results of the calculations cannot always be generalised to the population of patients using the drug (X). liver-function disorders are over-represented in a certain pharmacological group. the frequency of an adverse reaction in association with a certain drug is being compared with the other reports in the database.11 Other reports in the database are being used as a measure of the ‘background incidence’ of the relevant events. and it cannot provide accurate information about the incidence of the ADR. All other reports in the database are commonly used for reference purposes.General discussion effects. the association between various factors is expressed as a point estimate with corresponding confidence interval.10 The rationale for using this approach is that the composition of different types of reactions in the data set of an SRS is relatively constant. the number of ADRs needed for the drug in the index group to yield a statistically significant result will also increase. Although this is a large group. There is a chance that under these circumstances rare liver-function disorders associated with a new drug will be less easily detected. An example may illustrate the misinterpretation that may occur if results of the data of SRS are considered indicative of the incidence of ADRs. which has the advantage that smaller confidence intervals can be calculated. In other words. If we assume that two drugs have a similar pattern of ADRs. This problem of so-called underreporting refers to factor k2 in Figure 1. even if different incidences of the ADRs exist in the patient population that is using the drugs. like the incidence of the ADR. By calculating a measure of disproportionality. since the main goal of an SRS is to detect possible signals and not to estimate the incidence of the ADRs involved. In quantitative signal detection. Only a small proportion of the ADRs occurring in daily practice will be reported. but rather as a tool that facilitates the interpretation of the data in the context of other reports and co-factors in the database. interpretation of the quantitative results based on data sets of spontaneous reporting systems should be carried out by assessors who are experienced in this field and who have an expert knowledge of the composition of the database. Therefore. for instance. like the Reporting Odds Ratio. its composition is also heterogeneous. This does not necessarily have to be a problem. This implies that great care should be taken in choosing the reference group.

showed that general practitioners tend to only report 1 out of 24.322 ADRs to the pharmacovigilance centre. patient characteristics or the concomitant medication may act as a confounding factor.17 In the analysis of drug-drug interactions for instance. The best estimate may be obtained by an in-depth study of a sample of the relevant population. Stephens gave an overview of the various factors responsible for this phenomenon. drug-drug interaction. since it may contribute to the detection of risk factors for developing ADRs as was discussed in Chapter 3. In the latter study. time to onset and the presence of risk factors.. Reports sent to an SRS do not always represent true ADRs (so-called misclassification).14 A study by Moride et al. This does not necessarily have to be unfavourable. Several other types of selection bias may also jeopardise the validity of conclusions.16 If media attention is focussed on the specific association of the drug and the ADR. Also miscellaneous factors like legal implications may be involved or the doctor’s familiarity with the regulations. however.9 Underreporting can either be non-differential or differential. calculations of RORs are not necessarily influenced by non-differential underreporting but may be influenced by differential underreporting.2. the severity of the reaction or the unusual nature of the events). e. however. They may also represent symptoms or diseases incurred because of other reasons that cannot be distinguished from true ADRs on clinical 162 .2. the underreporting was lowest for serious and unlabelled ADRs and for newly marketed drugs. no general indication of the extent to which it occurs can be given. vary widely.General discussion be the cause of selection bias. For instance. this differential underreporting or bias will influence the ROR. unfortunately. a Swedish study showed that 80% of the cases of osteitis occurring after BCG vaccination were reported. which has a similar influence on both the numerator and denominator of the ROR. therefore. drug-related syndromes and drug-covariate interactions. This holds for the analysis of associations between ADR and suspected drug. An example of non-differential bias is media attention for a specific drug.15 Underreporting is inherent to spontaneous reporting and.9 Examples are factors associated with the drug.12 or factors associated with the ADR (e.13 Estimations of the degree of reporting. for example the possibility that the characteristics of reported cases differ from the non-reported cases in terms of severity of the ADR.g.g. the length of time the drug is on the market. As shown in Chapter 1.

for which it should be decided whether it represents a possible signal. like type B effects. findings in the literature. and false positive and false negative signals in order to establish an optimal signal-to-noise ratio. Factors involved in the analysis of possible signals An SRS should be considered as a diagnostic tool for health authorities. This decision is based on weighing clinical interpretation of the reports (and their context).e.17 Therefore. As was shown in Chapter 3. A large number of false positive signals on the other hand may indirectly cause harm in that they may withhold 163 . the results of which can be used to weigh benefit and safety of the drugs involved. decreased specificity) may occur.1.General discussion grounds. unless the symptom is very common. even for more common effects the number of coincidental reports remains low. the results of the quantitative analysis and other data available (e. Bégaud showed that the probability of a significant number of coincidental associations between a drug and suspected ADR is nil or remains low. False negative signals are potentially harmful since important ADRs may be missed. however. these various factors also play a role in the selection and dissemination of signals to the MEB. For rare ADRs. it is highly unlikely that more than three coincidental reports will be received. underreporting does not pose a fundamental threat to signal detection. health professionals. Clinical information Other data Quantitative analysis Figure 2. see Figure 2). For this reason some drug-event combinations may be purely coincidental18 and false positive signals (i.g. A balance must be found between the sensitivity. Due to the problem of underreporting. specificity. Practical considerations Every new case sent to an SRS concerns a possible association between suspected clinical events and drug(s). and the industry.

22 Apart from the ‘prescription-only’ drugs. as is the case in the treatment of AIDS or the eradication of Helicobacter Pylori. frequent drug-drug interactions occurred.23 the investigation of this type of interactions constitute a challenging new task for pharmacovigilance. the threshold in quantitative signal detection must be established very precisely.General discussion patients from an effective treatment. Examples are NSAIDs or H2 blocking drugs that may interact with prescribed drugs. the risk increases with the number of drugs prescribed and taken concurrently. and furthermore.10 Similar to the detection of ADRs. Because of the characteristics of the population in which mibefradil was used. the number of drug-drug interactions is likely to increase. From premarketing studies it was already known that mibefradil induced cytochrome P450 3A4 and 2D6. one of which was with HMG-co-A inhibitors. which may have serious implications for the drug(s) involved. Type B effects are related to 164 . Since it has been estimated that between 6 and 30 percent of all ADRs are due to drug-drug interactions.21.19 Eventually. are dose-related and tend to be fairly common. drug-drug interactions should be searched for in large databases that contain reliable information on all drugs being used by a particular patient. For this reason.3 have shown that data sets of spontaneous reporting systems can be used in the detection and analysis of drug-drug interactions. Another aspect that needs particular attention in the detection of drug-drug interactions is the choice of the suspected ADR(s). in the Netherlands many others are available as ‘over the counter’ medication. these frequent interactions caused this drug to be withdrawn from the market. An example is mibefradil. which was introduced in 1997. They represent the largest group of ADRs. drug-drug interactions may only be detected after drugs have been marketed. When a drug-drug interaction is already known and is monitored for in the pharmacist or physician information systems he or she is unlikely to report the interaction and thus selective underreporting may occur.2 and 2. the number of possible drugdrug interactions occurring is also likely to rise. Ideally. Since the relative number of elderly people is growing and since an increasing number of younger patients are taking combinations of drugs. Generally.24 Type A ADRs are related to the pharmacological effects of the drug. a distinction is made between various types of ADRs. The studies reported in Chapter 2.20 The elderly are at particular risk of adverse drug-drug interactions. Due to the growing availability of more potent drugs acting on fundamental biochemical pathways or receptors.

together. For this reason. the ADR should be associated with that particular drug. however.4 we showed that SRS data sets can be used in the detection and analysis of possible drug-related syndromes. constitute the picture of a disease. In the event that a possible drug-drug interaction causes a decreased effect of one of the drugs. A syndrome can be seen as a complex of signs and symptoms that. The net effect of the possible interaction. In a drug-drug interaction. in which a delayed withdrawal bleeding is taken to be indicative of a possible interaction. In the detection of drug-drug interactions.27 Another example is the serotonin syndrome associated with the use of various drug agents affecting the serotonergic system. They may concern allergic or idiosyncratic reactions and occur only in a minority of the patients.General discussion patient characteristics.2). and secondly to ensure that the combination of both drugs is mentioned in a sufficient number of reports. This was illustrated in Chapter 2.3 where the influence of concomitant use of diuretics and nonsteroidal anti-inflammatory drugs (NSAIDs) on symptoms indicating a decreased efficacy of diuretics was examined. since it determines the choice of the ADR(s) used for the detection of the interaction. An example is the possible interaction between oral contraceptives and the use of itraconazole (see Chapter 2. If an increased effect of one of the drugs is expected. among which the serotonin re-uptake inhibitors. each of which may lead to either an increase or decrease in the effect of one of the interacting drugs. both a pharmaceutical and a pharmacokinetic interaction as well as a pharmacodynamic interaction may be involved.25 Examples of drug-related syndromes are the oculomucocutaneous syndrome associated with the use of practolol26 and the possible association of Reye’s syndrome and aspirin use. An example is the possible association between oral contraceptives and breast tumours. the ADR used to identify the drug-drug interaction should be indicative of a lack of efficacy of one of the drugs involved. In Chapter 2. in the detection of ADRs in a spontaneous reporting system it is of less significance. it is essential to choose a relatively common ADR. The choice should be well founded to ensure that the ADR is dose-related with the pharmacological effects of the drugs involved.28 165 . is essential. Type C effects refer to the change in frequency of natural diseases. type B and type C effects are less suited for the detection of drug-drug interactions. Although the mechanism of action is important for an understanding of the nature of the interaction.

Recommendations for further research Studies into the validation of quantitative approaches are still sparse. publication of case reports in national and international journals is not primarily intended to give an early warning.29 The problem with validation studies is that they are hampered by the fact that there is. One of the possible approaches could be comparison with case reports of new ADRs published in the literature. the validity of a quantitative approach strongly depends on the data set to which it is applied. do not represent a particular clinical entity. however. no true gold standard available. 166 . Additionally.29 This implies that the goals of publication and signal detection in an SRS differ. also the implementation of data on drug utilisation would be useful to correct for the number of prescriptions of the drugs involved. although recently such a validation study has been conducted on the Bayesian Propagation Neural Network analysis of the WHO. Another possibility would be setting up an expert panel that is to decide whether or not the associations reported do indeed represent possible signals. validation of the quantitative signal detection within the Lareb data set and those of other SRSs needs to be corroborated by additional. The quantitative analyses of the data sets of SRSs give an impression about the occurrence of ADRs within the population of patients experiencing an ADR during the use of a certain drug. that similar results to those of the validation study on the WHO database will be found for the case/non-case design of the Lareb database. though closely interrelated. Rather than being part of a certain clinical syndrome. Nevertheless. has shown that the results of the various approaches are broadly comparable if associations are based on more than three reports (Chapter 1.General discussion Only a small part of the ADR clusters present in the database actually represent distinct clinical syndromes. in which the various methods of quantitative signal detection were compared. Therefore. an apparent clustering of symptoms may occur due to fact that the symptoms themselves are related. The decisions of this panel could subsequently be used as the validation standard. Adjusting for various covariates is already possible. as yet. Our study. which hampers the validation process. more elaborate studies.1). This does not necessarily imply. This is the case with nausea and vomiting or abdominal pain and diarrhoea. However. which symptoms.

General discussion

In the majority of reports sent to the Netherlands Pharmacovigilance Foundation information on the concomitant drug use is included. For the analysis of associations between ADR(s) and drugs, and for the analysis of drug-related syndromes, however, only the suspected medication was used in the studies presented in this thesis because the reporting health professionals did not attribute the suspected ADR(s) to the concomitant medication. If we would have included the concomitant medication in our investigations, which might then have yielded different results. This approach was for instance used by Moore in his analysis of the possible association between hypoglycaemia and the use of ACE inhibitors.30 The differences of these two approaches should be studied in more detail. The extent to which clinical information can be taken into account is limited in the current method of quantitative signal detection. At present, clinical information first has to be coded, for instance by using the WHO ADR terminology, before statistical analyses can be carried out. In the classical, empirical assessment methods, however, clinical information has a more prominent place. In order to improve the signal detection process, integration of clinical details in the quantitative approach is recommended. In the approaches discussed in the present thesis, the ATC code was used to code the suspected drug. With the ATC classification chemically related drugs can also be clustered to a certain extent. This may reveal associations between ADRs and groups of chemically related drugs. However, in addition to the chemical relationships, drugs may also be pharmacologically related even though they do not have a similar ATC code. For instance, despite the fact that selective serotonin reuptake inhibitors and triptans (serotonin agonists) belong to different ATC classes, they may still share the same ADRs because of their similar pharmacological activity. The use of different methods of clustering in quantitative signal detection may provide a deeper insight into the relationship between the chemical or pharmacological properties of the drugs involved and the occurrence of certain types of ADRs. Concerning the analysis of drug-drug interactions, aspects of metabolism of the drug under investigation can be studied, such as the metabolism by the various cytochrome subtypes. Finally, in Chapter 4.2 concepts of identifying risk factors in patients, like age, gender or concomitant diseases, was described. As with the analysis of drug-drug interactions, the relatively small size of the data set may hamper the analysis of

167

General discussion

these interactions and they should therefore preferably be carried out on larger data sets.

Final remarks
Although the first initiatives of spontaneous reporting date from the middle of the previous century, pharmacovigilance is continuously being refined. In addition to the classical approach, where for each case it is assessed if the reported association represents a new ADR, quantitative techniques are being developed to facilitate the signal detection process. These new techniques enable a more detailed analysis of patient characteristics, drugs and the ADRs involved, thus enhancing the quality and widening the scope of pharmacovigilance. Given the rapid advances in information technology, it is expected that quantitative signal detection will establish itself as a standard source of information in spontaneous reporting, which will clearly benefit the study into and the monitoring of the safety of drugs after marketing.

References
1. 2. Rossi AC, Hsu JP, Faich GA. Ulcerogenicity of piroxicam: An analysis of spontaneously reported data. Br Med J 1987;294:147-50. de Koning GHP. A regionalized spontaneous surveillance program for adverse drug reactions as a tool to improve pharmacotherapy. (Thesis) Utrecht University, Utrecht, the Netherlands, 1994. Egberts ACG Pharmacoepidemiologic approaches to the evaluation of antidepressant drugs. (Thesis) Utrecht University, Utrecht, the Netherlands, 1997. Meyboom RHB. Detecting adverse drug reactions - pharmacovigilance in the Netherlands. (Thesis) Catholic University of Nijmegen, Nijmegen, the Netherlands, 1998. Meyboom RHB. Pharmacovigilance in perspective. Drug Saf 1999;21:429-47. Rawlins MD, Thomson JW. Mechanisms of adverse drug reactions. In: Davis DM (ed). Textbook of adverse drug reactions. Oxford: Oxford University Press, 1991:18-45. Amery WK. Why there is a need for pharmacovigilance. Pharmacoepidemiol Drug Saf 1999;8:64. Talbot JC, Nilsson BS. Pharmacovigilance in the pharmaceutical industry. Br J Clin Pharmacol 1998;45:427-31. Stephens MDB. Introduction. In: Stephens MDB, Talbot JCC, Routledge PA (eds). Detection of new adverse drug reactions. New York: Grove's Dictionaries inc, 1998. Finney DJ. Systemic signalling of adverse reactions to drugs. Methods Inf Med 1974;13:110.

3. 4. 5. 6. 7. 8. 9. 10.

168

General discussion
11. 12. Evans S. Pharmacovigilance: a science or fielding emergencies? Stat Med 2000;19:3199209. Weber JCP. Epidemiology of Adverse Reactions to Nonsteroidal Anti-inflammatory Drugs In: Rainsford KD, Velo GP (eds). Advances in Inflammatory Research. New York: Raven Press, 1984:1-7. Griffin JP, Weber JC. Voluntary systems of adverse reaction reporting-Part II. Adverse Drug React Acute Poisoning Rev 1986;5:23-55. Bottiger M, Romanus V, de Verdier C, Boman G. Osteitis and other complications caused by generalized BCG-itis. Experiences in Sweden. Acta Paediatr Scand 1982;71:471-8. Moride Y, Haramburu F, Requejo AA, Begaud B. Under-reporting of adverse drug reactions in general practice. Br J Clin Pharmacol 1997;43:177-81. Stricker BHCh, Tijssen JGP. Serum sickness-like reactions to cefaclor. J Clin Epidemiol 1992;45:1177-84. Begaud B, Moride Y, Tubert-Bitter P, Chaslerie A, Haramburu F. False-positives in spontaneous reporting: should we worry about them? Br J Clin Pharmacol 1994;38:401-4. Venning GR. Validity of anecdotal reports of suspected adverse drug reactions: the problem of false alarms. Br Med J 1982;284:249-52. Po AL, Zhang WY. What lessons can be learnt from withdrawal of mibefradil from the market? Lancet 1998;351:1829-30. Routledge PA. Adverse drug reactions and interactions: Mechanisms, risk factors, detection, management and avoidance. In: Stephens MDB, Talbot JCC, Routledge PA (eds). Detection of New Adverse Drug Reactions. New York: Grove's Dictionaries inc, 1998. D'Arcy PF. Drug reactions and interactions in the elderly patient. Drug Intell Clin Pharm 1982;16:925-9. Veehof L, Stewart R, Haaijer-Ruskamp F, Jong BM. The development of polypharmacy. A longitudinal study. Fam Pract 2000;17:261-7. Pirmohamed M, Orme ML. Drug interactions of clinical importance In: Davies DM, Ferner RE, de Glanville H (eds). Davies's Textbook of Adverse Drug Reactions.London: Chapman & Hall Medical, 1998:888-912. Meyboom RH, Egberts AC, Edwards IR, Hekster YA, de Koning FH, Gribnau FW. Principles of signal detection in pharmacovigilance. Drug Saf 1997;16:355-65. Dirckx J (ed). Stedman's concise medical & allied health dictionary. Baltimore: Willams and Wilkins, 1997. Wright P. Untoward effects associated with practolol administration: oculomucocutaneous syndrome. Br Med J 1975;1:595-8. Halpin TJ, Holtzhauer FJ, Campbell RJ, Hall LJ, Correa-Villasenor A, Lanese R, Rice J, Hurwitz ES. Reye's syndrome and medication use. JAMA 1982;248:687-91. Ashton C, Young A. Drug-induced psychiatric disorders. In: Davies DM, Ferner RE , de Glanville H (eds). Davies's Textbook of Adverse Drug Reactions. London: Chapman & Hall Medical, 1998:669-731. Lindquist M, Ståhl M, Bate A, Edwards IR, Meyboom RH. A retrospective evaluation of a data mining approach to aid finding new adverse drug reaction signals in the WHO international database. Drug Saf 2000;23:533-42.

13. 14. 15. 16. 17. 18. 19. 20.

21. 22. 23.

24. 25. 26. 27. 28.

29.

169

Reports of hypoglycaemia associated with the use of ACE inhibitors and other drugs: a case/non-case study in the French pharmacovigilance system database. Br J Clin Pharmacol 1997.General discussion 30. Andrejak M. Moore N. Haramburu F.44:513-8. 170 . Ollagnier M. Noblet C. Kreft-Jais C. Begaud B.

and severe congenital malformations of the limbs of children whose mothers had used this drug in the early stages of pregnancy. Statistical analyses provide an answer to the question if the 171 . or quantitative signal detection. drug-related syndromes. Next to this classical case-by-case analysis. also the use of statistical approaches or quantitative signal detection may be an efficient tool to analyse reports sent to an SRS. they are generally difficult to detect or predict at this stage. At the time of marketing of the drug.Summary The development of a new drug is a time-consuming process. efficacy as well as safety. Health professionals are urged (in most countries) or obliged by law to report their suspicions of possible associations between drugs and ADRs to the centres managing spontaneous reporting systems (SRSs). the Pharmacovigilance Foundation Lareb is responsible for the management and maintenance of the SRS concerning drugs that have been approved for the Dutch market. and on behalf of the Medicines Evaluation Board. The application of numerical procedures in pharmacovigilance. and others only occur after prolonged use of the drug or are restricted to specific patient groups. In general. the individual reports are all being analysed by experienced assessors. the size as well as the duration of the trials is limited. but also for more complex relations like drug-drug interactions. Therefore. In the Netherlands. with the detection of the association between thalidomide. also after marketing. Furthermore. One of the aims of the SRSs is the detection of signals indicating a possible association between a suspected drug and reported ADR. a hypnotic drug. risk factors in patients and time trends after marketing of the drug under investigation. these studies are restricted to a selected group of patients. in the course of which attention is paid to quality. As some adverse drug reactions (ADRs) are rare or have a long latency period. can not only be used for signal detection and analysis of associations between a single ADR and a single drug. the information concerning the safety is generally limited and usually originating from a rather limited number of patients and studies. One of the ways of post-marketing surveillance is ‘spontaneous reporting’. The 1960s brought an awareness of the importance of a systematic surveillance of drugs for ADRs.

worded differently. the Poisson probability and Chi-square test are compared with the use of the Bayesian Confidence Propagation Neural Network analysis (BCPNN) of the WHO Collaborating Centre for International Drug Monitoring. examples of the application of quantitative signal detection are given. socalled non-selective underreporting does not influence the height of the ROR. different measures are used to quantify the extent to which an ADR is reported disproportionally to a certain drug compared to the generality of the database.2 shows that also in the analysis of complex relationships like drugdrug interactions and drug-related syndromes. The study in Chapter 1. the use of the Reporting Odds Ratio. In the second part of this thesis. Proportional Reporting Ratio.1 is to examine the level of concordance of the various measures when applied to the dataset of the Netherlands Pharmacovigilance Foundation. non-selective underreporting does not influence the height of the ROR. if the association has been reported disproportionally. The ROR can also be used in a logistic model in the analysis of clustering between drugs for studying possible drug-drug interactions as well as in the analysis of clustering of ADRs while studying possible drug-related syndromes. In general. The first part of this thesis consists of two studies focussing on methodological aspects of quantitative signal detection. The objective of this thesis is to explore the value and possibilities of quantitative signal detection in spontaneous reporting systems and to add further conceptual insight both into the analysis of the relationship between adverse drug reactions and a suspected drug as well as into complex relationships like drug-drug interactions and drug-related syndromes. this study shows that the different measures used are broadly comparable when four or more cases per combination have been collected. In various centres. Therefore. sensitivity is high and the specificity is fairly low with respect to the reference measure. The objective of the study in Chapter 1. These examples concern the analysis of the association of 172 .Summary number of observed cases exceeds the number of expected cases or. In the studies presented in this thesis the ROR is frequently used as a measure of disproportionality. Additionally. Yule’s Q. One of the advantages of the use of this measure is the fact that in the analysis of an association between a suspected drug and reported ADR.

In a logistic model. Since the introduction of the antimycotic drug terbinafine. In the dataset of the Netherlands Pharmacovigilance Foundation in reports indicating the use of both types of drugs. The objective of this study is to investigate whether the risk of anaphylactic reactions being reported during the use of various NSAIDs is greater than with other classes of drugs and if differences among NSAIDs exist.4. ibuprofen and naproxen. Although this method has been developed to detect unknown drug-drug interactions.3 the known interaction between diuretics and NSAIDs is used to illustrate this approach. The results of this study strengthen previous findings concerning the relative high risk of developing an anaphylactic reaction during the use of NSAIDs. in the study in Chapter 2.2 shows that datasets of SRS can be use to signal possible drug-drug interactions. the combined use of both drugs is examined. Like the analysis of concomitant use of drugs in the analysis of drug-drug interactions. This may subsequently cause an increase or decrease in the number of reported ADRs of the latter drug. one drug influences the effect of another drug.1 an example is given of the analysis of a possible association between a drug and possible ADR. In Chapter 2. In the event of a drug-drug interaction. which points at the existence of an actual drug-drug interaction. The results strongly suggest that in women using oral contraceptives and itraconazole concomitantly. the number of observed cases of a decreased efficacy of diuretics exceed the number of expected cases. datasets of SRS can be used to study the possibility of clustering of symptoms indicating the existence of a possible drug-related syndrome. As an example of this approach we analyse the ADR ‘delayed withdrawal bleeding’ during the concomitant use of oral contraceptives and the antimycotic drug itraconazole. the withdrawal bleeding may be delayed due to a drug-drug interaction. It is generally known that NonSteroidal Anti-inflammatory Drugs (NSAIDs) may rarely cause an anaphylactic reaction. The study in Chapter 2. as is illustrated in Chapter 2.Summary drug and an ADR as well as the analysis of drug-drug interactions and clustering of ADRs. which are not explicitly reported. the Netherlands Pharmacovigilance Foundation received eight reports of arthralgia during the use of this drug. In four reports the additional presence of 173 . The method described may enable a more active approach in the detection of drug-drug interactions after marketing. particularly diclofenac. The studies in both chapters illustrate that spontaneous reporting systems have a potential for signal detection and the analysis of possible drug-drug interactions.

Secondly. for example. Additionally. a ‘critical term’ of the WHO ADR terminology. is generated periodically. Goal of the study of Chapter 3. such as the fact whether the ADR or drug is new.2 the role and position of quantitative signal detection and the way it is applied at Lareb is described.Summary skin reactions is mentioned. Two patients who report arthralgia also have a fever. possibly an immunological reaction. These reports are described in more detail. quantitative signal detection is used to study signals in more detail. Subsequently. the ADR being unlabelled and the presence of a disproportionate association are all independently associated with the selection of a signal. on an ad hoc basis. the results of the statistical analyses are used as an aid in the traditional case-by-case review. and analysed in a logistic model in order to determine whether symptoms are interrelated. Logistic regression analysis is used to investigate the influence of various factors. First. two of the aforementioned factors have a sensitivity of 0. the seriousness of the reaction and factors related to the documentation of the reports. Finally.82 for the association to be selected for dissemination to the MEB. like drug-drug interactions and drug-related syndromes. two of these reports concern urticaria. The analysis shows that the presence of a ‘serious report’. the selection of signals is based on both qualitative and quantitative aspects. The statistical analysis shows a clustering of these symptoms among reports of patients using terbinafine. to the MEB.1 is to gain insight into factors contributing to the selection and dissemination of possible signals from the Netherlands Pharmacovigilance Foundation Lareb. Obviously.72 and a specificity of 0. which were not disseminated. the strength of the association. 174 . for example by adjusting for covariates or analysing more complex relationships. In the third part of this thesis the implementation of quantitative signal detection at the Netherlands Pharmacovigilance Foundation is discussed. these disproportionate associations needs to be analyse in more detail. signals disseminated to the MEB were compared with controls consisting of associations between drugs and ADRs. Among the many potential signals that are identified. In a case-control design. only a limited number of signals are important enough to inform the Dutch Medicines Evaluation Board (MEB) in a special procedure. a list of associations of ADRs and suspected drugs that are disproportionately present in the database. In Chapter 3. Presumably these symptoms have a shared aetiology. for datamining purposes. the findings show that the presence of.

The studies presented in this thesis show that the quantitative signal detection is an essential part of the analysis procedures applied by the Netherlands Pharmacovigilance Foundation Lareb. it is expected that quantitative signal detection will establish itself as a standard source of information in spontaneous reporting. and they cannot provide accurate information about the incidence of the ADR. every signal that has been generated by the quantitative approach should be analysed in more detail. It does not only facilitate the detection and analysis of unknown associations between a single drug and a single event. Additionally. Finally some suggestions are made for further research.Summary In Chapter 4. SRSs are primarily designed to generate hypotheses of possible ADRs. but also offers possibilities for the analysis of more complex relations such as drugdrug interactions and drug-related syndromes. The studies presented in this thesis show that the quantitative approach can be regarded as an additional source of information and not as a substitute for the classical case-by-case analysis. some additional methodological considerations and practical implications of quantitative signal detection are elaborated. 175 . the extent to which clinical information can be taken into account is limited in the current method of quantitative signal detection. For this reason. the General discussion. Given the rapid advances in information technology.

.

Samenvatting De ontwikkeling van een geneesmiddel is een langdurig proces waarin aandacht besteed wordt aan kwaliteit. is in dit stadium nog beperkt. De toepassing van statistische methoden in de geneesmiddelen177 . In Nederland wordt deze taak voor wat betreft geregistreerde geneesmiddelen uitgevoerd door de stichting Landelijke Registratie Evaluatie Bijwerkingen (Lareb) in opdracht van het College ter Beoordeling van Geneesmiddelen. Naast deze methode kan ook het gebruik van statistische technieken een efficiënt middel zijn om de binnengekomen meldingen te analyseren in relatie tot eerder gedane meldingen. Artsen en apothekers kunnen vermoedens van bijwerkingen op vrijwillige basis melden aan de hiervoor aangewezen instanties. is het niet mogelijk een volledig overzicht te hebben van alle bijwerkingen van het desbetreffende geneesmiddel op het moment dat het op de markt toegelaten wordt. werkzaamheid en veiligheid. Omdat deze onderzoeken bovendien in de regel betrekking hebben op een geselecteerde groep patiënten en ze beperkt zijn in zowel omvang als tijd. Met name de kennis over zeldzamere bijwerkingen. Een van de belangrijkste doelstellingen van spontane rapportage is het zo snel mogelijk genereren van signalen die kunnen wijzen op een mogelijke samenhang tussen een geneesmiddelgebruik en het optreden van de gemelde bijwerking. liet de noodzaak zien van een systematische bewaking van bijwerkingen ook nadat deze op de markt toegelaten zijn. Iedere melding van een mogelijke bijwerkingen wordt hiertoe na ontvangst afzonderlijk geanalyseerd door ervaren beoordelaars. is de informatie omtrent de veiligheid van geneesmiddelen in de regel beperkt en veelal afkomstig uit een naar verhouding gering aantal onderzoeken en patiënten. Op het moment waarop een geneesmiddel op de markt toegelaten wordt. Een van de vormen van geneesmiddelenbewaking is de zogenaamde ‘spontane rapportage’. bijwerkingen die pas na langere tijd optreden en bijwerkingen bij patiënten die meerdere geneesmiddelen tegelijk gebruiken of lijden aan chronische aandoeningen. De dramatische ervaringen met thalidomide (Softenon) in de jaren zestig. een slaapmiddel dat ernstige aangeboren afwijkingen kan veroorzaken bij kinderen wiens moeder het middel in de eerste maanden van de zwangerschap gebruikt.

Yules’ Q. Bij de onderzoeken in dit proefschrift wordt veelal gebruik gemaakt van de ROR als maat voor het bestaan van disproportionaliteit. Bij deze statistische analyses wordt gekeken of meldingen vaker voorkomen dan op grond van het toeval verwacht kan worden. kan niet alleen gebruikt worden voor de detectie en analyse van associaties tussen bijwerkingen en geneesmiddelen. de Proportional Reporting Ratio. ook wel kwantitatieve signaaldetectie genoemd. Het doel van de in dit proefschrift beschreven onderzoeken is de waarde en mogelijkheden van kwantitatieve signaaldetectie te exploreren en daarnaast meer inzicht te krijgen in de analyse van complexe relaties zoals bijvoorbeeld mogelijke geneesmiddelinteracties en de samenhang tussen verschillende gemelde bijwerkingen als uiting van een geneesmiddel-gerelateerd syndroom. Hiervoor worden de toepassing van de Reporting Odds Ratio (ROR). maar bovendien kunnen ook complexere relaties onderzocht worden zoals geneesmiddelinteracties. Een van de voordelen van het gebruik van de ROR is dat bij de analyse van de relatie tussen geneesmiddel en gerapporteerde bijwerking zogenaamde niet-selectieve onderrapportage geen invloed heeft op de hoogte van deze ROR. hierbij wordt geprobeerd een antwoord te geven op de vraag of de associatie disproportioneel vaak gemeld wordt. indien ze toegepast worden op de dataset van de Stichting Lareb. Bij vergelijking van de verschillende methoden met de BCPNN blijkt dat de sensitiviteit hoog en de specificiteit naar verhouding laag is. Bovendien laat dit onderzoek zien dat de resultaten van de verschillende methoden grote overeenkomsten met elkaar vertonen indien vier of meer meldingen per combinatie van geneesmiddel en bijwerking ontvangen zijn. Anders gezegd. In de eerste deel van dit proefschrift wordt een tweetal onderzoeken beschreven die ingaan op enkele methodologische aspecten van kwantitatieve signaaldetectie in geneesmiddelenbewaking. In het in Hoofdstuk 1. Deze maat kan in een logistisch model 178 . geneesmiddel gerelateerde syndromen en risicofactoren voor het ontwikkelen van bijwerkingen.Samenvatting bewaking. Verschillende centra voor geneesmiddelbewaking hebben ieder een eigen methode van statistische analyse.1 beschreven onderzoek wordt nagegaan wat de mate is van overeenstemming van de verschillende methoden. de Chi-kwadraat toets en de Poisson probability vergeleken met de toepassing van de Bayesian Propagation Neural Network analysis (BCPNN) die door het WHO Collaborating Centre for International Drug Monitoring gebruikt wordt.

Dit zou kunnen wijzen op een daadwerkelijk verhoogd risico op het krijgen van een anafylactische reactie bij het gebruik van deze geneesmiddelen.2 wordt aangetoond dat ook bij deze meer complexe analyses niet-selectieve onderrapportage geen invloed heeft op de hoogte van de ROR. In Hoofdstuk 2. Het onderzoek uit Hoofdstuk 2.2 laat zien dat datasets van spontane meldingsystemen gebruikt kunnen worden om mogelijke geneesmiddelinteracties. die niet expliciet zijn gemeld. te detecteren. In het tweede deel van dit proefschrift wordt een aantal praktische voorbeelden gegeven van de toepassing van kwantitatieve signaaldetectie. Men kan verwachten dat deze vervolgens ook meer gemeld worden.1 wordt een voorbeeld gegeven van de analyse van een mogelijke relatie tussen een geneesmiddel en vermeende bijwerking. De resultaten laten zien dat met name voor diclofenac. Het gelijktijdig gebruik van deze twee middelen blijkt inderdaad geassocieerd te zijn met een verhoogd aantal meldingen van een vertraagde onttrekkingsbloeding. Een 179 . Uitgangspunt hierbij is dat in het geval van een mogelijke geneesmiddel-interactie een of meer interacterende geneesmiddelen meer bijwerkingen kunnen veroorzaken. Met behulp van een logistisch model is de invloed van het gecombineerd gebruik van beide geneesmiddelen onderzocht. In Hoofdstuk 1. Deze voorbeelden hebben zowel betrekking op de analyse van een verband tussen geneesmiddel en mogelijke bijwerking. In dit onderzoek wordt nagegaan of de kans op het melden van een anafylactische reactie tijdens het gebruik van NSAIDs hoger is dan bij het gebruik van andere geneesmiddelen en of er onderlinge verschillen tussen de NSAIDs bestaan. als de analyse van meer complexere relaties zoals mogelijke geneesmiddel-interacties en clustering van mogelijke bijwerkingen.Samenvatting ook gebruikt worden voor de analyse van clustering tussen geneesmiddelen bij onderzoek naar mogelijke geneesmiddelinteracties en tussen gerapporteerde bijwerkingen bij de analyse van mogelijke syndromen. ibuprofen en naproxen een naar verhouding groot aantal en disproportioneel aantal meldingen ontvangen zijn. hetgeen een aanwijzing kan zijn voor het daadwerkelijk bestaan van geneesmiddel-interactie tussen deze stoffen. Het is bekend dat Nietsteroïde anti-inflammatoire geneesmiddelen (NSAIDs) in zeldzame gevallen een anafylactische reactie kunnen veroorzaken. Als illustratie van deze methode is de bijwerking ‘vertraagde onttrekkingsbloeding’ tijdens het gelijktijdig gebruik van orale anticonceptiva en het antimycoticum itraconazol geanalyseerd.

Niet alleen geneesmiddelen kunnen in samenhang met elkaar gemeld worden.4 besproken word. Twee patiënten hadden eveneens koorts. zijn slechts enkele belangrijk genoeg om in een speciale procedure doorgegeven te worden aan het College ter Beoordeling van Geneesmiddelen (CBG). waaronder het feit dat een geneesmiddel nieuw was. Mogelijk ligt een immunologische reactie aan deze bijwerkingen ten grondslag. Ook bij de meldingen in de Lareb databank waar zowel NSAIDs als diuretica gebruikt worden. Bij de Stichting Lareb zijn sinds de introductie van het antimycoticum terbinafine acht meldingen ontvangen van het optreden van arthralgieën. De onderzoeken uit beide hoofdstukken laten zien dat spontane rapportagesystemen mogelijkheden bieden voor zowel de detectie als analyse van geneesmiddelinteractie. Verschillende factoren werden onderzocht. Doel van het in Hoofdstuk 3.3 en laat zien dat ook een reeds bekende geneesmiddel-interactie teruggevonden kan worden aan de hand van de meldingen in de database. De statistische analyse versterkt de indruk dat de genoemde symptomen inderdaad verband met elkaar lijken te houden.1 beschreven onderzoek is inzicht te krijgen in de factoren die een rol spelen bij de selectie en verspreiding van signalen die voortkomen uit het meldingssysteem van Lareb. Het gelijktijdig gebruik van NSAIDs en diuretica kan leiden tot een verminderd effectiviteit van laatstgenoemde middelen. hetgeen zou kunnen wijzen op het bestaan van een syndroom. de sterke van de associatie en aspecten die betrekking hebben op de documentatie van de 180 . In een case-control studie werden signalen die aan het CBG doorgegeven zijn. Bij vier meldingen werden eveneens huidreacties gemeld. vergeleken met een controlegroep die bestond uit associaties die niet in een speciale procedure aan het CBG doorgegeven werden. Ook de bijwerkingen zelf kunnen vaak in combinatie met elkaar gemeld worden. Een voorbeeld hiervan is nader uitgewerkt in het onderzoek dat in Hoofdstuk 2. Van de vele signalen die geïdentificeerd worden. In het derde deel wordt de plaats van de kwantitatieve benadering bij Lareb nader beschreven. In het onderzoek worden de klinische details verder uitgewerkt en wordt aan de hand van een logistisch model gekeken of de genoemde symptomen met elkaar gerelateerd zijn. komen meldingen betreffende een mogelijke verminderde effectiviteit van diuretica vaker voor dan op grond van het toeval verwacht mag worden. of de bijwerking bekend was.Samenvatting tweede voorbeeld wordt gegeven in Hoofdstuk 2.

In de Discussie worden enkele aanvullende methodologische overwegingen gegeven. Een associatie die vaker voorkomt dan op grond van het toeval verwacht wordt. Niet alleen worden hierdoor de detectie en analyse van onbekende associaties tussen een geneesmiddel en mogelijke bijwerking verbeterd. Voor dit doel worden lijsten van geneesmiddel en mogelijke bijwerking gegenereerd. Bovendien kan klinisch inhoudelijke informatie op dit moment nog niet in het analyseproces betrokken worden. Hierop staat mate van disproportionaliteit van verschillende associaties vermeld.80 heeft voor het feit dat het signaal al dan niet geselecteerd wordt. De onderzoeken in dit proefschrift laten zien dat kwantitatieve signaaldetectie een essentieel onderdeel vormt van het analyseproces bij Lareb.Samenvatting meldingen. Daarnaast bleek dat bijvoorbeeld de aanwezigheid van een combinatie van twee van de bovengenoemde factoren een sensitiviteit van 0. De kwantitatieve benadering moet in de eerste plaats gezien worden als een 181 . Spontane rapportage systemen zijn primair ontworpen om hypothesen te genereren over het bestaan van mogelijke bijwerkingen.2 wordt een overzicht gegeven van de wijze waarop kwantitatieve signaal detectie bij Lareb toegepast worden. Tot slot wordt de kwantitatieve benadering bij Lareb gebruikt om meer complexe relaties te analyseren zoals geneesmiddelinteracties en geneesmiddel-gerelateerde syndromen. het feit dat een bijwerking vaker gemeld is dan op grond van het toeval verwacht kon worden.72 en een specificiteit van 0. alsmede het feit of een bijwerking bekend was. kan wijzen op een mogelijk signaal en dient daarom nader geanalyseerd te worden. maar het biedt mogelijkheden ook meer om complexe relaties zoals geneesmiddel-interacties en geneesmiddel-gerelateerde syndromen te onderzoeken. Blijkbaar spelen zowel kwantitatieve als kwalitatieve factoren een rol bij de selectie van signalen. Bij de afzonderlijke evaluatie van de verschillende binnengekomen meldingen vervult de kwantitatieve signaaldetectie een ondersteunende rol. Ze kunnen echter geen precieze informatie geven over de incidentie van een bijwerking. allen onafhankelijk geassocieerd waren met de selectie van een signaal. In Hoofdstuk 3. het feit dat een bijwerking gecodeerd was met een zogenaamde ‘critical term’ uit de bijwerkingenterminologie van de WHO. De multivariate analyse toont aan dat de aanwezigheid van een ‘ernstige’ melding. Daarnaast is deze methode ook een middel om actief te zoeken naar associaties in de databank die nadere aandacht verdienen.

Gezien de snelle ontwikkelingen in de informatietechnologie mag verwacht worden dat kwantitatieve signaaldetectie een standaardbron van informatie in de geneesmiddelenbewaking zal worden. Tot slot worden voorstellen gedaan voor verder onderzoek.Samenvatting additionele benadering en dient ieder gegenereerd signaal nader inhoudelijk geanalyseerd te worden. 182 .

We now write a general equation for the overall underreporting of the ADR experiencing population by the term kxya. with levels a (a =1. or possibly two covariates. and the other two variables are X with levels x (x = 1. the notation X c x for the underreporting of levels x = 1 and x = 2 of variable X.2) and Y with levels y (y =1.2).We denote the variables as follows: for the specific ADR we use A. for example.suspected population by txya. or one covariate and one drug in section 5. for example. set out for the situation of two variables and an ADR. We begin by defining a general loglinear model for the true frequencies t m xya in the ADR experiencing population as X A XY XA XYA log t xya = u + u x + uY + u a + u xy + u xa + uYA + u xya y ya (14) where the parameters add up to zero over each index. This leads to X Y A XY XA YA XYA k xya = cc x c y c a c xy c xa c ya c xya . As usual. (16) 183 . u = b . We use.2).Addendum to Chapter 1. These two variables can be either two specific drugs as in section 4. and the expected frequencies in the SRS population by mxya. X (15) X where. By taking exponents this equation can be reparametrized as X Y A XY XA YA XYA t xya = bb x b y b a b xy b xa b ya b xya .2 Integrating underreporting into loglinear models In this appendix we present our results in a more general way. The identifying restrictions in (15) are X Y A XY XY XA XA Õ x b x = Õ y b y = Õ a b a = Õ x b xy =Õ y b xy = Õ x b xa = Õ a b xa = YA YA XYA XYA XYA Õ y b ya = Õ a b a = Õ x b xya = Õ y b xya = Õ a b xya = 1. we denote the true frequencies in the ADR. exp u x = b x and so on.

e.t. In passing we indicate how these situations relate to the special cases that we discussed in section 4. every g-parameter is biased w. if a higherorder term is included. in the expected frequencies in the SRS the higher-order γ-parameters are equal 184 . b xya with a factor c xya . First we distinguish three situations.2 with restrictions similar to those for the b-parameters: X A XY XY XA XA Õ x c x = Õ y cY = Õ a c a = Õ x c xy =Õ y c xy = Õ x c xa = Õ a c xa = y XYA XYA XYA Õ y cYA = Õ a cYA = Õ x c xya = Õ y c xya = Õ a c xya = 1. In the first situation. its corresponding β-parameter by its corresponding underreporting constant c.e. loglinear) model for the expected values in the SRS population.t. XYA For example. Equation (17) also makes clear what happens when certain underreporting constants or certain loglinear β-parameters vanish by being equal to one. In our discussion we only deal with so-called hierarchical models. all the lower-order terms with indices that are a subset of the set of indices of the higher-order term are included. Note that the g-parameters specify a multiplicative (i. X Y A XY XA YA XYA (17) where the g-parameters are restricted by X Y A XY XY XA XA Õ x g x = Õ y g y = Õ a g a = Õ x g xy = Õ y g xy = Õ x g xa = Õa g xa = YA YA XYA XYA XYA Õ y g ya = Õa g a = Õ x g xya = Õ y g xya = Õa g xya = 1. The result is that. But equation (17) illustrates the fact that. a The expected frequencies mxya for the SRS are related to the true frequencies txya in the ADR experiencing population by mxya = k xya t xya X A XY XA XYA = (cb ) c x b x cY b y ca b a c xy b xy c xa b xa cYA b ya c xya b xya y ya X Y A XY XA YA ( )( )( )( )( )( )( XYA ) = gg x g y g a g xy g xa g ya g xya . i. Analysis of the observed frequencies of the SRS will yield estimates of the g-parameters. and 6 c a ´ c -1 = 1. some of the higher-order underreporting constants vanish where the corresponding β-parameters do not. ya a In sections 3. 4 and 5.Addendum to Chapter 1. these restrictions are satisfied for the underreporting constants because.1 we XYA XYA showed how this biases the odds-ratios in equations (6) to (11).r. in principle. In section 4. in Tables 5. for example.r. g xya is biased w.

2. the estimates of γ -parameters are in fact estimates of the corresponding underreporting constants. We refer to section 4. some of the βparameters are equal to one. and this XYA leads to the vanishing of g xya (they vanish by analysis). For example.r. This is discussed in detail in section 4. this was discussed in detail for the vanishing of b xya and XYA c xya (there denoted as udea and cdea) (they vanish by assumption). but XYA b xya ¹ 1 (there denoted by udea. SRS data are not useful for the assessment of causal links between drugs and ADRs.Addendum to Chapter 1. if in the ADR experiencing population in the true frequencies b xa = b xya = 1 . and the possible interpretation of odds-ratios estimated from SRS data.2).2 for a further discussion of this point. in estimating a loglinear model on the observed frequencies from the SRS. Thus the corresponding γ -parameters vanish. In this situation.2 to the β-parameters in the ADR suspected population.t. if a loglinear analyses is conducted on the observed frequencies in the SRS. the higher-order γ -parameters are unbiased w. but none of the underreporting constants vanish. In the third situation. In the second situation the corresponding β-parameters and underreporting constants vanish. This situation is worked out in observations 1 to 3 of section 4. This situation is not discussed in section 4. to the corresponding β parameters of interest. where XYA interest centred on the situation where c xya = 1 (there denoted by cdea). If the XA XYA XYA higher-order underreporting constants dominate the corresponding higherorder β-parameters. In section 3. observation 4.2. The result is that. then XA XYA XYA g xa = g xya = 1 and c xya = 1. 185 .

.

who developed pains of hands and feet 1-2 days after the start of terbinafine. C-reactive protein levels. GGT. rheumatological of allergic disorders. hips and elbows. no additional information was received. She fully recovered. Hb.4x109/l. 2 weeks after the start of terbinafine. liver of kidney function disorders. Unfortunately. one week this extended to her right knee. An oral contraceptive ethinyl estradiol/desogestrel was used concomitantly. He fully recovered. beclomethasone nasal inhaler and ibuprofen 400 mg three times daily when needed. lymfadenopathy or skin disorders.Addendum to Chapter 2. Laboratory examination revealed no abnormalities. The use of terbinafine was stopped and he was treated with nabumetone 1 g twice daily. The latter drug was used because of chronic backache. Concomitant medication used was doxycycline 100 mg (indication for use unknown) 7-14 days after the start of terbinafine. Concomitant medication used was miconazole cream. streptococcal antibody titre and rheumatoid factor were within normal limits. Terbinafine had not been used previously. 5 mm. There were no 187 . The complaints hampered her daily activities.9 mmol/l. There were no additional signs of arthritis. traumata. There was no fever. Patient D is a 41-year-old male who developed pain on flexion an extension of wrists and knees. Patient C is a 42-year-old female. The use of terbinafine was stopped and she was treated with ibuprofen 400 mg three times daily. 11 mmol/l. There was no history of joint disorders.4 Reports of arthralgia during the use of terfinafine Patient A is a 44-year-old woman who developed arthralgia of the right wrist 18 days after the start of terbinafine. All joints were symmetrical affected. She used no concomitant medication. 10 days after the start of terbinafine. ESR. 8. There were no joint disorders in the family history. No rechallenge was carried out. leukocyte count 5. Laboratory examination revealed: ESR. Patient B is a 41-year-old male who developed arthralgia of shoulders.

869 U/l. Patients A. leukocyte count was 7. GGT. He was treated with diclofenac and domperidone and fully recovered after 9 weeks. and blood smear). ESR.Addendum to Chapter 2. since the patient had moved and his former GP did not have his present medical history. kidney function normal. All patients were treated with terbinafine 250 mg once daily. vomiting. Concomitant medication used was budesonide inhaler. he fully recovered. One week after the start of terbinafine. Laboratory examination revealed no abnormalities (Hb. He fully recovered immediately after stopping the use of terbinafine. There were no signs of arthritis. He was treated with terfenadine. some atypical lymphocytes were found. 9. fever.8 x109/l. She was treated with ibuprofen 400 mg. Also a rash. The indication for use of patient G was not known. Unfortunately. He had a fever of 39oC. 10 days after the start of terbinafine. ESR. ALT and GGT had not increased. Hb. Patient F is 66-year-old male who developed urticarial rash. ALT. leukocytes. LDH. Patients A-E and H were treated for onychomycosis. On laboratory examination. who developed signs of ‘generalized’ arthralgia (no joints specified). After stopping terbinafine. D-F and H had no history of joint disorders. he developed anorexia. B. additional information could not be retrieved.4 additional signs of arthritis. and abdominal pain and complained of smell and taste disorders. stomatitis and peeling of the skin developed. who developed signs of arthritis (no joints specified) and generalized urticaria 10 days after he started using terbinafine. After one week re-examination of blood revealed no abnormalities. Laboratory tests revealed the following abnormalities: leukocyte count 4. On physical examination no abnormalities were found. headache. Patient G is a 34-year-old male. Concomitant medication used was miconazole cream. He was treated with terfenadine and prednisolone. 202 U/l. She fully recovered after 1 month. 5 mm. He used a homeopathic drug as concomitant medication. Patient E is 37-year-old female. vomiting and diffuse arthralgia (no joints specified) 2 weeks after starting terbinafine.2x109/l. No additional laboratory testing was done. 1672 U/l. Patient H is a 50-year-old male. The arthralgia was symmetrically distributed. In the blood smear.3 mmol/l. prednisone and promethazine. 188 . who developed pains in almost all joints and myalgia a 1-2 days after the start of terbinafine.

Neither were there any joint disorders in the family history in any of the patients. No rechallenge was done in any of the patients. 189 .4 traumata. rheumatological or allergic disorders. These data could not be retrieved for patients C and G. liver or kidney function disorders.Addendum to Chapter 2.

.

Peter van der Heijden en Stef van Buuren van de afdeling Preventie en Gezondheid van TNO waren in staat vanuit een andere invalshoek hun licht te laten schijnen op de kwantitatieve signaaldetectie. in te ruilen voor een breder perspectief. Ik wil daarom beginnen mijn dank uit te spreken aan mijn beide promotoren Bert Leufkens en Toine Egberts. AIO’s en andere medewerkers van de vakgroep farmaco-epidemiologie en farmacotherapie in Utrecht hebben laten zien dat een goede ontspannen werksfeer ervoor zorgt dat je met veel plezier kunt werken en optimaal kunt presteren. Hij hielp ervoor te zorgen dat de onderzoeken in het proefschrift nauw aansluiten bij de doelstellingen van de Stichting Lareb. Zijn politieke visie heeft ervoor gezorgd dat het werk van Lareb in een nog bredere kring erkenning heeft gekregen en zodoende een solide basis vormt van waaruit ideeën verder ontwikkeld kunnen worden. Jacques van der Hofstede. Zijn betrokkenheid. Zijn stimulerende en enthousiaste houding combineert hij met een grote vakinhoudelijke kennis. Ron Meyboom heeft met zijn grote kennis van de genees191 .Dankwoord Het is een goede traditie om op de laatste pagina’s van een proefschrift diegenen te bedanken die direct of indirect bijgedragen hebben aan de totstandkoming ervan. Kees van Grootheest heeft als directeur van de Stichting Lareb de mogelijkheden gecreëerd om binnen het kader van mijn werkzaamheden dit onderzoek uit te kunnen voeren. Bovendien bleek hij steeds in staat de soms wat engere visie die ik had wanneer ik me al te zeer met klein detail bezig hield. didactische vaardigheden en een uitstekende vakinhoudelijke kennis zijn een ideale combinatie van eigenschappen die ertoe bijgedragen hebben dat dit proefschrift ook inderdaad tot stand kwam. Hij mag daarom met recht de drijvende kracht achter dit project genoemd worden. Daarbij weet hij ook altijd openingen te creëren voor oplossingen van mogelijke problemen. Toine heb ik reeds leren kennen in 1994 toen wij beiden op dezelfde dag bij de Stichting Lareb kwamen werken. Collega’s en oud-collega’s bij de Stichting Lareb. Bert verstaat de kunst om gericht en kritisch naar je werk te kijken. Verschillende co-auteurs hebben een belangrijke inhoudelijke bijdrage geleverd aan dit proefschrift.

Zonder onze melders geen meldingen aan de Stichting Lareb en dus ook geen proefschrift. AW Broekmans. De levendige discussies met Andrew Bate van het Uppsala Monitoring Centre hebben laten zien dat onderlinge samenwerking tussen verschillende centra die zich bezig houden met spontane rapportage van grote toegevoegde waarde is. BHC Stricker waren in staat het proefschrift binnen korte termijn te beoordelen en bovendien waardevolle aanwijzingen te geven om het geheel verder te verbeteren. Dank ook aan al diegenen die in de laatste fase van het schrijven van het proefschrift een helpende hand geboden hebben met het corrigeren van foutjes. YA Hekster. het opmaken van de tekst en het verzenden van de proefschriften. Uit eigen ervaring weet ik dat het niet altijd vanzelfsprekend is een meldformulier in te vullen en te verzenden. Prof. Mijn grootste dank gaat uit naar Emmy en onze dochtertjes Linde en Sterre. dr. 192 . Mijn werk als huisarts is voor mij van grote waarde geweest de realiteit van alle dag niet uit het oog te verliezen.Dankwoord middelenbewaking en beschouwende houding een essentiële bijdrage geleverd aan de kwaliteit van het proefschrift. Prof. RHB Meyboom en Prof. dr. dr. IR Edwards. dr. De leescommissie waarin zitting hadden: Prof. Enthousiasme en geloof in het uiteindelijke doel van het melden van bijwerkingen zijn nodig om dit in de dagelijkse gang van zaken in te passen. De laatste maanden is er weinig tijd voor elkaar geweest. Hun steun en liefde hebben het mogelijk gemaakt dat de ruimte om dit proefschrift te schrijven er toch was. Mijn paranimfen Marcel Mutsaerts en Willem Diemont wil ik danken voor hun ondersteuning en hulp bij de organisatie van het feest. De prettige werksfeer waarin ik samen met mijn collega’s in Vught kan werken hebben me steeds laten zien wat het uiteindelijke doel is van onze exercities in de geneesmiddelenbewaking: een verbetering van de zorg voor de individuele patiënt. Artsen en apothekers die de moeite nemen om hun vermoedens van bijwerkingen te melden doen dit veelal in een relatieve anonimiteit. dr.

38:216-17. Egberts ACG. Du Buf-Vereijken PW. van Puijenbroek EP.110:300. J Intern Med 1996. van Doormaal JJ. Bouvy M. Huisarts en Wetenschap 1995. Egberts ACG. van der Laan JR. Verduijn MM. Meyboom RHB. Meyboom RHB. van Grootheest AC. NHG-standaard allergische en hyperreactieve rhinitis. Van Puijenbroek EP.52:1156-8. Headache 1996. N Z Med J 1997.131:551-4. Vinks MHAM. Possible increased risk of rhabdomyolysis during concomitant use of simvastatin and gemfibrozil. Pharmaceutisch weekbl 1995.List of publications Articles marked with an asterix (*) relate to work described in this thesis.130:477-9.16:459-62. In 't Veld BA. 193 . Steenvoorden MG. Bijwerkingen.141:1392-5. Trooster JF. van Puijenbroek EP.140:207-9. Overgevoeligheidsracties bij gebruik van mebeverine. Is het eerste-uitgifte signaal van nut voor postmarketing surveillance? Resultaten van een pilot study.240:403-4. Pharmaceutisch Weekbl 1996. Van Puijenbroek EP. van der Wal J. Stricker BHCh. Stemklachten als bijwerking van geneesmiddelen. Int J Clin Pharmacol Ther 1996. Lastdrager CJ. Toorenburg-Beijer B. Van Puijenbroek EP.36:48. Pharmaceutisch Weekbl 1996. Crohbach MJJS. Reversibele tandverkleuring tijdens oraal gebruik van antibiotica. Krom HJ.34:318. van Puijenbroek EP. Dekens-Konter JA. Huisarts en Wetenschap 1997. Ned Tijdschr Geneeskd 1997. Van Puijenbroek EP. van Puijenbroek EP. van Leeuwen JT. Ned Tijdschr Geneeskd 1996. Van Amerongen CA. Visual hallucinations and amnesia associated with the use of zolpidem. Disturbance of withdrawal bleeding during concomitant use of itraconazole and oral contraceptives. Geijer RMM. een verantwoordelijkheid van artsen? Medisch Contact 1997. Meyboom RHB. Van Puijenbroek EP.40:595-7. Van Puijenbroek EP. Van Puijenbroek EP. van Puijenbroek EP. Slaapstoornissen tijdens het gebruik van mefloquine. Reduction of migrainous headaches during the use of acenocoumarol. Smakman-Nossbaum E. Spooren PF. Zomerdijk J. Gerts BMF. Fixed drug eruption bij gebruik van norfloxacine.

Van der Linden PD.145:225-8. Nederl Tijdschr Geneeskd 1998. van Puijenbroek EP. *Van Puijenbroek EP. Verduijn MM. Sturkenboom MJ. Egberts ACG. Egberts ACG. Berghuis PH. Acute levercelnecrose bij gebruik van ketoconazol.List of Publications Van Puijenbroek EP. Leufkens HGM.142:590-2. Pulles-Heintzberger CFM. Eland IA.56:733-8. Meyboom RHB. Leufkens HG. Ned Tijdschr voor Epileptologie 2000. van Puijenbroek EP. Ned Tijdschr Geneeskd 1998.142(3):146-9.45:235-9. Different risks for NSAID-induced anaphylaxis.94:2417-22. Verstoring van de pilcyclus tijdens het gelijktijdig gebruik van itraconazol en orale anticonceptiva. Ned Tijdschr Geneeskd 2001. Artsen en apothekers kennen Lareb-is dat zo? Pharmaceutisch Weekbl 2000. Detecting drug-drug interactions using a database for spontaneous adverse drug reactions: an example with diuretics and non-steroidal anti-inflammatory drugs. van Puijenbroek EP.28:134-44.47:689-93. Int J Eat Disord 1998. Br J Clin Pharmacol 1999.24:111-3. Feenstra J. *Van Puijenbroek EP. van Puijenbroek EP. In 't Veld BA.142:2416-8. Veld BA. Leufkens HG. Hoekstra JB. Stricker BH. Van Puijenbroek EP. van Puijenbroek EP. Kuck EM. *Van Puijenbroek EP. Ned Tijdschr Geneeskd 1999. Ann Pharmacother: accepted for publication. Hekster YA. van Puijenbroek EP.143:2018-111. Meyboom RHB. Feenstra J. Metselaar HJ. Grootheest AC van.129:72. Pijlman AH. Sturkenboom MC. Semen-like urethral discharge during the use of mazindol. Arthritis Rheum 2001. Heeringa M. Stricker BHC.135:489-91. Stricker BHC. Kwee-Zuiderwijk WJM. 194 . waarschijnlijk uitgelokt door claritromycine. Acuut delier. van Puijenbroek EP. van Puijenbroek EP. Leufkens HGM. Drugassociated acute pancreatitis: twenty-one years of spontaneous reporting in The Netherlands. Heeringa M. Wilson JH. Toeval of toeval? Epilepsie en bijwerkingen. Stricker BHC. Neuropsychiatrische bijwerkingen toegeschreven aan het gebruik van oxybutynine. Van Puijenbroek EP. Ned Tijdschr Geneeskd 1998. Ann Intern Med 1998. Reversible dysgeusia attributed to losartan. Meyboom RHB. Tendon disorders attributed to fluoroquinolones: a study on 42 spontaneous reports in the period 1988 to 1998. Zondervan PE. Egberts ACG. Meyboom RHB. Verlenging van de QTc-tijd bij een pasgeborene tijdens gebruik van cisapride. Herings RM. Eur J Clin Pharmacol 2000. Diemont WL. Heerdink ER. Signalling possible drug-drug interactions in a spontaneous reporting system: delay of withdrawal bleeding during concomitant use of oral contraceptives and itraconazole. Am J Gastroenterol 1999.

van Grootheest AC.10:135-142. Diemont W. Egberts ACG.List of publications *Van Puijenbroek EP. Meyboom RHB. 195 . fever and urticaria: symptoms or syndrome? Pharmacoepidemiol Drug Saf 2001. *Van Puijenbroek EP. Egberts ACG. Leufkens. Determinants of signal selection in a spontaneous reporting system for adverse drug reactions Br J Clin Pharmacol: accepted for publication. Association between terbinafine and arthralgia. Meyboom RHB. Leufkens HGM.

.

197 .Curriculum vitae Eugène van Puijenbroek was born in Tilburg. dr. He is married with Emmy Konst. Following the completion of pre-university education at the Rijksscholengemeenschap Koning Willem II in Tilburg (HAVO and Atheneum B) in 1979. he started his medical study at the University of Nijmegen in the same year. From 1989 till 1991 he worked as a medical adviser at the ‘Sociaal Fonds Bouwnijverheid’. In November 1998 he started the research project described in this thesis in collaboration with the Department of Pharmacotherapy and Pharmacoepidemiology. the Netherlands on September 28. HGM Leufkens) at the University of Utrecht. 1960. Additionally he works as a general practitioner in the township of Vught. Utrecht Institute for Pharmaceutical Sciences (Head: Prof. He obtained his medical degree in 1987 after which he fulfilled his military service as a physician in the Dutch army. They have two daughters: Linde and Sterre. In 1994 he started his work as a research associate and subsequently as head of the analysing department of the Netherlands Pharmacovigilance Foundation Lareb. He was trained as general practitioner at the University of Maastricht from 1991 till 1993.

Sign up to vote on this title
UsefulNot useful