The Impact of Social Worker Referrals on Diabetes and Loneliness

This study is a pragmatic, single-centre, parallel-group randomized controlled trial evaluating the integration of social prescribing into primary care for older adults with diabetes who experience loneliness. The intervention is delivered within an academic family health team and leverages existing interprofessional resources to address social determinants of health as part of routine clinical care. Potential participants will be identified through electronic medical record (EMR) queries and clinic-based outreach. Loneliness screening will be conducted using the three-item UCLA Loneliness Scale (UCLA-3), administered via a secure electronic platform (REDCap). A flexible consent process, including both electronic and verbal options, will be used to accommodate varying levels of digital literacy within the target population. Following enrolment, participants will be assigned to study arms using simple randomization (coin toss). The intervention consists of a structured referral to a primary care social worker. The social worker will conduct an initial assessment of the participant's social context, including social supports, barriers to engagement, and unmet practical needs. Based on this assessment, a co-developed plan will be implemented, which may include connection to community resources, facilitation of social engagement activities, and support addressing social or logistical barriers. Follow-up will be individualized in frequency and modality (in-person or virtual), reflecting a pragmatic, patient-centred approach. Intervention delivery will not be protocolized beyond core components, in order to reflect real-world clinical practice. All social worker interactions will be documented within the EMR, allowing for characterization of intervention exposure (e.g., number and type of contacts). Participants in the comparator arm will continue to receive usual diabetes care during the study period, with access to social worker referral after study completion. Study data will be collected through a combination of participant-reported measures and EMR-derived clinical data. Data will be stored in REDCap using unique study identifiers, with linkage files maintained separately on secure, access-restricted servers. Analyses will use regression-based approaches to estimate between-group differences, adjusting for baseline values and relevant covariates. This study is designed to evaluate the feasibility and effectiveness of incorporating social care into chronic disease management within primary care settings. Findings will inform the scalability and implementation of social prescribing interventions aimed at reducing loneliness and improving health outcomes among older adults.