# Socio-geographical disparities in cardiometabolic multimorbidity in Sweden: an Intersectional Multilevel Analysis of Individual Heterogeneity and Discriminatory Accuracy (I-MAIHDA)

## TL;DR

This study maps health disparities in heart and metabolic diseases among middle-aged adults in Sweden, showing how factors like age, education, and location intersect to affect risk.

## Contribution

The study introduces an intersectional multilevel analysis to map cardiometabolic multimorbidity disparities in Sweden.

## Key findings

- 2.8% of participants had cardiometabolic multimorbidity, with the highest prevalence in specific socio-geographical strata.
- 60–64-year-old males with low education had the highest prevalence, regardless of country of birth.
- The VPC and AUC values highlight the importance of intersectional strata in explaining health disparities.

## Abstract

Mapping social and geographical disparities in health outcomes helps to identify vulnerable groups that should be targeted for intervention. This study aims to assess the disparities in cardiometabolic multimorbidity, the presence of at least two cardiometabolic diseases (CMD), including type 2 diabetes, heart disease, and stroke, among middle-aged adults across socio-geographical intersectional strata in Sweden.

This cross-sectional study used the first examination (2013–2018) of the Swedish CArdioPulmonary bioImage Study (SCAPIS), with a total sample of 29,093 individuals aged 50–64 years living in six areas in Sweden (Gothenburg, Linköping, Malmö/Lund, Stockholm, Umeå and Uppsala). Cardiometabolic multimorbidity was identified based on self-reported information, the National Patient Register, the Swedish Prescribed Drug Register, and examination of glycaemic status. We constructed ninety-six socio-geographical intersectional strata based on the combination of age (50–59/60–64 years), sex (females/males), education (low/high), country of birth (Swedish/foreign-born), and six geographical areas. Intersectional multilevel analysis of individual heterogeneity and discriminatory accuracy (I-MAIHDA) was used to map the predicted prevalence of cardiometabolic multimorbidity and quantify the variance partition coefficient (VPC) and area under the receiver operating characteristic curve (AUC).

24.4% of the participants had one CMD, and 2.8% had cardiometabolic multimorbidity. Across the six areas, strata of 60–64-year-old males with low education, irrespective of country of birth, had the highest prevalence of cardiometabolic multimorbidity. The highest prevalence was observed in 60–64-year-old foreign-born males with low education in Gothenburg (12.4%, 95% CI 7.1–19.3) and in 60–64-year-old Swedish-born males with low education in Malmö (8.6%, 95% CI 6.3–11.3). The VPC was high (15.0%, 95% CI 10.5–21.1), indicating the importance of intersectional strata in explaining disparities in cardiometabolic multimorbidity, with an AUC of 0.71 (95% CI 0.70–0.73).

By applying an intersectionality framework, our study provides a more nuanced map of disparities in cardiometabolic multimorbidity to inform preventive strategies aligned with proportionate universalism and precision of public health. The heterogeneity and discriminatory accuracy measures suggest the need for universally tailored intervention strategies to prevent cardiometabolic multimorbidity.

The online version contains supplementary material available at 10.1186/s12939-025-02684-z.