A wealth of emerging evidence on the associations between social determinants of health (SDH) (eg, food, housing, transportation, and education) and health outcomes1⇓⇓⇓⇓⇓–7 has fueled a wave of experimentation around identifying and addressing patients' SDH in the context of clinical care.8 The Centers for Medicare and Medicaid Services, the Centers for Disease Control and Prevention, and the National Academy of Medicine have recommended that high-quality primary care includes documentation of a core set of SDH measures, ideally in electronic health records (EHRs).9⇓⇓–12 Due in part to these recommendations, several health sector leaders have developed tools for identifying patients' SDH needs, using validated measures as available (Appendix).13,14 These social needs screening tools are now being used to inform clinical interventions, including providing social and economic resources on-site (eg, food boxes) or connecting patients with off-site community-based resources (eg, food banks). Collectively, SDH-focused interventions undertaken in medical settings have been referred to as “social prescribing.”15
In the United States, much of the experimentation around social prescribing takes place in community health centers16,17 where there is a historic precedent for addressing social and economic needs as a core part of primary care.18 The Health Resources and Services Administration Bureau of Primary Care requires that federally-qualified health centers provide some services under the umbrella of SDH, such as translation and transportation services that may help to address SDH-related barriers to care. These kinds of activities are expanding under new federal and state programs that leverage value-based payment models to incentivize more comprehensive, coordinated care, especially for high-risk beneficiaries. For instance, provision 2703 of the Affordable Care Act created an optional Medicaid State Plan benefit to support beneficiaries with chronic diseases19; the federal Comprehensive Primary Care+ (CPC+) demonstration project similarly includes a range of value-based payment incentives for improved care management and coordination.20 These models support primary care strategies that connect patients with nonclinical social services in addition to coordinating primary, acute, and behavioral health care services.
Despite growing interest in social prescribing, major evidence gaps persist in 2 key areas. First, although findings from some evaluations of SDH-related interventions suggest that specific programs can decrease social needs and improve health21⇓⇓–24, relatively little research addresses the impacts of social prescribing initiatives on patient and provider experience of care, health outcomes, health care costs, and utilization.25 Ideally, gaps in effectiveness research will be filled through federal demonstration project evaluations, including evaluations of Health Homes19, CPC+, and the newly launched Accountable Health Communities Program14, to the extent that social prescribing components can be distinguished from other care model components.25
As the effectiveness research grows in this field, a second major gap in research will become increasingly relevant to practitioners. This gap involves the evidence base on implementation strategies26 needed to put these interventions into practice and take them to scale in diverse settings. The rapid proliferation of social prescribing activities in the United States provides an important opportunity for implementation research in this area. This article highlights 3 areas where relevant implementation research is needed and examples of the types of research that could help fill these key evidence gaps (Table 1).
Opportunities for Implementation Science on Social Prescribing
Social Screening Research
Research is needed on the acceptability of social needs screening in the context of medical care, differences between tools used for capturing information on social needs, and ways that such tools can be adapted to optimize screening uptake in different settings and with different patient populations. New screening initiatives are multiplying around the country and provide ripe opportunities for this research. For example, the National Association of Community Health Centers and other partners have developed a screening tool (PRAPARE) that can help community health centers and other providers collect patients' SDH information. Building on PRAPARE, the nonprofit health care innovation center OCHIN is collaborating with the Kaiser Permanente NW Center for Health Research to study the implementation and use of PRAPARE and other EHR-based tools. (Table 1)
Workforce Research
Research is also needed on the workforce models that are most feasible and effective to carry out these activities in different settings. Models for social prescribing activities have included both clinical and nonclinical staff, including nurse and social worker case managers36,37, student volunteers38, community care coordinators39, and community health workers.40,41 Evidence is needed to better understand the benefits of different workforce models in different settings, identify core training and certification standards across programs, compare implementation strategies within these intervention models, and examine methods to retain and advance nonclinical staff. In one promising example of workforce implementation science, researchers at the University of Pennsylvania have explored strategies to adapt inpatient community health worker programs to outpatient settings. (Table 1)
Payment Models
Research could also assess how payment models can be structured to incentivize or otherwise support the adoption and spread of social prescribing programs.42 These models include federal programs such as CPC+, Health Homes, and Accountable Health Communities, and state/regional programs, including alternative payment methodology demonstrations focused on community health centers43, state Medicaid SDH risk adjustment initiatives44, and some Medicaid waiver demonstrations.37 These programs offer both site-specific and cross-site opportunities to explore which models can catalyze and sustain social prescribing activities, including the workforce and technologic infrastructure needed for intersectoral work. Hennepin Health's payment model, for example, incorporates risk-sharing strategies between participating entities and the reinvestment of annual cost savings into new social prescribing interventions. Together, these have enabled more workforce and data-sharing innovations. Implementation research on this project is ongoing, focused on the impact and sustainability of the interventions developed through reinvestment and on how this model could be replicated in other settings.35 (Table 1)
Looking to the Future
Given the mounting evidence linking SDH and health, health care delivery systems must ask what their roles are in identifying and addressing patients' social and economic needs. The growth of social prescribing pilot programs across the United States can and should be leveraged to explore this looming question. Studies of effectiveness, however, are both necessary and insufficient; they must be aligned with and followed by implementation research that examines program feasibility, including whether and how social prescribing activities can be implemented, disseminated, and sustained in real-world clinical settings. This research should address questions such as (1) which social screening tools are most appropriate for which settings, (2) which implementation approaches maximize adoption, (3) how health information technology could facilitate social screening and community resource linkages, (4) how different workforce models could be leveraged in diverse clinical contexts, and (5) how federal, state, and local payment models can support social prescribing activities over time.
The implementation research examples highlighted here provide insights into how implementation science can support social prescribing's transition from innovative pilot work to sustainable primary care practice. Many other opportunities exist for new research to be conducted in this rapidly evolving field. Advancing such research will require a sustained commitment from many stakeholders, including innovators, health professional organizations, and agencies focused on developing health care standardization and improving quality. Capitalizing on existing practice-based experiments will also require strong research-practice partnerships, which can be facilitated through practice-based research networks with expertise in practice-based methodologies.45 The Agency for Health Care Research and Quality has been a strong supporter of such implementation science in primary care.46 Recent threats to their budget and overall sustainability directly conflict with this implementation research agenda.47⇓⇓–50 Primary care providers and researchers investing in SDH work will need sufficient funding to ensure that the rigorous implementation science needed can prosper in this nascent field.
Acknowledgments
The authors acknowledge the Starfield Health Equity Summit team Jennifer Devoe, MD, PhD, FAAFP; Viviana Martinez-Bianchi, MD, FAAFP; Ronya Green, MD; Jennifer Edgoose, MD, MPH; and Sonja Likumahuwa-Ackman, MID, MPH, and CFAR staff, Malachi O'Connor, PhD; Ashleigh Reeves, MA; and Jason Ring, for their contributions to manuscript development. The authors also wish to thank Stephanie Chernitskiy for her assistance in manuscript preparation.
Appendix
Notes
This article was externally peer reviewed.
Funding: This work was partially supported by Kaiser Permanente, grant CRN-5374–7544-15320 (LG, CF) and the National Institute of Diabetes and Digestive and Kidney Disease, grant R18DK105463 (EKC, RG).
Conflict of interest: none declared.
To see this article online, please go to: http://jabfm.org/content/31/3/315.full.
- Received for publication June 28, 2017.
- Revision received October 2, 2017.
- Accepted for publication October 4, 2017.