Last update:
31-May-2023

The extensive use of logistic regression models not only in analytic epidemiology but also in randomized clinical trials often creates inflated estimates of the relative risk. Particularly, in cases where a binary outcome has a high incidence in the respective populations (>10%), the bias in assessing the relative risk may be very large. Meta-analysis studies have shown that about 40% of the relative risk estimates in prospective investigations through logarithmic models lead to extensive bias of the population parameters. The problem of the inflation of risk estimates also appears in cross-sectional studies with binary outcomes, where the parameter of interest is the prevalence ratio. As an alternative to the use of logistic regression models, in both longitudinal and cross-sectional studies, the modified Poisson regression model is proposed.

Key words: Etiological studies, Logistic regression, Poisson regression, Relative risk.