Logistic Regression Analysis: When the Odds Ratio Does Not Work: An Example Using Intimate Partner Violence Data
Document Type
Article
Publication Date
10-2000
Publication Title
Journal of Interpersonal Violence
Abstract
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's 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 (<10%). 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.
Repository Citation
Mcnutt, L., Holcomb, J. P., Jr., , & Carlson, B. E. (2000). Logistic Regression Analysis: When the Odds Ratio Does Not Work: An Example Using Intimate Partner Violence Data. Journal Of Interpersonal Violence, 15(10), 1050-1059.
Original Citation
Mcnutt, L., Holcomb, J. P., Jr., , & Carlson, B. E. (2000). Logistic Regression Analysis: When the Odds Ratio Does Not Work: An Example Using Intimate Partner Violence Data. Journal Of Interpersonal Violence, 15(10), 1050-1059.
DOI
10.1177/088626000015010003
Volume
15
Issue
10
Comments
This study was supported by the Preventive Health Services Block Grant, New York State Department of Health.