Correspondence

Reporting on the modes In addition, we encourage medical The Lancet has rightly emphasised
journals to pay more attention to that authors of papers must make
of data collection the way in which the methods are a substantial contribution.3 It is also
To gather more insight into the reported to improve the possibilities important that all major contributors
methods of data collection currently of critical appraisal. are recognised, either as an author or
used in medical research, we analysed We declare that we have no conflicts of interest. with an acknowledgment.
research studies published in four I have received payment for lectures or speakers’
high-impact general medical journals
*Marleen M H J van Gelder, bureaux from Dako, Ipsogen, Hammersmith Hospital,
Reini W Bretveld, Nel Roeleveld and Netherlands Cancer Institute.
and three epidemiological journals
[Link]@[Link]
in 2008–09.1 Two epidemiologists Andrew H S Lee
Department of Epidemiology, Biostatistics and HTA,
independently assessed the modes Radboud University Nijmegen Medical Centre,
[Link]@[Link]
of data collection reported in each 6500 HB Nijmegen, Netherlands Department of Histopathology, Nottingham
research paper. 1 Van Gelder MMHJ, Bretveld RW, Roeleveld N. University Hospitals, City Hospital Campus,
Surprisingly, the proportion of Web-based questionnaires: the future in Nottingham NG5 1PB, UK
epidemiology? Am J Epidemiol 2010; 1 Kang S. Anecdotes in medicine—15 years of
inconsistencies between the two 172: 1292–98. Lancet case reports. Lancet 2010; 376: 1448–49.
reviewers was high, especially for 2 Armstrong BK, White E, Saracci R. Principles of 2 Wilson R, Liston J, Cooke J, et al. Clinical
papers published in general medical exposure measurement in epidemiology. guidelines for breast cancer screening
New York: Oxford University Press, 1992. assessment. Sheffield: NHS Cancer Screening
journals (about 30%). Further 3 Rothman KJ, Greenland S, Lash TL. Validity in Programmes, 2005.
examination revealed that these epidemiologic studies. In: Rothman KJ, 3 The Lancet. The role and responsibilities of
inconsistencies were mainly due to Greenland S, Lash TL. Modern epidemiology, coauthors. Lancet 2008; 372: 778.
3rd edn. Philadelphia: Lippincott Williams &
unclear reporting of the methods used, Wilkins, 2008: 137–46.
with phrases such as “Information
was collected on [list of variables]”,
“Race/ethnicity was assessed by the All important
investigator or study coordinator”, and
“Sociodemographic, clinical, treatment
contributions to papers
(…), and laboratory data are collected” should be recognised
without any specification. Did they
use questionnaires or interviews, were In her Comment (Oct 30, p 1448),1
any measurements taken, or was it all Seema Kang discusses Case Reports
Science Photo Library

hearsay? in The Lancet according to specialty

The choice of the method of data interest and mentions that “more than
collection for a particular study one specialty might be involved in
Department of Error
depends on several factors, including, managing a case”. She comments on the
Tatem AJ, Smith DL, Gething PW, Kabaria CW,
but not limited to, the type of study, absence of contributions from general Snow RW, Hay SI. Ranking of elimination
sensitivity of the topic of interest, practice, but makes no comment on feasibility between malaria-endemic countries.
and costs of the measurements.2 Valid the contribution of other specialities Lancet 2010; 376: 1579–91—In this Series
paper (Nov 6), the fourth sentence under the
measurement of exposures, outcomes, such as radiology and pathology. Many heading “Estimation of intensity of endemic
and potential confounders is essential patients require multidisciplinary in- P falciparum transmission” should have read:
in medical research to prevent biased vestigation or management. For “R0 is a measure of maximum potential
transmission, and if R0 is 1 or greater, then
results.3 Since different methods of example, in the diagnosis of breast
endemic malaria transmission can be
data collection yield various amounts disease, the triple approach of clinical sustained”. Additionally, the title for figure 1
of measurement error, detailed examination, radiology, and pathology should have read “Categorical map of
is recommended.2 Plasmodium falciparum reproductive number,
reporting on the methods used is of
PfRc, indicating the extent to which
great importance to assess the quality Over the past 2 years (November, transmission needs to be reduced for
of the study by both readers and 2008, to October, 2010) histopathology elimination”. These corrections have been
reviewers or editors. Additionally, an had an important role in the diagnosis made to the online version as of Dec 31, 2010.

adequate description of the methods in 36 Case Reports in The Lancet. A The InFACT Global H1N1 Collaboration. InFACT:
of data collection used enables other histopathologist was an author in 20. a global critical care research response to H1N1.
research groups to replicate the In the other 16 there was no apparent Lancet 2010; 375: 11–13—In this Comment
(Jan 2), the name of the third member of the
original study. recognition of histopathology in terms InFACT Global H1N1 Collaboration was mis-
In our view, researchers should of authorship or acknowledgment, spelled. The correct spelling is “Neill K J
improve the description of the modes although it had an essential role in ten Adhikari”. This correction has been made to
the online version as of Dec 31, 2010.
of data collection used in their studies. and was contributory in six.

30 [Link] Vol 377 January 1, 2011