Measuring Additive Interaction Using Odds Ratios
Author Information
Author(s): Kalilani Linda, Atashili Julius
Primary Institution: Department of Epidemiology, School of Public Health, University of North Carolina at Chapel Hill, USA
Hypothesis
How do measures of additive interaction differ when using odds ratios versus risk ratios?
Conclusion
Using odds ratios instead of risk ratios to calculate measures of additive interaction can lead to misleading conclusions.
Supporting Evidence
- The study found that the discrepancy between interaction measures increases with the baseline risk.
- AP was found to be the most robust measure when substituting odds ratios for risk ratios.
- The analysis showed that using odds ratios can lead to incorrect conclusions about the presence of interaction.
Takeaway
This study shows that using one type of measurement instead of another can give you the wrong idea about how two things work together.
Methodology
The study used simulations to compare the interaction measures calculated using odds ratios and risk ratios across various baseline risks.
Potential Biases
Potential bias arises from using odds ratios in place of risk ratios, especially when the outcome is not rare.
Limitations
The analysis assumes complete assessment of the population and does not account for confounding factors.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website