Measuring social needs among older adults: psychometrics and policy

Indiana University Indianapolis, Indianapolis IN

Investigators

Abstract

Health-related social needs (HRSNs), sometimes called social determinants, are drivers of morbidity, mortality, healthcare utilization, and costs. Per the Centers for Medicare & Medicaid Services (CMS) HRSN data will be used in quality reporting metrics, in referrals, in adjusting payments, for public reporting, and in value-based care. To better inform service delivery and policy, it is crucial to understand whether screening tools result in valid measurement of HRSNs. Problematically, current HRSN screening tools may be inaccurate and have unknown performance among older adults. Our long-term goal is to enable effective care delivery and optimal policy decisions that leverage information on the social needs that drive health. The objective of this proposal is to advance the science of HRSN measurement and assess the implications of current HRSN screening approaches for Medicare policy. Our central hypothesis is that current HRSN screening tools disadvantage older adults covered by Medicare and will undermine efforts to create effective patient care and population health. Drawing on strong preliminary data and extensive domain expertise, our multidisciplinary team proposes the following specific aims. 1) Compare the validity of health-related social needs screening tools for older adults will establish the sensitivity, specificity, and positive predictive values of 4 screening tools (CMS, Epic, PRAPARE, Health Leads) against reference (i.e. “gold”) standard measures in a sample of 4,000 Medicare beneficiaries. We hypothesize that HRSN screening tool performance among older adults will be worse than the accepted performance thresholds for clinical utility. 2) Compare the performance of health-related social needs screening tools in predicting healthcare quality and costs for older adults will test the performance of the 4 HRSN screening tools and reference standard measures in predicting subsequent quality and costs outcomes. We hypothesize that the reference standard measures will explain more of the variance in costs and quality outcomes than the HRSN screening tools. 3) Evaluate the content validity of health-related social need screening for older adults will use key informant interviews with older adults, their caregivers/family members, and their clinicians, we will assess the appropriateness and relevance of different HRSN screening tools. This project will advance practice from a “one-size fits all” screening policy to a scientifically justifiable approach that leverages the most effective screening tools and incorporates HRSNs most relevant to older adults. This project is significant as: 1) CMS policy prioritizes HRSN screening as a means to better population health; 2) HRSNs are critical to the well-being of older adults; and 3) unvalidated HRSN screening tools could lead to undesirable downstream effects on future operation and policy decisions. This proposal is innovative because it rigorously evaluates HRSN screening tool validity against reference standard measures and considers on the appropriateness of HRSN screening tools.