Development of social risk screening tool for primary care: Systematic review and Delphi study

Problem

Social determinants of health (SDOH) critically influence population and individual health outcomes. Despite this, information on individual-level social risks are not collected routinely in primary care. General Practice is often the primary contact point for individuals facing social issues. Enhancing information on individual social risks facilitates improved personalised care, targeted support for those with unmet social need, and better information for resource planning and policy decisions. Existing social risk screening tools are primarily designed for North American audiences. Our aim was to develop a screening tool through a systematic review of existing tools/questions and a Delphi study to identify acceptable questions for use in a UK setting.

Approach

1) Systematic reviewWe searched English language literature using Medline, CINAHL, Embase, Web of Science, Social science and practice databases from 2002-2022 for questions/tools that collected information on people’s social risk from the following pre-specified domains: finance, housing condition, homelessness, food security, transport, utilities, education, neighbourhood safety, social connectedness, childcare and employment. We excluded any questions/tools that were not relevant to UK settings.Each publication was reviewed to assess whether the 8 Gold Standard Steps of Measure Development (GSSoMD) were reported which included: generation of initial questions using experts, pilot testing, validity and reliability testing and reporting of psychometric properties. Questions/tools that were used for screening and met at least one step from the 8 GSSoMD proceeded to the Delphi study.2) The Delphi study consisted of 3 rounds with stakeholders (PPIE members, GPs, public health, social workers, allied health professionals and third sector workers). In each survey round, participants were presented with questions in each of the domains and were asked to rate or rank which questions they felt were most acceptable to ask in each domain. Each round led to item reduction and after the final round one question per domain remained.PPIEKeele PPIE members reviewed research questions, study design and survey instruments.

Findings

1) Systematic review106 studies were included after full text review, the majority from USA (n=84). Tools/questions were predominantly from hospital settings (n=37) and primary care (n=26). Screening was usually in written format (n=35), face-to-face (n=23) or electronic (n=19) and self-complete (n=44).2) Results from the Delphi study are currently being analysed. In total, 27 participants completed Survey 1, the majority were aged 45-54 years (n=9), female (n=20) and white ethnicity (n=15).

Consequences

Our 11-item, yes/no answer, evidence-based, social risk screening tool has been developed for use in primary care. It gathers comprehensive information on people’s social circumstances. Where a need is identified people can be signposted to services or referred to social prescribers. Further research is needed to test reliability and validity of the tool, understand the extent of unmet social need and resource implications of screening.