The essay by Merlo reflects on his 20+ years of work on MAIHDA, or multilevel analysis of individual heterogeneity and discriminatory accuracy.1 Starting with his original motivating question (how can neighbourhoods meaningfully shape health if they explain almost none of the variation between individuals?), Merlo traces the development of this paradox into a systematic framework for examining whether and to what extent contexts organise individual heterogeneity. Along the way, he clarifies key MAIHDA concepts, including the specific contextual effect (SCE), general contextual effect (GCE), and discriminatory accuracy, and emphasises that MAIHDA is agnostic to both social theory and statistical techniques. He ultimately calls for greater attention to organised heterogeneity within mainstream, mean-centric epidemiology.
Over the past decade, much has been written about MAIHDA and particularly its intersectional application, I-MAIHDA (intersectional-MAIHDA), originally advanced by Evans et al and then further refined by Merlo2 3; excellent discussions of…