doi: 10.1111/j.1472-8206.2011.00924.

ORIGINAL Medication prescribing errors pertaining

ARTICLE
to cardiovascular/antidiabetic medications:
Fundamental & Clinical Pharmacology

a prescription audit in primary care

Khalid A.J. Al Khajaa*, Reginald P. Sequeiraa, Awatif H.H. Damanhorib
a
Department of Pharmacology & Therapeutics, Arabian Gulf University, Manama, Kingdom of Bahrain
b
Primary Care, Ministry of Health, Manama, Kingdom of Bahrain

Keywords
Bahrain,
cardiovascular/antidiabetic
medications,
prescribing errors,
prescription audit,
primary care
Received 12 July 2010;
revised 4 November 2010;
accepted 15 December 2010
*Correspondence and reprints:
khlidj@agu.edu.bh
ABSTRACT
This study was carried out to identify the medication prescribing errors (MPEs)
pertaining to cardiovascular/antidiabetic medications in prescriptions issued to
hypertensive and diabetic hypertensive patients. A retrospective, nationwide audit of
prescriptions (n = 2773) issued by primary care physicians (n = 194) of 20 health
centres in Bahrain was carried out. Approximately one-quarter of prescriptions
ordered by two-thirds of primary care physicians had errors. No significant
differences with respect to overall errors were evident in prescriptions ordered by
the family physicians and general practitioners. The most common error (in 8.0% of
prescriptions) was prescribing b-blockers or diuretics (thiazide) or their combinations
to patients on lipid-lowering drugs. Prescribing multiple antihypertensives, often with
a similar mechanism, accounted for 2.2% errors: approximately half of these (1.45%)
were two angiotensin-converting enzyme inhibitors (ACEIs) co-prescribed and/or
ACEIs plus angiotensin-II receptor blockers. In 0.7% of prescriptions, b-blockers were
ordered to patients on salbutamol treatment. High-dose metformin (3 g/day) was
prescribed to approximately 4% diabetic hypertensives; of these, many were elderly
patients. Prescribing high-dose glibenclamide (median dose 15 mg) to the elderly
accounted for 3.6% of the overall errors. Polypharmacy, such as aspirin along with
an immediate-release dipyridamole, was prescribed occasionally (0.25%), particularly
by the general practitioners (P = 0.0139). MPEs are common in primary care, in
Bahrain. Some of these prescribing errors have the potential to harm patients.
Effective measures to detect and prevent such errors are needed to improve the
quality of health care. Standard treatment guidelines and educational interventions
are important strategies to achieve these goals.

of medication errors, are preventable [4]. Medication

INTRODUCTION
The rate of prescribing errors has been estimated to be
11% of prescriptions in primary care [1] in contrast to
1.5% in hospitals [2]. Errors resulting in preventable
adverse drug events (ADEs) occurred most often at the
stages of ordering (56%) or administering (34%),
whereas transcribing (6%) and dispensing errors (4%)
were less common [3]. It has been estimated that
approximately one-third of ADEs in outpatients, because
errors and ADEs are common in the ambulatory setting
and are estimated to cost the US health care system $
177 billion annually [5]. Nonetheless, little is known
about the type or frequency of medication errors in
primary care, where the majority of patients are in
contact with health care providers [1,4,6].
Medication error is defined as ‘a failure in the
treatment process that leads to, or has the potential to
lead to, harm to patient’ [7]. A failure in the treatment

ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
410 Fundamental & Clinical Pharmacology 26 (2012) 410–417
Prescribing errors in primary care
process includes any error in the prescribing, transcrib-
ing, manufacturing or compounding, dispensing, admin-
istrating and monitoring of a drug [7]. Based on a
psychological analysis, errors are classified into errors in
planning action and errors in executing correctly
planned actions. Errors in planning actions can be
further classified into knowledge-based errors (KBEs) and
rule-based errors [7]. ‘A prescribing error occurs when,
as a result of a prescribing decision or prescription
writing process, there is an unintentional significant
(i) reduction in the probability of treatment being timely
and effective or (ii) increase in the risk of harm’[8]. This
definition has been further divided into two categories:
(i) decision making–based error that is equivalent to
knowledge-based error (KBE) defined by Aronson et al.
[7] and (ii) prescription writing error.
Prescribing errors have been evaluated at primary
care setting in Bahrain [9,10]. A comprehensive under-
standing of the frequency and nature of medication
prescribing errors (MPEs) along with prescription writing
errors is essential in developing health policies to reduce
actual or potential therapy-related harm to patients.
Hence, a multicentric primary care–based study has
been carried out to specifically identify the frequency and
nature of MPEs pertaining to cardiovascular/antidiabetic
medications in prescriptions issued by primary care
physicians to patients with hypertension and diabetic
hypertension.
METHODS
Setting
The Kingdom of Bahrain is a group of islands located at
Arabian Gulf with a population of 1 046 814 at the time
of the study. A primary health care organization with a
network of 20 health centres spread across the country
provides subsidized access to health care including
dispensing essential drugs. The number of primary care
physicians in each health centre varied between 4 and
11, with a randomly mixed ratio of family physicians
(FPs) and general practitioners (GPs). Patients requiring
special investigations and consultations or admission are
referred to Salmaniya Medical Complex, a hospital that
provides secondary/tertiary care.
Study population and variability
All patients with hypertension who received antihyper-
tensive drugs and all diabetic hypertensives who received
antidiabetics (both oral antidiabetics and insulin) and
antihypertensive drugs concomitantly were included in
ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
Fundamental & Clinical Pharmacology 26 (2012) 410–417
411
the study. Prescriptions issued in September and October
2007 with a refill order–based requests of 3 months
were collected in November and December 2007,
respectively, by the pharmacists incharge at these health
centres. This design was used to avoid duplication of
prescriptions, to exclude drugs prescribed as short-term
trial therapy and to reflect the actual prescribing
behaviour of primary care physicians with respect to
antihypertensives and antidiabetics as a part of ongoing
comprehensive study.
Operational definitions
A patient was identified as a hypertensive if he or she
received one or more antihypertensives and was identi-
fied as diabetic with hypertension if he or she received
one or more antihypertensives together with one or
more antidiabetic drugs. FPs are physicians who had
successfully completed the Family Practice Residency
programme in Bahrain[10]. This four-year programme is
recognized by the Arab Board Council for Family and
Community Medicine and is affiliated with the Irish
College of General Practitioners and the Royal College of
Surgeons, Ireland. The FPs comprise two-thirds of
primary care physicians. GPs are medical graduates
licensed to practice in Bahrain.
Medication prescribing errors in this study were
defined as any errors identified only in the process of
prescribing (writing the prescription) and classified as
follows: (i) KBE defined as an error in planning action [7]
or an error in decision-making [8] that lead to, or has the
potential to lead to, harm to patient [7]. Potential drug–
drug interactions or drug allergies that may reflect a
failure of the prescriber to integrate information about
the patient or his/her history (i.e. error of integration)
are some examples for KBEs. (ii) Prescribing errors that
involved elements of commission errors such as wrong
dose or dosage forms, wrong/inappropriate drug selec-
tion or its indication, wrong quantity or duration of
therapy and drug duplication are defined as inconsistent
prescribing errors. Errors of omission are defined as
absence, vague, incomplete and/or illegibility of any
components of the body of the prescription [10] and are
excluded in this study.
Data validation and coding
Based on defining criteria, the eligible prescriptions were
carefully audited by the first author and then indepen-
dently reviewed by second and third authors. Discrep-
ancies were resolved by discussion. Errors were coded,
and data were entered for analysis by the first author.
412
Medications
Single antihypertensive drugs available at the health
centres were angiotensin-converting enzyme (ACE)
inhibitors (captopril, cilazapril, enalapril, lisinopril and
perindopril); angiotensin-II receptor blockers (ARBs)
(valsartan); b-blockers (atenolol, propranolol); calcium
channel blockers (CCBs) [amlodipine, nifedipine (sus-
tained release), diltiazem]; diuretics [chlorthalidone,
hydrochlorothiazide (HCTZ), indapamide (sustained
release) and furosemide; and others (methyldopa, hydral-
azine)]. Antihypertensive fixed-dose combinations were
as follows: HCTZ + triamterene (dyazide); reserpine +
dihydroergocristine + clopamide (Brinerdin); HCTZ +
valsartan (Co-Diovan); perindopril + indapamide
(Preterax and Bi-preterax). Among antidiabetic drugs,

biguanide (metformin), second-generation sulphonylu-
rea drugs (glibenclamide, gliclazide, glipizide) and
various preparations of short- and immediate-acting
insulins were available.
Statistical analysis
Data were analysed using the Statistical Package for
Social Sciences (SPSS/PC+; version 14.0; SPSS Inc.,
Chicago, IL, USA). Descriptive statistics such as percent-
age, mean and standard deviation were used to test the
difference between proportions, and two-tailed t-test was
used for continuous variables. A P-value <0.05 was
considered statistically significant.
RESULTS
Of 5992 prescriptions collected in November and
December 2007, 2936 (49.0%) had a refill order of
3 months that fulfilled the inclusion criteria. About one-
quarter (26.4%) of prescriptions that met inclusion
criteria (733 of 2773) had MPEs and were prescribed
Table I Prescriber (physician) and prescription characteristics.
Number of prescribers
Number and percentage of prescribers committing MPEs (%)
Number of prescriptions issued by each prescriber category
Number and percentage of MPEs in each prescriber category (%)
MPEs, Medication prescribing errors;
a
Total # of prescriptions was 2936, of these 163 were without physician’s stamp (unknown category).
b
Difference between FPs and GPs (P = 0.449).
c
Total # of MPEs was 761, of these 28 were without physician’s stamp (unknown category).
ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
K.A.J. Al Khaja et al.
by approximately two-thirds of primary care physicians
(Table I). No significant differences with respect to overall
MPEs were evident in prescriptions issued by the FPs and
GPs (Table I and II). Prescribing of b-blockers (notably
atenolol) or diuretics (thiazide and thiazide-like diuretics)
or their co-prescribing to patients on lipid-lowering drugs
was the most common inconsistent prescribing errors in
8% of prescriptions (Table II). Prescribing of multiple
antihypertensives that act through a similar mechanism
accounted for 2.2% of errors: approximately half of these
prescriptions (1.45%) comprised two ACE inhibitors co-
prescribed and/or ACE inhibitors plus ARBs (Table II). In
0.7% of prescriptions, b-blockers were ordered to patients
on salbutamol treatment as well. High-dose metformin
(3 g/day) was prescribed to approximately 4% diabetic
hypertensives; of whom, many were elderly patients.
Prescribing high-dose glibenclamide (median dose
15 mg) to the elderly accounted for 3.6% of the overall
MPEs. Polypharmacy such as aspirin along with imme-
diate-release dipyridamole was prescribed occasionally
(0.25%), by GPs rather than FPs (0.58 vs. 0.06;
P = 0.0139).
DISCUSSION
Auditing MPEs is an important strategy to improve
patient care, promote rational prescribing and enhance
the quality of prescribing. This study provides an insight
into the nature of MPEs by primary care physicians in
prescribing cardiovascular and antidiabetic medications
to hypertensive and diabetic hypertensive patients.
Approximately one-quarter (26.4%) of prescriptions
had errors, and these medications were prescribed by
two-thirds of primary care physicians. No significant
differences between GPs and FPs were observed regard-
ing overall prescribing errors. Prescribing b-blockers,
Physician category
Family Physicians (FPs) General practitioners (GPs) Total
114 80 194
84 (73.3) 55 (68.8) 139 (71.6)
1732 1041 2773a
449 (25.9) 284 (27.3b) 733c (26.4)
Fundamental & Clinical Pharmacology 26 (2012) 410–417
Prescribing errors in primary care 413

Table II Types of knowledge-based errors of family physicians (FPs) and general practitioners (GPs) in prescriptions issued for hypertensive
and diabetic hypertensive patients.

Prescriptions according to physician category

Inappropriate prescribing FPs (1732) GPs (1041) Unknowna (163) Total (2773)

Antihypertensives
b-blockersb/diureticsc to patients on lipid- lowering drugs 127 (7.33) 95 (9.13)p 6 222 (8.01)
b-blockersb/diureticsc to diabetic hypertensive patients 60 (3.46) 43 (4.13) 7 103 (3.71)
Multiple antihypertensives with similar mechanisms
ACEId + ARBe 21 (1.21) 8 (0.77) – 29 (1.05)
ACEId + ACEId 6 (0.35) 5 (0.48) – 11 (0.4)
Diureticc + Diureticc 6 (0.35) 5 (0.48) – 11 (0.4)
Calcium channel blockers
DHPf + DHPf 3 (0.17) 1 (0.10) – 4 (0.14)
DHPf + Non-DHPg 2 (0.12) – – 2 (0.07)
b-blockerh + b-blockeri 4 (0.23) 1 (0.10) – 5 (0.18)
b-blockersb/diureticsc to diabetic patients on lipid-lowering drugs + allopurinol 19 (1.09) 15 (1.44) 4 34 (1.23)
Diureticsc to patients on allopurinol 18 (1.04) 12 (1.15) 2 30 (1.08)
b-blockersh,i with salbutamol 16 (0.92) 3 (0.29)q – 19 (0.69)
High dose of diureticsj in malesk 7 (0.40) 1 (0.10) 1 8 (0.29)
High dose of diureticsj in femalesl 8 (0.46) 1 (0.10) – 9 (0.32)
b-blockers with non-DHPh calcium channel blockers 6 (0.35) – – 6 (0.22)
High dose of furosemidem to hypertensives 2 (0.12) – – 2 (0.07)
Carvedilol in dosage form unavailable 2 (0.12) – – 2 (0.07)
Indapamide in dosage form unavailable 2 (0.12) – – 2 (0.07)
Antidiabetics
High dose of metformin (3 g daily in divided dose)n 73 (4.21) 38 (3.65) 5 111 (4.00)
Glibenclamideo to ‡ 65 year old patients 57 (3.29) 42 (4.03) 3 99 (3.57)
Glibenclamideo 20 mg/day + atenolol 100 mg/day to ‡65 year old patients 5 (0.29) 6 (0.58) – 11 (0.4)
Glibenclamideo 20 mg/d + insulin to 65 ‡ year old patients 3 (0.17) 1 (0.10) – 4 (0.14)
Insulin + atenolol 100 mg/day to ‡65 year old patients 1 (0.06) 1 (0.10) – 2 (0.07)
Antiplatelets
Aspirin with dipyridamole 1 (0.06) 6 (0.58)r – 7 (0.25)
Total n (%) 449 (25.93) 284 (27.31) 28 (0.95)s 733 (26.43)

a
Prescriptions without physician’s stamp.
b
Atenolol ‡50 mg.
c
Indapamide 1.25 mg/hydrochlorothiazide ‡25 mg/chlorthalidone ‡25 mg/dyazide 25 mg.
d
Angiotensin-converting enzyme inhibitors.
e
Angiotensin-II receptor blockers.
f
Nifedipine/amlodipine.
g
Diltiazem.
h
Atenolol.
i
Bisoprolol/carvedilol.
j
Hydrochlorothiazide/chlorthalidone ‡37.5 mg.
k
Mean age of male patients (year) and dose administered (mg) were 54.8 ± 12.7 and 51.4 ± 4.2, respectively.
l
Mean age of female patients (year) and dose administered (mg) were 57.2 ± 10.9 and 47.2 ± 5.5, respectively.
m
Furosemide 80 mg bid.
n
28 (24.1%) cases were patients ‡65 year old.
o
Standard formulation.
p
P = 0.0967 (difference between FPs and GPs).
q
P = 0.0573 (and difference between FPs and GPs),
r
P = 0.0139 (differences between FPs and GPs);
s
Prescriptions with errors without physicians’ stamp (unknown category).

ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
Fundamental & Clinical Pharmacology 26 (2012) 410–417
414 K.A.J. Al Khaja et al.

notably atenolol, or diuretics (thiazide and thiazide-like [21]. Until this dual RAAS blockade is proved beneficial,
diuretics) or their combinations to patients with hyper- they should be avoided in patients with uncomplicated
tension and diabetic hypertension on lipid-lowering hypertension. A subanalysis of data revealed that the
drugs was the most common inconsistent prescribing. prevalence of antihypertensive combinations, particu-
With the exception of b-blockers such as carvedilol and larly irrational multiple diuretics (with the exception of
nebivolol, other b-blockers, particularly atenolol, can potassium sparing and thiazide combinations) and
worsen dyslipidaemia and increase the incidence of new controversial combinations of RAAS inhibitors, was
onset diabetes [11,12]. Nonetheless, b-blockers are found to be associated with prescriptions comprising
strongly recommended in patients with angina and fixed-dose preparations, such as Bi-preterax, Preterax
post-myocardial infarction, unless contraindicated [13]. and Co-Diovan (81.1 and 55.2% in case of multiple
Thiazide diuretics are less expensive than angiotensin- diuretics and RAAS-based combinations, respectively;
converting enzyme inhibitors (ACEIs), ARBs and CCBs Table III). The plausible explanation for such irrational
but have dyslipidaemic and diabetogenic effects partic- and controversial combinations perhaps is because of a
ularly at high doses [11]. In long term, diuretics can lack of physicians’ awareness about the active constit-
reduce the benefit of treatment and increase the cost uents of these fixed-dose preparations, resulting in
owing to the need for further pharmacological therapy polypharmacy (Table III).
that is often needed to treat the metabolic abnormalities Of note, dual CCBs therapy has been reported to be an
[14] The metabolic adverse effects of both b-blockers and option for individuals who are refractory to usual
thiazide diuretics may be even more profound when stepped care of antihypertensive drug regimens [24].
these are co-administered [13]. Our study showed that dual nifedipine–diltiazem therapy
The optimal combination of antihypertensives is based
on selecting at least two antihypertensives with comple-
mentary mechanisms to provide additive or synergistic
Table III Pattern of multiple diuretics and multiple RAAS inhibitor
antihypertensive effect with minimal likelihood of dose- related knowledge-based errors.
dependent adverse effects [13,15]. However, combina-
tion of antihypertensives that act through a similar Inappropriate prescribing Frequency
mechanism may result in less than additive antihyper-
Diuretic + Diuretic
tensive effects and increases the risk of adverse effects, Indapamide + Hydrochlorothiazide 1
particularly if the drugs combined have similar side Indapamide + Furosemide 1
effects [13]. In this study, prescribing of multiple Indapamide + Co-Diovana 4
antihypertensives with a similar mechanism accounted Furosemide + Co-Diovana 2
for 2.2%; approximately half of these prescriptions were Furosemide + Preterxb 1
Hydrochlorothiazide + Bi-preteraxc 2
with multiple drugs interfering with renin-angiotensin-
Total (n) 11
aldosterone system (RAAS), particularly with ACEIs and
Mean antihypertensive combination ± SD (range) 3.9 ± 0.8 (3–5)
ARBs (Table II). Although a dual blockade of RAAS has a Mean drugs per prescription (range) 6.5 ± 2.8 (4–12)
positive effect on proteinuria in both diabetic and non- Proportion of fixed-dose combinations 81.8
diabetic nephropathies [16–18], and additive benefits in ACEIsd + ARBse
heart failure–related morbidity and mortality [19,20], it Enalapril/Lisinopril/Perindopril + Valsartan/Irbesartan 13
is not recommended in adults with hypertension without Captopril/Enalapril/Perindopril + Co-Diovana 8
Cilazapril + Perindopril + Co-Diovana 1
compelling indications for specific antihypertensive
Valsartan + Bi-preteraxc 5
agents [15,21,22]. This view can be justified because a
Bi-preteraxc + Co-Diovana 2
combination of two antihypertensives with overlapping Total (n) 29
mechanism of action (an ACE with an ARB) is deemed Mean antihypertensive combinations ± SD (range) 3.9 ± 1.3 (2–6)
controversial. Also, large randomized clinical trials to Mean drugs per prescription (range) 7.1 ± 2.5 (3–15)
confirm the additive or synergistic effect of dual blockade Proportion of fixed-dose combinations 55.2
of RAAS in controlling blood pressure are lacking [23]. a
Valsartan 80 mg + hydrochlorothiazide 12.5 mg;
Dual RAAS blockade in adults with hypertension was b
Perindopril 2 mg + indapamide 0.625 mg;
associated with more adverse effects such as hypotensive c
Perindopril 4 mg + indapamide 1.25 mg;
symptoms, syncope, renal dysfunction and hyperkala- d
Angiotensin-converting enzyme inhibitors;
e
emia without offering an increase in beneficial effects Angiotensin-II receptor blockers.

ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
Fundamental & Clinical Pharmacology 26 (2012) 410–417
Prescribing errors in primary care
accounted for 0.07% of overall MPEs; it was only
prescribed by FPs. It has been reported that co-prescrip-
tion of felodipine with diltiazem provides significant
additional antihypertensive effects [25]. However, Kar-
otsis and his colleagues [26] reported that despite dual
CCBs beneficial effects, approximately 29% of treated
patients stopped treatment because of intolerable ankle
oedema. Whereas controlled data for evaluating dual
CCBs as standard form of antihypertensive therapy are
sparse, we opine that this approach, particularly for
treating uncomplicated hypertension, would remain
controversial and premature.
Our study showed that some patients were prescribed
high doses of thiazide diuretics (Table II). The high dose
of thiazides should be replaced by that of optimal dose
(6.25–25 mg/day hydrochlorothiazide equivalent),
whether alone or in combination with possibly RAAS
inhibitors, to achieve optimal blood pressure control with
minimal electrolyte disturbances and metabolic adverse
effects [14]. It is well known that use of thiazides is
associated in some patients with erectile dysfunction that
can be averted if the dose is kept as low as possible. Of
note, chlorthalidone prescribed at a dose even <25 mg/
day has been reported to cause erectile dysfunction
[27,28]. Lack of gender-based differences with respect to
prescribing high doses of thiazides in our study is an
issue that needs further attention.
b-blockers are contraindicated in patients with
obstructive airway disease [29,30]. This KBE, with a
potential to cause clinically significant drug–disease
interaction, was more common with FPs than GPs
(P = 0.053) who prescribed moderate-to-high mean
dosages of atenolol (69.1 mg) and carvedilol (25 mg;
for referral patients; data not shown).
Metformin was prescribed at a dose of 3 g/day,
although the usual maximum dose recommended is
2 g/day [29]. Higher doses are often associated with
frequent gastrointestinal adverse effects [30] and anae-
mia caused by vitamin B12 deficiency [31]. Approxi-
mately one-quarter (28 of 116) of metformin-treated
patients receiving 3 g/day were elderly (Table II). As
vitamin B12 deficiency occurs frequently in the elderly
[32] and can further be exacerbated by metformin, the
dose of metformin, particularly for elderly, should be
conservative and not titrated up to 3 g/day [33].
The rate of severe hypoglycaemia, in an attempt to
achieve tight glycaemic control, increases exponentially
with age as a result of associated physiological changes
influencing pharmacokinetics and pharmacodynamics of
antidiabetic drugs [34]. Because of long half-life, active
ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
Fundamental & Clinical Pharmacology 26 (2012) 410–417
415
metabolites and its greater tendency to suppress hepatic
glucose production, glibenclamide is associated with a
greater risk of hypoglycaemia [35,36]. Short-acting
sulphonylureas are better alternatives [29]. The current
study revealed that prescribing glibenclamide to elderly
accounted for approximately 3.6% of overall MPEs
(Table II). Although comparable findings have been
reported previously [37,38], continuation of such irra-
tional practice suggests that a little or no changes have
occurred in prescribing patterns.
Glibenclamide was prescribed to elderly diabetic
hypertensives at a mean dose of 13.9 mg ± 6.2 (med-
ian = 15 mg; data not shown), although the incidence
of hypoglycaemic episodes with glibenclamide is consid-
erably higher even at submaximal therapeutic dosage
that ranged between 5 and 10 mg/day [39]. Ageing is
associated with a decline in adrenergic receptors that
may compromise the response to hypoglycaemia in
elderly, presenting with neuroglycopenic symptoms than
adrenergic symptoms. Concurrent use of b-blockers,
which masks the adrenergic symptoms, may further
impair the hypoglycaemic response [40]. Caution should
be exercised with diabetic patients in whom b-blockers
are used concomitantly with either long-acting sulpho-
nylureas or insulin (Table II) because such inappropriate
combinations may lead to severe hypoglycaemic epi-
sodes.
A combination of aspirin (30–325 mg; median =
75 mg) with extended release dipyridamole (400 mg/day)
provided a significant reduction in secondary prevention of
stroke when compared to aspirin alone [41]. The superi-
ority of such combination has been supported by a meta-
analysis [42]. However, combination of aspirin (330 mg)
with immediate-release dipyridamole (225 mg/day) pro-
duced no significant beneficial effect compared to aspirin
alone [43]. Prescribing aspirin (153.4 ± 103 mg; med-
ian = 81 mg) plus an immediate-release dipyridamole
(171.4 ± 36.6 mg; median = 150 mg) observed in our
study is an irrational practice favoured by some GPs
rather than FPs (Table II; P = 0.0139) that has to be
discouraged; neither the dose of dipyridamole nor the
formulation used conforms with ESPRIT study recom-
mendations [41].
LIMITATIONS OF THE STUDY
Selection of prescriptions pertaining to cardiovascular/
antidiabetic medications with refill order–based requests
of three months may result in underestimated MPEs,
particularly those with actual consequences which
416
would call for the discontinuation of treatment during
this period.
CONCLUSION
Medication prescribing errors are common in primary
care in Bahrain. Some of these errors have the potential
to harm patients. No significant differences between GPs
and FPs were observed regarding overall MPES of
prescribing. Although many of these errors have been
reported earlier in studies conducted at the same primary
care setting, most of these errors continue unabated.
Effective measures to detect and prevent such errors are
needed to improve the quality of health care. Educational
intervention and determining the outcome are important
strategies to achieve this goal.
ACKNOWLEDGEMENT
We acknowledge the help and assistance given to us by
Mr. Moh’d Ghali Rashid (reference librarian) for infor-
mation retrieval and to Mrs. Radha Raghavan in
preparing and typing of this manuscript.
COMPETING INTEREST
There is no competing interest to declare. This work is
not supported or funded by any drug company.
FUNDING
Nil.
REFERENCES
1 Sandars J., Esmail A. The frequency and nature of medical error
in primary care: understanding the diversity across studies.
Fam. Pract. (2003) 20 231–236.
2 Dean B., Schachter M., Vincent C., Barber N. Prescribing errors
in hospital inpatients: their incidence and clinical significance.
Qual. Saf. Health Care (2002) 11 340–344.
3 Bates D.W., Cullen D.J., Laird N. et al. Incidence of adverse drug
events and potential adverse drug events : implications for
prevention. ADE Prevention Study Group. JAMA (1995) 274
29–34.
4 Gandhi T.K., Weingart S.N., Borus J. et al. Adverse drug events
in ambulatory care. N. Eng. J. Med. (2003) 348 1556–1564.
5 Ernst F.S., Grizzle A.J. Drug – related morbidity and mortality:
updating the cost of illness model. J. Am. Pharm. Assoc. (2001)
41 156–157.
6 Zavaleta–Bustos M., Castro–Pastrana L.I., Reyes–Hernandez I.,
Lopez-Luna M.A., Bermudez-Camps I.B. Prescription errors in
ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
K.A.J. Al Khaja et al.
primary care university unit: urgency of pharmaceutical care
in Mexico. Braz. J. Pharm. Sci. (2008) 44 115–125.
7 Ferner R.E., Aronson J.K. Clarification of terminology in
medication errors. Drug Saf. (2006) 29 1011–1022.
8 Dean B., Barber N., Schachter M. What is a prescribing error?
Qual. Health Care (2000) 9 232–237.
9 Al Khaja K.A.J., Al-Ansari T.M., Sequeira R.P. An evaluation of
prescribing errors in primary care in Bahrain. Int. J. Clin.
Pharmacol. Ther. (2005) 43 294–301.
10 Al Khaja K.A.J., Sequeira R.P., Al-Ansari T.M., Damanhori
A.H.H. Prescription writing skills of residents in a family
practice residency programme in Bahrain. Postgrad. Med. J.
(2008) 84 198–204.
11 Mancia G., Gassi G., Zanchetti A. A new onset diabetes and
antihypertensive drugs. J. Hyperten. (2006) 24 3–10.
12 Opie L.H., Schall R. Old antihypertensives and new diabetes.
J. Hypertens. (2004) 22 1453–1458.
13 World Health Organization/International Society of Hyperten-
siion (WHO/ISH). Guidelines for the management of mild
hypertension. J. Hypertens. (1999) 17 151–183.
14 Salvetti A., Ghiadoni L. Thiazide diuretics in the treatment of
hypertension: an update. J. Am. Soc. Nephrol. (2006) 17 S25–
S29.
15 Mancia G., de Backer G., Dominiczak A. et al. Guidelines for the
Management of Arterial Hypertension. The Task Force for
the Management of Arterial Hypertension of the European
Society of Hypertension (ESH) and of the European Society of
Cardiology (ESC). J. Hypertens. (2007) 25 1105–1187.
16 Arici M., Erdem Y. Dual blockade of the renin-angiotensin
system for cardiorenal protection: an update. Am. J. Kidney Dis.
(2009) 53 332–345.
17 Mogensen C., Neldam S., Tikkanen I. et al. Randomized
controlled trial of dual blockade of renin-angiotensin systemin
patients with hypertension, micro-albuminurea and non-insu-
lin dependent diabetes: the candesartan and lisinopril micro-
albuminurea (CALM) study. BMJ (2000) 321 1440–1444.
18 Nakao N., Yoshimura A., Morita H., Takada M., Kayano T.,
Ideura T. Combination treatment of angiotensin-II receptor
blocker and angiotensin-converting enzyme inhibitor in non-
diabetic renal disease (COOPERATE): a randomized controlled
trial. Lancet (2003) 361 117–124.
19 McMurray J.J., Ostergren J., Swedberg K. et al. Effects of
candesartan in patients with chronic heart failure and reduced
left ventricular systolic function taking angiotensin-converting
enzyme inhibitors: the CHARM-added trial. Lancet (2003) 362
767–777.
20 Krum H., Carson P., Farsang C. et al. Effect of valsartan added
to background ACE inhibitor therapy in patients with heart
failure: results from VaL-HeFT. Eur. J. Heart Fail. (2004) 6 937–
945.
21 Yousuf S., Teo K.K., Pogue J. et al. (ONTARGET Investigators)
Telmisartan, ramipril, or both in patients at high risk for
vascular events. N. Eng. J. Med. (2008) 358 1547–1559.
22 On behalf of the Canadian Hypertension Education Program
(CHEP). 2009 CHEP recommendations. An annual update. Can.
Fam. Physician (2009) 55 697–700.
Fundamental & Clinical Pharmacology 26 (2012) 410–417
Prescribing errors in primary care
23 Kolasinka-Malkowska K., Filipiak K.J., Gwizdala A., Tykarski A.
Current possibilities of ACE inhibitor and ARB combination
in arterial hypertension and its complications. Expert Rev.
Cardiovasc. Ther. (2008) 6 759–771.
24 Handler J. Dihydropyridine/Nonhydropyridine calcium channel
blocker combination therapy. J. Clin. Hypertens (2005) 7 50–
53.
25 Sassen J.J., Carter B.L., Brown T.E., Elliot W.J., Black H.R.
Comparison of nifedipine alone and with diltiazem or verapamil
in hypertension. Hypertension (1996) 28 109–114.
26 Karotsis A.K., Symeonidis A., Mastorantonakis S.E., Stergiou
G.S., Home-Di-Plus Study Group. Additional antihypertensive
effect of drugs in hypertensive subjects uncontrolled on
diltiazem monotherapy: a randomized controlled trial using
office and home blood pressure monitoring. Clin. Exp. Hyper-
tens. (2006) 28 655–662.
27 Grimm R.H. Jr, Grandits G.A., Prineas R.J. et al. Long-term
effects on sexual function of five antihypertensive drugs and
nutritional hygienic treatment in antihypertensive men and
women. Treatment of Mild Hypertension Study (TOMHS).
Hypertension (1997) 29(1pt1) 8–14.
28 Wassertheil-Smoller S., Blaufox M.D., Oberman A. et al. Effect
of antihypertensives on sexual function and quality of life:
The TAIM study. Ann. Intern. Med. (1991) 114 613–620.
29 Anonymous. British National Formulary (BNF) No. 58. BMJ
Group and RPS Publishing, London, 2009.
30 Sweetman S.C. Martindale. The Complete Drug Reference.
Pharmaceutical Press, London, 2007.
31 Ting R.Z., Szeto C.C., Chan M.H., Ma K.K., Chow K.M.
Risk factors of vitamin B12 deficiency in patients receiving
metformin. Arch. Int. Med. (2006) 166 1975–1978.
32 Metthews J.H. Cobalamin and folate deficiency in elderly.
Baillieres Clin. Haematol. (1995) 8 679–697.
ª 2011 The Authors Fundamental and Clinical Pharmacology ª 2011 Société Française de Pharmacologie et de Thérapeutique
Fundamental & Clinical Pharmacology 26 (2012) 410–417
417
33 Lee F.T.H. Advances in diabetes in the elderly. J. Pharm. Pract.
Res. (2009) 39 63–67.
34 Gregoria F., Ambrosi F., Fillipponi P., Manfrini S., Testa I. Is
metformin safe enough for aging type 2 diabetic patients?
Diabetes Metab. (1996) 22 43–50.
35 Groop L., Luzi L., Melander A. et al. Different effects of
glibenclamide and glipizide on insulin secretion and hepatic
glucose production in normal and NIDDM subjects. Diabetes
(1987) 36 1320–1328.
36 Cox A.R., Ferner R.E. Prescribing errors in diabetes. Br. J.
Diabetes Vas. Dis. (2009) 9 84–88.
37 Al Khaja K.A.J., Sequeira R.P., Mathur V.S. Prescribing patterns
and therapeutic implications for diabetic hypertension in
Bahrain. Ann. Pharmacother. (2001) 35 1350–1359.
38 Al Khaja K.A.J., Sequeira R.P., Mathur V.S., Damanhori A.H.H.,
Abdul Wahab A.W. Family physicians; and general practitio-
ners’ approaches to drug management of diabetic hypertension
in primary care. J. Eval. Clin. Pract. (2002) 8 19–30.
39 Papa G., Fedele V., Rizzo M.R. et al. Safety of type 2 diabetes
treatment with repaglinide compared with glibenclamide in
elderly people. Diabetes Care (2006) 29 1918–1920.
40 Hornick T., Aron D.C. Managing diabetes in the elderly: go easy,
individualize. Cleve. Clin. J. Med. (2008) 75 70–78.
41 The ESPRIT Study Group. Aspirin plus dipyridamole versus
aspirin alone after cerebral ischemia of arterial origin (ESPRIT):
randomized controlled trial. Lancet (2006) 367 1665–1673.
42 Verro P., Gorelick P.B., Nguyen D. Aspirin plus dipyridamole
versus aspirin for prevention of vascular events after stroke or
TIA: a meta-analysis. Stroke (2008) 39 1358–1363.
43 Bousser M.G., Eschwege E., Haguenayu M. et al. ‘AICLA’
controlled trial of aspirin and dipyridamole in the secondary
prevention of athero-thrombotic cerebral ischemia. Stroke
(1983) 14 5–14.