Subtyping Social Determinants of Health inAll of Us: Network Analysis and Visualization Approach

Abstract

Project Background. Social determinants of health (SDoH), such as unstable employment during the pandemic, account for between 30-55% of people's health outcomes. While many studies have identified strong associations among specific SDoH and health outcomes, most people experience multiple SDoH in their daily lives. Analysis of this complexity requires the integration of personal, clinical, social, and environmental information from a large cohort of underrepresented populations, which is only recently being made available through the All of Us research program. However, little is known about the range and response of SDoH in All of Us, and how they co-occur to form subtypes, which are critical for designing precision medicine interventions. Research Questions. (1) What is the range and response to survey questions related to SDoH? (2) How do SDoH co-occur to form subtypes, and what are their risk for adverse health outcomes? Methods. For Question-1, we characterized the range of SDoH questions across the surveys, and analyzed their responses. For Question-2, we used the following steps: (1) due to the missingness across the surveys, selected all participants with valid and complete SDoH data, and used inverse probability weighting to adjust their imbalance in demographics, compared to the full cohort; (2) asked domain experts to map the SDoH questions to SDoH subdomains, for enabling a more consistent granularity; (3) used bipartite modularity maximization to identify SDoH biclusters, their significance, and their replicability; (4) measured the association of each bicluster with 3 outcomes (depression, delayed medical care, emergency room visits in the last year) using multiple data types (surveys, electronic health records, and zip codes mapped to Medicaid expansion states), and (5) asked 3 domain experts to infer the subtype labels, their mechanisms, and potential targeted interventions. Results. For Question-1, we identified 110 SDoH questions across 4 surveys, categorized into 18 SDoH subdomains covering all 5 domains in Healthy People 2030 (HP-30). However, there was a large degree of missingness in survey responses (1.76%-84.56%), with later surveys having significantly fewer responses compared to earlier ones, and significant differences in race, ethnicity, and age of participants when compared to the full cohort. For Question-2, the subtype analysis (n=12,913, d=18) identified 4 biclusters with significant biclusteredness (Q=0.13, random-Q=0.11, z=7.5, P<0.001), and significant replication (Real-RI=0.88, Random-RI=0.62, P<.001). Furthermore, there were significant associations of specific subtypes with the outcomes and with Medicaid expansion, each with meaningful interpretations and potential precision interventions. For example, the subtype Socioeconomic Barriers included the SDoH subdomains employment, food security, housing, income, literacy, and education attainment, and had a significantly higher odds ratio (OR=4.2, CI=3.5-5.1, P-corr<.001) for depression, when compared to the subtype sociocultural barriers. Individuals that match this subtype profile could be screened early for depression and referred to social services to address combinations of SDoH such as housing and income. Finally, the identified subtypes spanned one or more HP-30 domains, revealing the difference between the current knowledge-based SDoH domains, versus the data-driven subtypes, reflecting the complexity of how SDoH co-occur in the real world, and their potential use in designing interventions. Community Impact. While several SDoH models including the Dahlgren-Whitehead conceptual model have identified SDoH domains, they have emphasized that real-world SDoH span multiple domains with complex interactions and feedback loops. However, this phenomenon has been difficult to analyze given the lack of large cohorts with underrepresented populations characterized by a wide range of SDoH and datatypes. The results from analyzing SDoH using the All of Us cohort provided direct evidence for this real-world phenomenon by showing that data-driven SDoH subtypes span one or more of the SDoH domains defined by Healthy People 2030. This result provides testable hypotheses in future studies that SDoH models based on data-driven subtypes will be more accurate and interpretable for predicting adverse health outcomes, when compared to existing models that use the knowledge-driven domains. Furthermore, the characterization of the range and response to SDoH across the entire All of Us cohort using over one hundred SDoH, should enable researchers to use the approach for characterizing other cohorts for identifying and addressing missingness. Finally, our workbench which focuses on subtyping SDoH, provides generalizable and scalable machine learning methods that can be used to periodically rerun the analysis as the All of Us cohort continues to evolve.

The Role of Social Determinants of Health and Social Position in Mental Health: An Examination of the Moderating Effects of Race, Ethnicity, and Gender on Depression through the All of Us Dataset

“…We acknowledge that while we based our operationalization strategy based on a careful review of extant literature and SDoH screeners [18,[34][35][36][37] and the sources of the instruments from which AoU drew the questions, other choices might have been made. See Bhavnani and colleagues (2023) for another approach to subtyping SDoH subdomains from the measurements provided by AoU [38].…”

Joint Association of Polygenic Risk and Social Determinants of Health with Coronary Heart Disease in the United States

“…SDOH data in AoU cover the five SDOH domains defined by Healthy People 2030 (HP-30). These include 1) economic stability, 2) education access and quality, 3) healthcare access and quality, 4) neighborhood and built environment, and 5) social and community context (15)(16)(17).…”