Back

Expert panel identified health-related social needs and methodological considerations for a polysocial risk score

Vest, J. R.; McNamee, C.; Musey, P. I.

2023-10-17 health systems and quality improvement
10.1101/2023.10.17.23297142 medRxiv
Show abstract

ObjectivesA polysocial risk score, which summaries multiple different health-related social needs (HRSNs) into a single likelihood of risk, could support more effective patient and population health management. Nevertheless, developing a polysocial risk score faces uncertainties and challenges due the HRSNs differing etiologies and interventions, cooccurrence, and variation in information availability. MethodsA 3-round Delphi technique to elicit expert opinion and develop a preliminary polysocial risk score approach. Expert panel members included physicians (n=8), social service professionals & staff (n=9), and patients (n=6). For physicians, the primary qualification was direct patient care experience in screening or asking about patients health-related social needs. Social service professionals & staff had titles of: nurse, patient care assistant, patient advocate, community health worker, director of community services. Round 1 obtained an initial importance of HRSNs on general health & well-being and total healthcare cost. Panelists also suggested additional HRSNs Responses served as discussion points for Round 2. Five focus groups explored how HRSNs should be ranked; additional HRSNs to include; timing of measurements; management of non-response and missing data; and concerns about bias and equity. We analyzed the transcripts using a consensus coding approach. Panelists then completed a follow-up survey. ResultsPanelists identified 17 HRSNs relevant to health and well-being for inclusion in a polysocial risk score. Methodology concerns ranging from the sources and quality of data, non-random missing information, data timeliness, and the need for different risk scores by population. Panelist also raised concerns about potential bias and missaplication of a polysocial risk score. ConclusionA polysocial risk score is a potentially useful addition to the growing methodologies to better understand and address HRSNs. Nevertheless, development is potentially complicated and fraught with challenges.

Matching journals

The top 7 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.