Toward Successful and Sustainable Statewide Screening for Social Determinants of Health: Testing the Interest of Hospitals

Abstract

The impact of social and behavioral factors on health outcomes are well defined in the field of public health. Additionally, characteristics such as race, ethnicity, and language have been proven to affect an individual's capacity to address health care needs. While these nonclinical components affect care, variations in screening methodology between organizations make it difficult to analyze data broadly. Standardized approaches can mitigate the impact of these factors but may be difficult to incorporate into an established workflow. The Connecticut Hospital Association identified social determinants of health (SDOH) as a factor affecting patient outcomes during a statewide collaborative on asthma. The goal of this quality improvement project was to explore change in workflow as a barrier to screening for SDOH in hospitals. Four hospitals participated in the pilot using a standardized screening tool to assess 662 patients; 62% (n = 410) were White, 11% (n = 76) were Black, 5% (n = 31) were classified as other, and 22% (n = 145) were in unknown race categories. Of those reporting needs, 438 (66%) had food-, housing-, or transportation-related needs. Qualitative interviews with staff from pilot hospitals were conducted. There were 3 main themes: the screening tool was easy to use; patients could be reluctant to reveal SDOH information; and lack of a standardized referral process made patient screening difficult to sustain or justify. The volume and magnitude of SDOH needs identified, along with the sense of helplessness expressed in qualitative interviews, reinforced the decision to implement a technology platform for screening, closed-loop referral, and outcome measurement.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 19 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.