Prevalence odds ratio versus prevalence ratio: choice comes with consequences

Area of Science:

• Epidemiology
• Biostatistics

Background:

• Measures of association like odds ratio, risk ratio, and prevalence ratio are crucial for quantifying relationships between variables in research.
• The selection of an appropriate measure often depends on the study design, leading to ongoing debate in the scientific community.

Purpose of the Study:

• To illustrate the analytic implications of selecting specific outcome categories and reference groups for independent variables in statistical modeling.
• To demonstrate the practical application of choosing prevalence ratio over odds ratio in a cross-sectional study setting.

Main Methods:

• A cross-sectional study design was utilized for the demonstration.
• The study compared the results of using prevalence ratio versus odds ratio.
• Analysis focused on the impact of selecting different outcome categories and reference levels for independent variables.

Main Results:

• The choice of statistical measure (prevalence ratio vs. odds ratio) significantly affects the interpretation of the association between independent and dependent variables.
• Altering the modeled outcome category or the reference group for independent variables can alter statistical significance.

Conclusions:

• The selection of the outcome category and reference level for independent variables has demonstrable analytic implications.
• Researchers should carefully consider these choices to ensure accurate and robust findings, particularly in cross-sectional studies where prevalence ratio may be more appropriate.