Abstract
Objectives: The case-control study is commonly used to examine adverse drug events, in which prevalence of exposure in the source population is frequently very low. The objective of the current study was to examine the bias inherent in the odds ratio assessing the association between exposure and an adverse outcome when prevalence of exposure in the source population is extremely low.
Design: Monte Carlo simulations examined the effect of sample size, exposure prevalence, and magnitude of the underlying odds ratio on the bias of the estimated risk ratio, and the power to detect a non-zero risk ratio.
Results: Once the underlying odds ratio was at least four, the adverse effects of low prevalence of exposure was minimal. Studies with small sample sizes and low prevalence of exposure, coupled with small to moderate effect sizes, can result in biased estimates of association between exposure and disease status. With a sample size of 200 and an exposure prevalence of 0.5% in the control population, the bias in the estimated odds ratio can be as large as 115%. However, bias becomes negligible as sample size becomes large (n ≥ 2000), even when prevalence of exposure is very low. Once the expected number of exposed controls is at least eight, the bias in the estimated odds ratio was no more than 5%.
Conclusions: Studies with small sample sizes and low prevalence of exposure, coupled with small to moderate effect sizes can result in biased estimates of association between exposure status and adverse drug effects. However, bias becomes negligible as sample size becomes large.
Similar content being viewed by others
References
Rothman KJ, Greenland S. Modern epidemiology. New York: Lippincott Williams & Wilkins, 1998
Breslow NE, Day NE. Statistical methods in cancer research. vol 1: the analysis of case-control studies. Lyon: IARC, 1980
Roujeau JC, Kelly JP, Naldi L, et al. Medication use and the risk of Stevens-Johnson syndrome or toxic epidermal necrolysis. N Engl J Med 1995; 333: 1600–7
Kernan WN, Viscoli CM, Brass LM, et al. Phenylpropanolamine and the risk of hemorrhagic stroke. N Engl J Med 2000; 343: 1826–32
Abenhaim L, Moride Y, Brenot F, et al. Appetite-suppressant drugs and the risk of primary pulmonary hypertension. N Engl J Med 1996; 335: 609–16
Zornberg GL, Jick H. Antipsychotic drug use and the risk of first-time idiopathic venous thromboembolism: a case-control study. Lancet 2000; 356: 1219–23
International Agranulocytosis and Aplastic Anaemia Study. Risk of agranulocytosis and aplastic anaemia in relation to use of antithyroid drugs. BMJ 1988; 297: 262–5
Rawson NSB, Rutledge Harding S, Malcolm E, et al. Hospitalizations for aplastic anemia and agranulocytosis in Saskatchewan: incidence and associations with antecedent prescription drug use. J Clin Epidemiol 1998; 51: 1343–55
Kelly JP, Kaufman DW, Shapiro S. Risks of agranulocytosis and aplastic anemia in relation to the use of cardiovascular drugs: the International Agranulocytosis and Aplastic Anemia Study. Clin Pharmacol Ther 1991; 49: 330–41
Pastuszak AL, Schuler L, Speck-Martins CE, et al. Use of misoprostol during pregnancy and mobius’ syndrome in infants. N Engl J Med 1998; 338: 1881–5
Jick H. The discovery of drug-induced illness. N Engl J Med 1977; 296: 481–5
de Abajo FJ, Rodriguez LAG, Montero D. Association between selective serotonin reuptake inhibitors and upper gastrointestinal bleeding: population-based case-control study. BMJ 1999; 319: 1106–9
Peduzzi P, Concato J, Kemper E, et al. A simulation study of the number of events per variable in logistic regression analysis. J Clin Epidemiol 1996; 49: 1373–9
Peduzzi P, Concato J, Feinstein AR, et al. Importance of events per independent variable in proportional hazards regression analysis II: accuracy and precision of regression estimates. J Clin Epidemiol 1995; 48: 1503–10
Becker S. A comparison of maximum likelihood and jewell’s estimators of the odds ratio and relative risk in single 2x2 tables. Stat Med 1989; 8: 987–96
Walter SD, Cook RJ. A comparison of several point estimators of the odds ratio in a single 2 × 2 contingency table. Biometrics 1991; 47: 795–811
Haldane JBS. The estimation and significance of the logarithm of a ratio of frequencies. Ann Hum Genet 1956; 20: 309–11
Jewel NP. On the bias of commonly used measures of association for 2 × 2 tables. Biometrics 1986; 42: 351–8
Gart JJ, Zweifel JR. On the bias of various estimators of the logit and its variance with application to quantal bioassay. Biometrika 1967; 54: 181–7
Greenland S, Schwartzbaum JA, Finkle WD. Problems due to small samples and sparse data in conditional logistic regression analysis. Am J Epidemiol 2000; 151: 531–9
Breslow NE, Day NE. Statistical methods in cancer research, vol II: the analysis of cohort studies. Lyon: IARC, 1987
Mathsoft Inc. S-Plus, version 5.1. 1999
SAS Institute Inc. SAS, version 6.12. Cary (NC), 1999
Doll R, Hill AB. Smoking and carcinoma of the lung: preliminary report. BMJ 1950; II: 739–48
Psaty BM, Koepsell TD, LoGerfo JP, et al. Beta-blockers and primary prevention of coronary heart disease in patients with high blood pressure. JAMA 1989; 261: 2087–94
Hu S, Hertz-Picciotto I, Siemiatycki J. When to be skeptical of negative studies: pitfalls in evaluating occupational risks using population-based case-control studies. Can J Public Health 1999; 90: 138–42
Lee-Feldstein A. A comparison of several measures of exposure to arsenic: matched case-control study of copper smelter employees. Am J Epidemiol 1989; 129: 112–24
Jarup L, Pershagen G. Arsenic exposure, smoking, and lung cancer in smelter workers: a case-control study. Am J Epidemiol 1991; 134: 545–51
Czeizel AE, Rockenbauer M, Sorenson HT, et al. A population-based case-control teratologic study of furazidine, a nitrofuranderivative treatment during pregnancy. Clin Nephrol 2000; 53: 257–63
Thomas DG. Exact confidence intervals for the odds ratio in a 2x2 table. Appl Stat 1971; 10: 105–10
Stata Corporation. Stata 7.0. Texas: College Station, 2001
Acknowledgements
The study was funded by the Institute for Clinical Evaluative Sciences. The funding arrangement allowed the authors freedom to design, analyse, and publish the results.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Austin, P.C., Mamdani, M. & Williams, I.J. Adverse Effects of Observational Studies When Examining Adverse Outcomes of Drugs. Drug-Safety 25, 677–687 (2002). https://doi.org/10.2165/00002018-200225090-00006
Published:
Issue Date:
DOI: https://doi.org/10.2165/00002018-200225090-00006