A.Y. Mo1, A. Mahajan2, P. Podugu2, P. Mallu1, Y. Hu1, S.A. Sweeney1,2, V.P. Ho1,2 1Case Western Reserve University School Of Medicine, Cleveland, OH, USA 2MetroHealth Medical Center, Surgery, Cleveland, OH, USA
Introduction: Current metrics such as the Social Vulnerability Index (SVI) and the Area Deprivation Index (ADI) are widely used to estimate social drivers of health (SDOH) for health services research. However, debate exists over which metric better represents social need, as inaccurate estimates may lead to resource misallocation. Our public safety-net institution implemented a self-reported SDOH screen in 2019. Our study aims to evaluate how well ADI and SVI reflect the reported social needs of trauma patients. We hypothesized that ADI and SVI would both correlate well with self-reported social needs from our SDOH screen.
Methods: We identified all trauma patients from 2022 who had completed a SDOH screen within a year of the trauma, either before or after. ADI and SVI scores were estimated for each patient using zip codes. Our SDOH screen assessed 11 determinants individually and produces a summative score. We used state-specific decile rankings for ADI and the SVI's sub-indices: Socioeconomic Status, Household Characteristics, Racial & Ethnic Minority Status, and Housing Type & Transportation. We created a Spearman correlation matrix for summative and individual components of SDOH, ADI, and SVI. We also averaged SDOH scores by zip code and correlated them with ADI and SVI using a Pearson correlation (r) matrix, excluding zip codes with fewer than five respondents.
Results: Our study included 608 patients. The self-reported SDOH screen’s individual components and summative scores did not correlate with ADI decile, SVI sub-indices, or the summative SVI score, with correlation values only reaching 0.21 (Figure). ADI and SVI correlated well with each other (r=0.85). The SDOH screen's individual measures also showed poor correlation with each other, with the highest being 0.64 for Food Insecurity and Financial Resource Strain. When examining median SDOH scores for each zip code, the institutional SDOH screen did not demonstrate a strong positive relationship with either ADI (r=0.42) or SVI (r=0.39).
Conclusion: In our cohort, ADI and SVI did not accurately reflect individuals’ self-reported social needs or reliably capture patients’ socioeconomic realities. Starting this year, hospitals are incentivized to perform SDOH screening by new federal reimbursement initiatives. Our study underscores the need to prioritize the use of self-reported SDOH data in future research for the development of practical interventions to address social needs and promote better health outcomes.