cations. About 8% of our population have diabetes, among
whom the great majority have type 2.1,2The 2 most common
oral pharmacologic strategies to manage type 2 diabetes in-
clude the use of agents that promote insulin release (e.g., sul-
fonylureas) or improve insulin sensitivity (e.g., metformin
and thiazolidinediones).3These strategies reduce hypergly-
cemia to a similar degree; how they affect mortality, however,
Since publication of the results of the University Group Di-
abetes Project (UGDP),4the relation between the use of sulfo-
nylureas and the risk of cardiovascular events has been ques-
tioned. Subjects treated with tolbutamide in the UGDP had a
significantly higher cardiovascular rate of death than those
Despite the UGDP observations, however, sulfonylureas
have been the mainstay of therapy for the past 40 years.5This
treatment choice persists probably because the UGDP study
design has had some controversy6–9and perhaps because a
plausible mechanism remains inevident.10–13Moreover, the
largest (3867 subjects) and longest (median follow-up 10.0
years) study of glucose control in people with type 2 diabetes,
the United Kingdom Prospective Diabetes Study (UKPDS),
showed that intensive therapy with glibenclamide (glyburide)
significantly reduced the rate of microvascular events.14Con-
trary to the UGDP results, sulfonylurea therapy was not asso-
ciated with an increased risk of death.14
In 2 large population-based cohort studies,15,16we previ-
ously observed that the use of metformin by people with newly
treated type 2 diabetes was associated with reduced risks of
morbidity and death compared with sulfonylurea monothera-
py. Questions remain, however, as to whether the observed re-
lation resulted from a protective effect imparted by metformin,
or a potentially harmful effect of the sulfonylureas.12,17,18The
purpose of the study we report here was to determine whether
the risk of death changes with the level of exposure to sulfon-
ylureas. We hypothesized that increased exposure to a harmful
medication would be associated with an increased risk of
death from all causes or from an acute ischemic event.
iabetes mellitus is a chronic, progressive disease
characterized by deteriorating glucose control and
increased risk of micro- and macrovascular compli-
• January 17, 2006 • 174(2) | 169
© 2006 CMA Media Inc. or its licensors
Scot H. Simpson, Sumit R. Majumdar, Ross T. Tsuyuki, Dean T. Eurich, Jeffrey A. Johnson
See related article page 185
Dose–response relation between sulfonylurea drugs
and mortality in type 2 diabetes mellitus:
a population-based cohort study
Background: Over the past 30 years, the relation between
use of sulfonylureas to treat type 2 diabetes and the risk of
cardiovascular events has been vigorously debated. The pur-
pose of this study was to determine if the risk of death
changes with level of exposure to sulfonylurea drugs.
Methods:This was a retrospective, inception cohort study us-
ing administrative data from Saskatchewan Health (1991–
1999). The 5795 subjects, identified by their first-ever dispen-
sation for an oral antidiabetic agent, were grouped according
to their use of such agents during follow-up. Potential sub-
jects using insulin or combination therapy were excluded. Ex-
posure level was defined by daily dose and degree of adher-
ence. Separate multivariate Cox proportional-hazard models
were constructed for each monotherapy group and used to
calculate the risk of death associated with higher versus
lower exposure category. Disease severity indicators were
identified among the administrative data and entered as co-
variates in each model. The main outcomes were all-cause
mortality and death from an acute ischemic event.
Results: The mean age of the cohort members was 66.3
(standard deviation [SD] 13.4) years; 43.4% were female;
and their mean duration of follow-up was 4.6 (SD 2.1) years.
First-generation sulfonylureas were used exclusively by 120
subjects; glyburide, by 4138; and metformin, by 1537. A
greater risk of death was associated with higher daily doses
of the first-generation sulfonylureas (adjusted hazard ratio
[HR] 2.1, 95% confidence interval [CI] 1.0–4.7) and glybur-
ide (HR 1.3, 95% CI 1.2–1.4), but not metformin (HR 0.8,
95% CI 0.7–1.1). Similar associations were observed for
death caused by an acute ischemic event.
Interpretation: Higher exposure to sulfonylureas was asso-
ciated with increased mortality among patients newly
treated for type 2 diabetes. The same relation was not ob-
served with metformin. This implies that the manner in
which blood glucose concentration is lowered may be as im-
portant as achieving recommended glucose targets.
We carried out a retrospective inception cohort study using
the administrative databases of Saskatchewan Health. These
databases record health care services used by registered bene-
ficiaries in the province who are elegible for prescription drug
benefits, who account for 91% of the provincial population of
about 1000000 people.19Saskatchewan residents 30 years of
age and older were eligible for inclusion if they were dis-
pensed an oral antidiabetic agent anytime during 1991–1996
(inclusive). These inclusion criteria were used to reduce the
chances of including any people with type 1 diabetes in the
cohort. The index date was defined as the first dispensation
date for an oral antidiabetic agent during the index period;
subjects were considered “new users” if they were not dis-
pensed antidiabetic agents for 1year before the index date.
Findings from the UKPDS and other investigations, includ-
ing our previous studies, suggest that strategies that do not use
sulfonylureas may be associated with different rates of death
and cardiovascular events.15,16,20,21This variation may be caused
by differences in patient characteristics or the pharmacologic
properties of the agents. Therefore, to avoid confounding by
pharmacologic effects, we identified individuals who used one
oral antidiabetic agent exclusively during the observation
period. Subjects were excluded if they were dispensed insulin
or oral antidiabetic agents from 2 or more classes. Three
groups were established according to exclusive dispensation
histories of metformin, glyburide or first-generation sulfonyl-
ureas (either chlorpropamide or tolbutamide). Study subjects
were monitored from their index date until Dec. 31, 1999,
death or termination of Saskatchewan Health coverage.
Two measures of level of exposure — dose and adherence
— were considered. To determine the average daily dose, we
divided the duration of follow-up in days into the total weight
of active drug dispensed therein. We identified a median daily
dose for each monotherapy group and assigned subjects to
the higher- or lower-dose subgroup if their average daily dose
was above or below the median, respectively. An adherence
rate was calculated by dividing the total expected duration for
all dispensations by the subject’s duration of follow-up.22We
developed an algorithm to estimate the expected duration for
each dispensation based on the number of tablets dispensed
and the feasible daily regimens taken from each antidiabetic
agent’s product monograph. Subjects with an adherence rate
over 0.8 were considered to have good adherence; this cut-off
is a well-accepted threshold.23,24In addition, we assumed that
subjects had discontinued oral antidiabetic drug therapy if
the interval between the last dispensation and the end of fol-
low-up was 6 months or more. Given that diabetes is a
chronic, progressive disease and that these subjects stopped
all antidiabetic drug therapy, including insulin, we included
these subjects in the poor adherence group. To test the sensi-
tivity of our definition for adherence, we applied different
thresholds for adherence rate (0.66–0.9) and discontinuance
For statistical analysis, the primary outcome was time
from the index drug-dispensation date to death from any
cause. The secondary outcome was time to death attributable
to an acute ischemic event. Cause of death was determined
from the vital statistics files of Saskatchewan Health; deaths
attributable to myocardial infarction (i.e., recorded with the
International Classification of Diseases, 9th revision [ICD-9]
code 410) or another ischemic cardiovascular event (ICD-9
codes 411–414) were noted. Multivariate Cox proportional-
hazards models were used to estimate the hazard ratio (HR)
for death while controlling for potential confounding factors.
Separate analysis models for each monotherapy group were
constructed. Subjects with the lowest level of exposure (either
lower daily dose or poor adherence) served as the reference
group for each HR reported.
Potential confounding variables that were entered into the
models included age, sex, nitrate use, chronic disease score,
number of physician visits and hospital admissions. The use
of nitrates has been used previously as a marker for the pres-
ence of cardiovascular disease.25,26The chronic disease score
uses pharmacy dispensation information for selected drugs
(e.g., angiotensin-converting-enzyme inhibitors, β-blockers,
inhaled bronchodilators, antiparkinsonian medications) to
estimate the presence of chronic diseases.27A relative weight
of disease burden was calculated based on the number of
chronic diseases identified by drug therapy during the follow-
up period of the subject; all subjects therefore had a mini-
mum chronic disease score of 2 because they were dispensed
oral antidiabetic agents. Interaction terms between each vari-
able and level-of-exposure variable were examined; but be-
cause none reached statistical significance (p < 0.05), they
were not considered in the final models.
As with any observational study, confounding of the statis-
tical analysis by indication, whereby “sicker” patients would
more likely receive higher doses and also be at higher risk of
death, was a major concern. We followed the methodology
proposed by Sackett and colleagues28and used available data
to control for this source of confounding. Assuming that
sicker patients would visit their physician more frequently
and be admitted more often to hospital, we entered the num-
ber of physician visits and hospital admissions into the multi-
variate analysis. We also calculated a propensity score to rep-
resent the likelihood of an individual receiving a higher drug
dose, given his or her characteristics.29As we had also ob-
served in another study with the same database,16the addition
of a propensity score did not make a substantial change to the
observed association; we therefore excluded this score from
the final models of analysis.
We identified 12 272 Saskatchewan residents who were new
users of oral antidiabetic agents. Of these, 1443 received in-
sulin and 4885 were dispensed oral antidiabetic agents from
2 or more different classes during the observation period. Be-
cause their information was insufficient to estimate a level of
exposure, we excluded 149 others who were monitored for
less than 6 months. Of the remaining 5795 subjects, 120 used
a first-generation sulfonylurea exclusively (607 accumulated
person-years of follow-up); 4138, glyburide monotherapy
(19 298 person-years of follow-up); and 1537, metformin
• January 17, 2006 • 174(2) | 170
monotherapy (6995 person-years of follow-up).
The mean duration of follow-up was 4.8 years (standard
deviation [SD] 2.1 yr) for subjects taking lower daily doses,
compared with 4.5 (SD 2.1) years for those taking higher
doses (p < 0.001). The mean age of those taking lower daily
doses was 65.6 (SD 14.1) years; higher daily doses, 67.0 (SD
12.6; p < 0.001). The average burden of chronic diseases, esti-
mated with chronic disease scores, was 8.1 (SD 4.3) among
subjects with lower daily doses and 8.2 (SD 4.0) among those
with higher doses (p = 0.36). Table 1 displays patient charac-
teristics stratified by monotherapy groups.
There were 1503 deaths during the study period, of which
372 (24.8%) were attributable to an acute ischemic event.
First-generation sulfonylurea monotherapy users had the
highest mortality (67.6 deaths per 1000 person-years), com-
pared with glyburide monotherapy users (61.4 deaths per
1000 person-years) and metformin monotherapy users (39.6
deaths per 1000 person-years).
The high-dose groups using first-generation sulfonylurea
monotherapy and glyburide monotherapy had higher rates of
death than their counterparts who took lower doses (Fig. 1A).
This statistical association was maintained after controlling
for age, sex, comorbidities, physician visits and hospital ad-
missions. Higher daily doses of metformin were not associ-
ated with an increased risk of death. Similar patterns of asso-
ciation were observed for deaths attributable to an acute
ischemic event (Fig. 1B).
To confirm the observed relation between daily dose and
increased risk of all-cause risk of mortality, we repeated the
analyses using adherence rate to dichotomize the monother-
apy groups. Good adherence to prescribed drug treatment
was associated with a higher risk of death in the first-genera-
tion sulfonylurea and glyburide groups (Fig. 1C), but not in
the 2 metformin subgroups. In our sensitivity analyses, these
relations did not change substantially with the any of the vari-
ations in threshold that we applied to define good versus poor
adherence or discontinuance.
This study used information from administrative health data-
bases to further explore the relation between individual anti-
diabetic drug classes and the risk of death among people with
type 2 diabetes. We found that the risk of death was dose-
related for chlorpropamide, tolbutamide and glyburide, even
after controlling for demographic variables and comorbidi-
ties. In contrast, higher doses of metformin were not associ-
ated with an increased risk of death.
Our observation of a dose–response relation for sulfonyl-
ureas fulfills another element of the Bradford–Hill conditions
to establish causality from observational studies.30Subjects
exposed to higher levels of sulfonylurea monotherapy had a
higher mortality. In contrast, the level of exposure to metfor-
min monotherapy had no apparent effect on mortality. The
results generated from the UGDP more than 30 years ago
suggested that sulfonylurea use may be associated with an in-
creased risk of cardiovascular events and death.4Since then,
investigations in animal and human models and post-hoc
analyses of clinical trial data and observational studies have
shown a consistent but perhaps underappreciated association
between the use of sulfonylurea drugs and poor cardiac out-
One biological mechanism for this relation, as hypothes-
ized by several authors,10,32,33may be impairment of ischemic
preconditioning. Sulfonylureas promote release of insulin
from pancreatic β cells by binding to the sulfonylurea recep-
tor and maintaining closure of the adenosine triphosphate
(ATP)-sensitive potassium channel.32In cardiac myocytes and
smooth muscle cells, the closure of ATP-sensitive potassium
channels impairs ischemic preconditioning, a phenomenon
that enables myocardial cells to survive brief periods of ische-
mia.10,33The effect on ischemic preconditioning varies sub-
stantially among the sulfonylureas, likely because ofthe wide
range of binding affinities for sulfonylurea receptors on car-
diac myocytes.12For example, glyburide has a high affinity for
• January 17, 2006 • 174(2) | 171
Table 1: Demographic and clinical characteristics of patients with type 2 diabetes mellitus prescribed drug monotherapy, by
median split of average daily dose*
First-generation sulfonylureaGlyburide Metformin
(n = 46)
(n = 74)
(n = 2071)
(n = 2067)
(n = 769)
(n = 768)
Age, mean (SD), yr
Female, no. (%)
Nitrate use, no. (%)
Follow-up, mean (SD), yr
Chronic disease score, mean (SD)
Physician visits, mean (SD)
Median no. of visits (IQR)
Hospital admissions, mean (SD)
Note: SD = standard deviation, IQR = interquartile range.
*Whether drug exposure was higher or lower than the median for the group of patients taking that drug.
†‡p < 0.05 when lower and higher daily dose groups within a treatment cohort are compared †by analysis of variance or ‡by χ2 test.
• January 17, 2006 • 174(2) | 172
Fig. 1: Hazard ratios among patients with type 2 diabetes mellitus, comparing 2 subgroups within each drug-monotherapy group:
1A, hazard ratios for deaths from all causes, according to median split for drug exposure (i.e., whether their individual daily dose was
more than the group median [higher subgroup] or less [lower subgroup — used as the reference]); 1B, for deaths attributable to an
acute ischemic event, also according to median split for drug exposure; and 1C, for all-cause mortality, according to drug-treatment ad-
herence (patients were assigned to the poor-adherence [reference] subgroup if their adherence rate was < 0.8 or if they stopped ther-
apy > 6 months before end of follow-up). Error bars indicate 95% confidence intervals. *Either chlorpropamide or tolbutamide.
Adjusted for age, sex, chronic disease
score (CDS) and nitrate use
Adjusted for age, sex, CDS, nitrate use,
physician visits and hospital admissions
n = 120
n = 4138
n = 1537
Deaths per 1000 person-
years, in subgroup shown
both pancreatic and cardiac sulfonylurea receptors, whereas
gliclazide and nateglinide have high selectivity for pancreatic
sulfonylurea receptors.12Equally plausible explanations for
sulfonylurea toxicity include direct arrhythmogenic effects,
associated weight gain, hypoglycemia and the toxicity of elev-
ated insulin levels.10,33
This study builds on observations from our previous analy-
ses illustrating a difference in the risks of morbidity and
death among classes of oral antidiabetic drugs used by pa-
tients.15,16The reason for this risk difference has been the
topic of much recent discussion.12,17,18The focus appears to
be on differences in pharmacologic effects, whereby sulfonyl-
ureas may be cardiotoxic and metformin may provide some
degree of cardioprotection.17Although the potential cardio-
toxic effects of sulfonylureas have been debated over the last
30 years,4,13the potential cardioprotective effects of metfor-
min, first shown by the UKPDS,20have not been well des-
cribed. In our current study, stratification by monotherapy
group eliminated confounding from pharmacologic differ-
ences between drug classes. This allowed us to evaluate the
effect of different levels of exposure by comparing the risk of
death among users of the same antidiabetic agent. Although
our study provides additional evidence to support the theory
of a cardiotoxic effect of sulfonylureas, it does not resolve the
question of whether the risk difference between antidiabetic
drug classes is the result of different pharmacologic effects.
There are several limitations to this observational study.
First, our observations are based on administrative data. Such
databases do not record clinical information such as the sub-
ject’s height, weight, blood pressure, cholesterol profiles or
glucose control. We believe that information on glucose con-
trol would not change our observations substantially, given
that the largest randomized controlled study involving people
with type 2 diabetes did not demonstrate that good glucose
control reduced the risk of all-cause mortality.14Nevertheless,
residual confounding, such as differences in underlying car-
diovascular risk between exposure groups, may partially
explain our observed differences in risk of death. We used
available prescription drug data (i.e., nitrate use to identify
subjects with cardiovascular disease, and the chronic disease
score as a measure of chronic disease burden), demographic
information (age and sex) and health resource utilization in-
formation (i.e., physician visits and hospital admissions) to
control for these underlying risks and to adjust for case mix.
Second, our measures for exposure share the same limita-
tions as other pharmacy claims databases.22We assumed that
drug acquisition is a surrogate marker for consumption and
thereby may have overestimated actual exposure.
Third, subjects using higher doses of oral antidiabetic
agents may have required this level of drug use to manage
higher blood-glucose levels or more advanced diabetes — a
form of confounding by indication. If confounding by indica-
tion were present, the positive association observed in the un-
adjusted model could be expected to move toward unity with
adjustment for disease severity.28The addition of physician
visits and hospital admissions to the multivariate models,
however, did not have a substantial effect on the observations.
Furthermore, enrolment criteria for our inception cohort con-
trolled for duration of disease by identifying subjects filling
their first prescription for an oral antidiabetic agent. Intrigu-
ingly, subjects in the group taking lower daily doses had a
longer follow-up than subjects prescribed higher daily doses.
Fourth, the index period for our data set predated the avail-
ability of the thiazolidinediones acarbose and repaglanide. If
ischemic preconditioning, rather than the other postulated
mechanisms, is responsible for the potential toxicity of the
older but commonly used sulfonylureas that we studied, then
this may become less of an issue as accepted practice shifts to
the newer sulfonylureas.12,13
In conclusion, we observed a dose–response relation be-
tween sulfonylurea exposure and risk of death. This evidence,
taken within the context of observations collected over the
last 30 years, suggests that clinicians should carefully assess
the need for sulfonylurea therapy in subjects at high risk of
cardiovascular events — particularly now, when several other
classes of antidiabetic oral medications are available.
1. Health Canada Centre for Chronic Disease Prevention and Control. Diabetes in
Canada. 2nd ed. Ottawa: The Centre; 2002.
2. Wild S, Roglic G, Green A, et al. Global prevalence of diabetes: estimates for the
year 2000 and projections for 2030. Diabetes Care2004;27:1047-53.
3. Inzucchi SE. Oral antihyperglycemic therapy for type 2 diabetes: scientific review.
• January 17, 2006 • 174(2) | 173
This article has been peer reviewed.
From the Institute of Health Economics (all authors); the Faculty of Phar-
macy and Pharmaceutical Sciences (Simpson, Tsuyuki); the Divisions of Gen-
eral Internal Medicine (Majumdar) and Cardiology (Tsuyuki), Department of
Medicine; and the Department of Public Health Sciences (Majumdar,
Tsuyuki, Eurich, Johnson), University of Alberta, Edmonton, Alta.
Competing interests: None declared.
Contributors: Scot Simpson was responsible for development of the study
concept and design and the preparation of the first draft of this manuscript.
Sumit Majumdar and Jeffrey Johnson contributed substantially to study de-
sign and data acquisition. All authors participated in the analysis and inter-
pretation of the data and critical revision of the manuscript, and all support
the findings of the study.
Acknowledgements: This study is based, in part, on de-identified data pro-
vided by the Saskatchewan Department of Health. The interpretation and
conclusions contained herein do not necessarily represent those of the Gov-
ernment of Saskatchewan or the Saskatchewan Department of Health.
Scot Simpson is a New Investigator supported by the Canadian Institutes
of Health Research (CIHR). Sumit Majumdar is a New Investigator supported
by CIHR and a Population Health Investigator with the Alberta Heritage
Foundation for Medical Research (AHFMR). Ross Tsuyuki holds the Merck
Frosst Chair in Patient Health Management from the Faculty of Pharmacy and
Pharmaceutical Sciences and the Faculty of Medicine and Dentistry, Univer-
sity of Alberta. Dean Eurich holds a Doctoral Research Award with CIHR and
a Full-time Studentship in Health Research with AHFMR. Jeffrey Johnson
holds a Canada Research Chair in Diabetes Health Outcomes, is a Health
Scholar supported by AHFMR and is Chair of a New Emerging Team (NET)
grant to the Alliance for Canadian Health Outcomes Research in Diabetes
(ACHORD). The ACHORD NET grant is sponsored by the Canadian Diabetes
Association, the Heart and Stroke Foundation of Canada, the Kidney Founda-
tion of Canada and CIHR (the Institute of Nutrition, Metabolism and Dia-
betes, and the Institute of Circulatory and Respiratory Health).
This study was funded in part by grants from the Institute of Health Econ-
omics and AHFMR. These study sponsors did not play any role in the study
design or conduct; the collection, analysis, and interpretation of data; the
writing of the report; or the decision to submit the paper for publication.
4. Meinert CL, Knatterud GL, Prout TE, et al. A study of the effects of hypoglycemic Download full-text
agents on vascular complications in patients with adult-onset diabetes. II. Mortal-
ity results. Diabetes1970;19(Suppl):789-830.
5. Groop LC. Sulfonylureas in NIDDM. Diabetes Care1992;15:737-54.
6. Cornfield J. The University Group Diabetes Program: a further statistical analysis
of the mortality findings. JAMA1971;217:1676-87.
7. Schor S. The University Group Diabetes Program: a statistician looks at the mortal-
ity results. JAMA1971;217:1671-5.
8. Seltzer HS. A summary of criticisms of the findings and conclusions of the Univer-
sity Group Diabetes Program (UGDP). Diabetes1972;21:976-9.
9. Report of the Committee for the Assessment of Biometric Aspects of Controlled
Trials of Hypoglycemic Agents. JAMA1975;231:583-608.
Engler RL, Yellon DM. Sulfonylurea KATP blockade in type II diabetes and precon-
ditioning in cardiovascular disease: time for reconsideration. Circulation 1996;94:
11.Riveline JP, Danchin N, Ledru F, et al. Sulfonylureas and cardiovascular effects:
from experimental data to clinical use. Available data in humans and clinical appli-
cations. Diabetes Metab2003;29:207-22.
12.Meier JJ, Gallwitz B, Schmidt WE, et al. Is impairment of ischaemic precondition-
ing by sulfonylurea drugs clinically important? Heart2004;90:9-12.
Schwartz TB, Meinert CL. The UGDP controversy: thirty-four years of contentious
ambiguity laid to rest. Perspect Biol Med2004;47:564-74.
14.UK Prospective Diabetes Study (UKPDS) Group. Intensive blood-glucose control
with sulphonylureas or insulin compared with conventional treatment and risk of
complications in patients with type 2 diabetes (UKPDS 33). Lancet1998;352:837-53.
Johnson JA, Majumdar SR, Simpson SH, et al. Decreased mortality associated with
the use of metformin compared with sulfonylurea monotherapy in type 2 diabetes.
Johnson JA, Simpson SH, Toth EL, et al. Reduced cardiovascular morbidity and
mortality associated with metformin use in subjects with type 2 diabetes. Diabet
17.Sasali A, Leahy JL. Is metformin cardioprotective? Diabetes Care2003;26:243-4.
18. Bell DSH. Secretagogues and cardiac risk. Endocrinologist2004;14:33-7.
19.Downey W, Beck P, McNutt M, et al. Health databases in Saskatchewan. In: Strom
B, editor. Pharmacoepidemiology. Chinchester (UK): Wiley; 2000. p. 325-45.
20. UK Prospective Diabetes Study Group. Effect of intensive blood-glucose control
with metformin on complications in overweight patients with type 2 diabetes
(UKPDS 34). Lancet1998;352:854-65.
21.McGuire DK, Newby LK, Bhapkar MV, et al. Association of diabetes mellitus and
glycemic control strategies with clinical outcomes after acute coronary syndromes.
Am Heart J2004;147:246-52.
22. Steiner JF, Prochazka AV. The assessment of refill compliance using pharmacy
records: methods, validity, and applications. J Clin Epidemiol1997;50:105-16.
23. Haynes RB, Gibson ES, Taylor DW, et al. Process versus outcome in hypertension:
a positive result. Circulation1982;65:28-33.
24. Luscher TF, Vetter H, Siegenthaler W, et al. Compliance in hypertension: facts and
concepts. J Hypertens Suppl1985;3:S3-9.
25. Tsuyuki RT, Olson KL, Teo KK. Nitroglycerin as an indicator for the presence of
coronary heart disease. Can Pharm J1999;132:32-6.
26. Cannon PJ, Connell PA, Stockley IH, et al. Prevalence of angina as assessed by a
survey of prescriptions for nitrates. Lancet1988;1:979-81.
27. Von Korff M, Wagner EH, Saunders K. A chronic disease score from automated
pharmacy data. J Clin Epidemiol1992;45:197-203.
28. Sackett DL, Shannon HS, Browman GW. Fenoterol and fatal asthma. Lancet 1990;
29. Rosenbaum P, Rubin D. Reducing bias in observational studies using subclassifi-
cation on the propensity score. J Am Stat Assoc1984;79:516-24.
30. Hill AB. The environment and disease: association or causation? Proc R Soc Med
Lin WH, Wright JM. Risk of sulfonylurea drugs is underappreciated. J Am Coll Car-
32. Boyd AE III, Aguilar-Bryan L, Nelson DA. Molecular mechanisms of action of gly-
buride on the beta cell. Am J Med1990;89(2A):3S-10S.
33. Leibowitz G, Cerasi E. Sulphonylurea treatment of NIDDM patients with cardio-
vascular disease: a mixed blessing? Diabetologia1996;39:503-14.
• January 17, 2006 • 174(2) | 174
Correspondence to: Dr. Jeffrey A. Johnson, 10405 Jasper Ave.,
Rm. 1200, Edmonton AB T5J 3N4; fax 780 448-0018;
• Existing evidence suggests that in patients with type 2
diabetes, the risk of death is higher among those who
are taking sulfonylurea monotherapy than among
those taking metformin.
• In this study, the authors used administrative data to
confirm previous findings: they found that rates of
mortality per 1000 patient-years were 67.6 among
users of first-generation sulfonylureas, compared with
39.6 among users of meformin. Additionally, the
authors have demonstrated that the risk of death was
higher in patients taking high doses of sulfonylureas
than taking low doses, suggesting a causal link.
• As is typical in studies that use administrative data-
bases, detailed clinical data were unavailable for
analysis. It is therefore possible that important con-
founding variables were unaccounted for.
Implications for practice:The results of this study should
add to existing caution about prescribing sulfonylurea
monotherapy for patients with type 2 diabetes. Clinicians
should weigh the need for sulfonylurea drugs in light of
their possible risks, given the availability of other oral