Logistic Regression Analysis: When the Odds Ratio Does Not Work: An Example Using Intimate Partner Violence Data

The odds ratio is one of the most common measures used to assess the relationship between exposure to violence and adverse health outcomes, adjusting for possible confounding factors. A reason for the odds ratio\u27s popularity is that it is relatively easy to calculate from the coefficients of a logistic regression model. For most etiologic studies of disease, the odds ratio is a suitable estimate of risk because incidence or prevalence of disease is rare (\u3c10%). However, health outcomes studied in violence research are often more prevalent (e.g., fatigue, insomnia, stomach pain, and shortness of breath). In these cases, the odds ratio usually overestimates the strength of association, sometimes erroneously tripling the magnitude. Data from a study measuring the health effects of intimate partner violence are used to illustrate the problem of incorrectly using odds ratios. Methods to calculate relative risks and prevalence ratios from logistic regression models are presented