Quantitative methods for descriptive intersectional analysis with binary health outcomes

Abstract

Intersectionality recognizes that in the context of sociohistorically shaped structural power relations, an individual's multiple social positions or identities (e.g., gender, ethnicity) can interact to affect health-related outcomes. Despite limited methodological guidance, intersectionality frameworks have increasingly been incorporated into epidemiological studies, both to describe health disparities and to examine their causes. This study aimed to advance methods in intersectional estimation of binary outcomes in descriptive health disparities research through evaluation of 7 potentially intersectional data analysis methods: cross-classification, regression with interactions, multilevel analysis of individual heterogeneity (MAIHDA), and decision trees (CART, CTree, CHAID, random forest). Accuracy of estimated intersection-specific outcome prevalence was evaluated across 192 intersections using simulated data scenarios. For comparison we included a non-intersectional main effects regression. We additionally assessed variable selection performance amongst decision trees. Example analyses using National Health and Nutrition Examination Study data illustrated differences in results between methods. At larger sample sizes, all methods except for CART performed better than non-intersectional main effects regression. In smaller samples, MAIHDA was the most accurate method but showed no advantage over main effects regression, while random forest, cross-classification, and saturated regression were the least accurate, and CTree and CHAID performed moderately well. CART performed poorly for estimation and variable selection. Sensitivity analyses examining the bias-variance tradeoff suggest MAIHDA as the preferred unbiased method for accurate estimation of high-dimensional intersections at smaller sample sizes. Larger sample sizes are more imperative for other methods. Results support the adoption of an intersectional approach to descriptive epidemiology.

Keywords: Biostatistics; CART, classification and regression tree; CHAID, chi-square automatic interaction detector; CTree, conditional inference trees; Epidemiological studies; Health equity; Intersectionality; MAD, mean absolute deviation; MAIHDA, multilevel analysis of individual heterogeneity and discriminatory accuracy; NHANES, National Health and Nutrition Examination Study; Research design; SD, standard deviation; U.S., United States; VIM, variable importance measure.

Fig. 1

A to 1.D. Boxplots of the mean absolute deviation (MAD) of intersection estimations for four different sample sizes (graph excludes outliers) 1.A. Common outcome with categorical inputs 1.B. Rare outcome with categorical inputs 1.C. Common outcome with mixed inputs 1.D. Rare outcome with mixed inputs. Abbreviations: CART = classification and regression tree; CHAID = chi-square automatic interaction detector; CTree = conditional inference trees; MAIHDA = multilevel analysis of individual heterogeneity and discriminatory accuracy.

Fig. 1

A to 1.D. Boxplots of the mean absolute deviation (MAD) of intersection estimations for four different sample sizes (graph excludes outliers) 1.A. Common outcome with categorical inputs 1.B. Rare outcome with categorical inputs 1.C. Common outcome with mixed inputs 1.D. Rare outcome with mixed inputs. Abbreviations: CART = classification and regression tree; CHAID = chi-square automatic interaction detector; CTree = conditional inference trees; MAIHDA = multilevel analysis of individual heterogeneity and discriminatory accuracy.

Fig. 2

A to 2.C. Prevalence of high blood pressure by intersection. Abbreviations: CART = classification and regression tree; CHAID = chi-square automatic interaction detector; CTree = conditional inference trees; MAIHDA = multilevel analysis of individual heterogeneity and discriminatory accuracy.

Fig. 2

A to 2.C. Prevalence of high blood pressure by intersection. Abbreviations: CART = classification and regression tree; CHAID = chi-square automatic interaction detector; CTree = conditional inference trees; MAIHDA = multilevel analysis of individual heterogeneity and discriminatory accuracy.

Figure 3

A to 3.D. Boxplots of the MAD of intersection-specific estimations for two different small sample sizes, and a simulated outcome prevalence of 50% (graph excludes outliers) A. Categorical inputs B. categorical inputs with larger effect sizes only for interaction effects C. Mixed inputs D. Mixed inputs with larger effect sizes only for the interaction effects. Abbreviations: CART = classification and regression tree; CHAID = chi-square automatic interaction detector; CTree = conditional inference trees; MAIHDA = multilevel analysis of individual heterogeneity and discriminatory accuracy.

Figure 3

A to 3.D. Boxplots of the MAD of intersection-specific estimations for two different small sample sizes, and a simulated outcome prevalence of 50% (graph excludes outliers) A. Categorical inputs B. categorical inputs with larger effect sizes only for interaction effects C. Mixed inputs D. Mixed inputs with larger effect sizes only for the interaction effects. Abbreviations: CART = classification and regression tree; CHAID = chi-square automatic interaction detector; CTree = conditional inference trees; MAIHDA = multilevel analysis of individual heterogeneity and discriminatory accuracy.