How can researchers better measure overlapping social inequalities? Evans et al. (2024) offer an accessible tutorial on I-MAIHDA, a multilevel approach designed to examine health, social, and economic disparities across intersecting groups.
Intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (I-MAIHDA) is an innovative approach for investigating inequalities across health, disease, psychosocial, and socioeconomic outcomes. Evans et al. explain that I-MAIHDA and related methods offer clear conceptual and practical advantages over standard single-level regression models. By helping researchers examine how multiple, interlocking systems of marginalization and oppression produce inequalities—and by overcoming the typical challenges of analyzing interactions—it serves as a valuable tool for social epidemiology, health psychology, precision medicine, public health, and environmental justice. The method allows researchers to estimate average differences between intersectional groups, evaluate interaction effects in detail, and separate individual variation into differences within groups and between groups.
Because of the lack of practical guidance on running and interpreting MAIHDA models across different studies, Evans et al. bring this knowledge together into a clear conceptual and practical guide for both continuous and binary outcomes. While they focus on I-MAIHDA, the tutorial also sheds light on related methods, such as multicategorical MAIHDA used in clinical research. To help researchers run their own models or understand the concepts, the authors offer step-by-step guidance alongside a practical example using simulated health data, providing all data and code so readers can replicate the analyses.
Learn more about this article here:
Evans, C. R., Leckie, G., Subramanian, S. V., Bell, A., & Merlo, J. (2024). A tutorial for conducting intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (MAIHDA). SSM – Population Health, 26, 101664. https://doi.org/10.1016/j.ssmph.2024.101664