To use or not to use the odds ratio in epidemiologic analyses?
 Markku Nurminen
 … show all 1 hide
Rent the article at a discount
Rent now* Final gross prices may vary according to local VAT.
Get AccessAbstract
This paper argues that the use of the odds ratio parameter in epidemiology needs to be considered with a view to the specific study design and the types of exposure and disease data at hand. Frequently, the odds ratio measure is being used instead of the risk ratio or the incidenceproportion ratio in cohort studies or as an estimate for the incidencedensity ratio in casereferent studies. Therefore, the analyses of epidemiologic data have produced biased estimates and the presentation of results has been misleading. However, the odds ratio can be relinquished as an effect measure for these study designs; and, the application of the casebase sampling approach permits the incidence ratio and difference measures to be estimated without any untenable assumptions. For the Poisson regression, the odds ratio is not a parameter of interest; only the risk or rate ratio and difference are relevant. For the conditional logistic regression in matched casereferent studies, the odds ratio remains useful, but only when it is interpreted as an estimate of the incidencedensity ratio. Thus the odds ratio should, in general, give way to the incidence ratio and difference as the measures of choice for exposure effect in epidemiology.
 Pearce, N (1993) What does the odds ratio estimate in a casecontrol study?. Int J Epidemiol 22: pp. 11891192
 Lee, J (1994) Odds ratio for crosssectional data?. Int J Epidemiol 23: pp. 201203
 Miettinen, OS (1985) Theoretical epidemiology: Principles of occurrence research in medicine. Delmar Publishers, Albany, NY
 Breslow, NE, Day, NE (1980) Statistical methods in cancer research, Vol. 1: The design and analysis of casecontrol studies. International Agency for Research on Cancer, Lyon
 Breslow, NE, Day, NE (1987) Statistical methods in cancer research, Vol 2: The design and analysis of cohort studies. International Agency for Research on Cancer, Lyon
 Gail, MH (1991) A bibliography and comments on the use of statistical models in epidemiology in the 1980s. Stat Med 10: pp. 18191885
 Sinclair, JC, Bracken, MB (1994) Clinically useful measures of effect in binary analyses of randomised trials. J. Clin Epidemiol 47: pp. 881889
 Greenland, S (1987) Interpretation and choice of effect measures in epidemiologic analyses. Am J Epidemiol 125: pp. 761788
 Miettinen, OS, Cook, EF (1981) Confounding: Essence and detection. Am J Epidemiol 114: pp. 593603
 Greenland, S, Robins, J (1986) Identifiability, exchangeability and epidemiologic confounding. Int J Epidemiol 15: pp. 413419
 Gail, MH, Wieand, S, Piandatosi, S (1984) Biased estimates of treatment effect in randomised experiments with nonlinear regressions and omitted covariables. Biometrika 71: pp. 431344
 Savitz, DA (1992) Measurements, estimates, and inferences in reporting epidemiologic study results [editorial]. Am J Epidemiol 135: pp. 223224
 Axelson O, Fredrikson M, Ekberg K. A comment on the implications of using odds ratios or prevalence ratios in crosssectional studies. In: Hemon D (ed), Book of abstracts of the 8th International Symposium Epidemiology in Occupational Health, 10–12 September 1991, Paris, 1991: 23.
 Axelson, O, Fredrikson, M, Ekberg, K (1994) Use of the prevalence ratiov the prevalence odds ratio as a measure of risk in cross sectional studies. Occup and Environ Med 51: pp. 574
 Leino T, Tammilehto L, Luukkonen R, Kanerva L, Nordman H. Respiratory and skin symptoms in hairdressers. In: Program résumé of the 43rd Nordic Work Environment Meeting, 28–30 August 1994, Loen, Norway, 1994: 87.
 ViikariJuntura, E, Riihimäki, H, Tola, S, Videman, T, Mutanen, P (1994) Neck trouble in machine operating, dynamic physical work and sedentary work: A prospective study on occupational and individual risk factors. J Clin Epidemiol 47: pp. 14111422
 Greenland, S (1994) Modelling risk ratios from matched cohort data: An estimating equation approach. Appl Statist 43: pp. 223232
 Greenland, S, Thomas, DC (1982) On the need for the rare disease assumption in casecontrol studies. Am J Epidemiol 116: pp. 547553
 Wacholder, S (1986) Binomial regression in GLIM: Estimating risk ratios and risk differences. Am J Epidemiol 123: pp. 174184
 Greenland, S (1979) Limitations of the logistic analysis of epidemiologic data. Am J Epidemiol 110: pp. 693698
 Manton, KG, Stallard, E (1988) Chronic disease modelling: Measurement and evaluation of the risks of chronic disease processes. Charles Griffin & Company Ltd, London
 Berry, G, Mandryk, J, Mock, P Presentation of results of logistic regression: Analyses of cohort and crosssectional studies. In: Francis, IS, Manly, BFJ, Lam, FC eds. (1986) Pacific Statistical Congress. Elsevier Science Publishers (NorthHolland), Amsterdam, pp. 7982
 Cornfield, J (1951) A method of estimating comparative rates from clinical data: Applications to cancer of the lung, breast and cervix. J Natl Cancer Inst 11: pp. 12691275
 Greenland, S, Thomas, DC, Morgenstern, H (1986) The raredisease assumption revisited: A critique of ‘estimators of relative risk for casecontrol studies’. Am J Epidemiol 124: pp. 869876
 Kupper, LL, McMichael, AJ, Spritas, R (1975) A hybrid epidemiologic study design useful in estimating relative risk. Am Stat Assoc 70: pp. 524528
 Miettinen, OS (1976) Estimability and estimation incasereferent studies. Am J Epidemiol 103: pp. 226235
 Miettinen, O (1982) Design options in epidemiologic research: an update. Scand J Work Environ Health 8: pp. 714
 Miettinen, O, Nurminen, M (1985) Comparative analysis of two rates. Stat Med 4: pp. 213226
 Greenland, S (1986) Adjustment of risk ratios in casebase studies (hybrid epidemiologic designs). Stat Med 5: pp. 579584
 Miettinen, OS, Caro, JJ (1989) Principles of nonexperimental assessment of excess risk, with special reference to adverse drug reactions. J Clin Epidemiol 42: pp. 325331
 Nurminen, M (1989) Analysis of epidemiologic casebase studies for binary data. Stat Med 8: pp. 12411254
 Greenland, S (1982) Interpretation and estimation of summary ratios under heterogeneity. Stat Med 1: pp. 217227
 Nurminen, M (1992) Assessment of excess risk in casebase studies. J Clin Epidemiol 45: pp. 10811092
 Wacholder, S (1991) Practical considerations in choosing between the casecohort and nested casecontrol designs. Epidemiology 2: pp. 155158
 Wacholder, S, Boivin, JF (1987) External comparisons with the casecohort design. Am J Epidemiol 126: pp. 12081209
 Rothman, KJ (1986) Modern epidemiology. Little, Brown and Company, Boston, MA
 Prentice, R (1976) Use of the logistic model in retrospective studies. Biometrics 32: pp. 599606
 Prentice, RL (1986) A casecohort design for epidemiologic cohort studies and disease prevention trials. Biometrika 73: pp. 111
 Schouten, EG, Decker, JM, Kok, FJ, Cessie, S, Houwelingen, HC, Pool, J, Wandenbroucke, JP (1993) Risk ratio and rate ratio estimation in casecohort designs: Hypertension and cardiovascular mortality. Stat Med 12: pp. 17331745
 Title
 To use or not to use the odds ratio in epidemiologic analyses?
 Journal

European Journal of Epidemiology
Volume 11, Issue 4 , pp 365371
 Cover Date
 19950801
 DOI
 10.1007/BF01721219
 Print ISSN
 03932990
 Online ISSN
 15737284
 Publisher
 Kluwer Academic Publishers
 Additional Links
 Topics
 Keywords

 Biometry
 Epidemiologic methods
 Odds ratio
 Risk difference
 Risk ratio
 Industry Sectors
 Authors

 Markku Nurminen ^{(1)}
 Author Affiliations

 1. Department of Epidemiology and Biostatistics, Finnish Institute of Occupational Health, and Department of Public Health, University of Helsinki, Helsinki, Finland