Social determinant of health patterns and mortality outcomes in US adults

TL;DR

This study identifies five patterns of social determinants of health in US adults and finds that being unmarried and unemployed or facing multiple social challenges is linked to higher mortality risks.

Contribution

The study introduces a novel approach to clustering multiple social determinants of health into distinct patterns and evaluates their joint effects on mortality outcomes.

Findings

Five distinct SDOH patterns were identified in US adults aged 18–79.

Patterns 4 and 5 showed worse mortality outcomes, including all-cause, cancer-specific, and cardiovascular disease-specific mortality.

Comprehensive screening of SDOH profiles is recommended to understand cumulative health impacts.

Abstract

Social Determinants of Health (SDOH) influences healthcare access, especially in patients with chronic diseases. However, SDOHs were often investigated as single variables. Combination patterns and joint effects of multiple SDOHs are much understudied. This study seeks to identify SDOH patterns in the US general population and their influence in all-cause and specific mortality.

This study included US adults aged 18 to 79 from 2002 to 2018 National Health Interview Survey (NHIS) and NHIS Linked Mortality Files. 12 SDOHs from 5 domains (healthcare access, education and literacy, economic stability, social isolation, neighborhood cohesion) were selected and binarized from the NHIS, including: material, psychological, and behavioral medical financial hardship, delayed care due to transportation and due to non-transportation factors, education, employment, food security, income, housing security, marital status, and neighborhood cohesion. Key outcomes, including all-cause, cancer-specific mortality, and cardiovascular disease-specific death were identified at quarter and year of death.

From the 105,824 younger adults (18–64 years), and the 23,825 older adults (65–79 years), five distinct SDOH patterns were identified: pattern 1 (31%, few barriers); 2 (20%, unmarried); 3 (17%, unemployed); 4 (15%, both unmarried and unemployed); and 5 (16%, with relatively high rate of non-married status, housing insecurity, and material, psychological, or behavioral medical financial hardships). Compared to pattern 1, pattern 4 and 5 had worse prognosis in all mortality outcomes in both age groups, including all-cause mortality, cancer-specific mortality, and cardiovascular disease-specific mortality in both age groups.

In this study, we found that SDOHs could be clustered into five distinct patterns. Patients who were unmarried and unemployed (pattern 4) or with multiple concurrent adverse SDOHs (pattern 5) had poorest key health outcomes. These findings support comprehensive screening for SDOH profiles to understand cumulative influences of SDOHs on quality of life and clinical outcomes of patients with chronic medical conditions.

The online version contains supplementary material available at 10.1186/s12889-025-24126-9.

Figures