Adapting a Complex, Integrated Health and Social Services Intervention in Two Communities. (20th August 2020)

• Record Type: Journal Article
• Authors: Nichols, L.; Knighton, A.; Brunisholz, K.; Elbel, R.; Smith, G.; Choberka, A.; Belnap, T.; Allen, T.; Moore, M.; Srivastava, R.
• Is Part Of: Health services research. Volume 55:Number 1(2020)
• Page Start: 111
• Page End: 112
• Publication Date: 2020-08-20
• Subjects: Medical care -- Periodicals; Medical care -- Evaluation -- Periodicals; Hospital care -- Periodicals; Health services administration -- Periodicals; 362
• DOI: 10.1111/1475-6773.13489
• Languages: English
• ISSNs: 0017-9124

Abstract

Abstract : Research Objective: Poor population health is not randomly distributed but follows a social gradient with adversity clustering in specific communities. Under the Alliance for Determinants of Health program, Intermountain Healthcare is leading a three‐year demonstration project designed to improve population health outcomes through integrating health care and social services. Recent federal policy efforts are testing interventions to address health inequities, yet little is known about how to effectively adapt these complex, highly contextual interventions to local settings. Study Design: A mixed methods analysis was used to adapt and deploy an integrated health and social services solution in two diverse communities using an intervention mapping approach. Specific steps taken to effectively adapt the intervention included the following: a needs assessment using local health needs and actuarial claims data, ethnographic field results, and key informant interviews; definition of change and performance objectives; theory and methods selection; program plan and production; and implementation and evaluation planning. Population Studied: Key national, state, and local government and community leaders, health departments, social service providers, health care providers, police/justice departments, education systems, advocacy organizations, and community members in two regionally and ethnically diverse communities in Utah. Principal Findings: The needs assessment identified meaningful variation in local needs and available infrastructure across zip‐code‐based areas that guided selection of two communities. Program change objectives were defined, including improved local health and social services integration across sectors involving more than 50 service partners. Given the need for strong local program engagement, a multi‐ecological approach using theory‐informed methods was selected to guide local intervention development. An adapted Accountable Health Communities Model was developed including local community‐led steering committees to provide strategic oversight to local needs. Intermountain provided $6 million in funding to each community distributed by local community‐led finance committees to address local social service gaps. Weekly program huddles with Alliance partners are used to report results and to continuously improve and resolve local barriers. An external team was hired to conduct the evaluation. During the first year, 1321 eligible members underwent initial screening at an Alliance provider location; 139 members were navigated to one or more service partners demonstrating local integration feasibility. The most frequent navigation requests were for transportation (16%), dental care (15%), food (15%), clothing (9%), rent (9%), and mental health care (8%). Conclusions: Local communities vary considerably in social service needs, capacity, and existing health care infrastructure. Intervention mapping provides a robust methodology for adapting complex integrated health and social services interventions that are responsive to the needs of diverse communities. Implications for Policy or Practice: Implementing programs that integrate health and social services is complex and multifaceted and should be designed with the local implementation in mind. Rigorous implementation science methods may reduce the risk of implementation failure and strengthen the potential generalizability of results.