An Effectiveness-implementation Hybrid Study of Social Prescribing in a Singapore Community Hospital Setting

The proposed study will adopt a sequential mixed-methods design combining:

1. An open-label RCT to evaluate the effects of Social Prescribing on participants' general wellbeing, social support and healthcare utilisation. This open-label design approach is chosen in view of the challenge to blind the participants, investigators, and all peripheral staff to the treatment allocation in a practical clinical setting.

Potential patients will be recruited from 3 community hospitals (namely Bright Vision Hospital, Sengkang Community Hospital and Outram Community Hospital)

Randomisation Patients will be randomised to either the intervention or control group. An independent administrator in a central office will generate the entire randomisation sequence using GraphPad randomisation sequence software© 2017 before recruitment commenced. This data will be password protected and only accessible by the administrator. The individual allocation will be revealed to each patient, their ward doctor/clinician, the deprescribing and study teams just before the first deprescribing round, hereby ensuring allocation concealment.

Intervention

Key intervention elements will include:

• Upon enrolment, WBCs co-develop a personalised social prescription plan with the participant based on Social Determinants of Health (SDoH) using SBAR4;
• WBCs refer the participant to inpatient activities that suit the participant's interests and preferences; and · WBCs work with a local anchor support network, Community Network for Seniors (CNS), to link the participant to community assets based on the social prescription plan co-developed. This will take place from seven days prior to patient discharge to about 2-month post-discharge. WBCs will stop following up with the participant in community once CNS has successfully taken over the case.

This study will use a SBAR4 tool, an evolved version of the well-known SBAR framework, to facilitate the assessment. This tool was previously developed by the Principal Investigator and his research team for caring for patients with complex comorbidities and has been widely adopted both locally and internationally. The R4 components - Recommendation (i.e. an action plan for each SDOH of the patient), Resources (i.e. social and medical resources are mustered to support the patient), Responsibilities (i.e. role of the patient, stakeholders and care providers, and how they can be activated), and Relationships (i.e. connections with patient, caregivers, team members and service providers, are sustained to facilitate addressing SDOH) will also guide the WBC to co-develop the Social Prescribing plan with the patient.

The control group will receive usual care.(including educational pamphlets informing about community resources and to improve wellbeing)

The study aims to recruit 804 participants in total (402 in each group).

Inpatient The WBCs will conduct the first visit to the newly admitted patients within 3 days of patients' admission to screen patients' eligibility for the study.

For the intervention group, the WBCs will conduct approximately daily visits to the participants for:

1. collecting demographics and baseline data as described in F10;
2. co-developing the personalised social prescription plan with the participant based on Social Determinants of Health (SDoH) using SBAR4;
3. referring and accompanying the participant to attend inpatient activities that suit the participant's interests and preferences; and
4. informing and seeking the participant's agreement with community assets identified for the participant together with CNS. The duration of each visit will vary ranging from 15 minutes to 1 hour, depending on the comfort level of the participant.

For the control group, the interviewer will conduct approximately 2 visits to collect demographics and baseline data. The duration of each visit will range from 15 to 30 minutes.

Community phase For the intervention group, approximately at 1-week post-discharge, the WBCs will telephone the participant to check his/her transition back to community and readiness to start attending community activities or receiving services, as per the social prescription plan developed during the participant's inpatient stay. For community activities, upon receiving positive responses from the participant, the WBCs and CNS will visit the participant, accompanying him/her to the activity premise on the first day of the activity session and send the participant home after the session. The activity session will normally take about 1 hour.

For the control group, no intervention will be administered.

Approximately 2 months after the discharge, the interviewers will telephone the participants to schedule a telephone or face-to-face session for the post-discharge data collection, depending on the participants' preference. At the agreed timing, the interviewers will telephone or meet the participant and proceed with the quantitative survey, similar to the baseline data collection. The duration will be approximately 30 mins.

For the participants selected for the qualitative interview, the interviewers will also arrange a telephone or face-to-face interview session, depending on the participants' preference. At the agreed timing, the interviewers will telephone or meet the participant and conduct the semi-structured interview. The duration will be approximately 30 minutes.

Data Collection The participants' general well-being will be measured by the Brief Inventory of Thriving (BIT), a 10-item self-administered instrument consisting of 5 psychological well-being constructs in positive psychology: Positive emotions, Engagement, Relationship, Meaning and Accomplishment (PERMA), which establishes the theoretical basis of Social Prescribing under study. Though this tool has not been validated in a Singapore context, it is the only tool that the study team has found in literature that centres around PERMA and feasible to administer in a practical clinical setting. The participants' social support will be measured by the modified Medical Outcomes Study- Social Support Survey (mMOS-SS), an 8-item self-administered instrument covering 2 domains: emotional/informational support and tangible/instrumental support. The modified version is chosen over the full 19-item version because of its relevance and easiness of administration. Baseline data of the intervention group at enrollment of the study will be collected by the WBCs, as the information likely help the WBCs to formulate the social prescription plan. Baseline data of the control group and the 2-month (approximately) post-discharge data of the intervention and control groups will be collected by interviewers. 2 months have been set based on a previous prototype of Social Prescribing implemented in Bright Vision Hospital. It was found that patients generally took about 1 month to transit back home after discharge from the hospital and start attending community activities or receiving intervention sessions from the second month. Both instruments have been tested in wards to ensure that the WBCs and interviewers understand the questions correctly and explain to patients in a similar manner. The participants' health utilisation will include the polyclinic visit, emergency department visit,Specialist Outpatient Clinic (SOC) visit, unplanned admission and average length of stay. The 12-month pre-enrollment and 12-month post-enrollment utilisation data will be extracted from the national public healthcare administrative database (i.e. Omnibus). In addition, participants' functional status (measured by the Modified Barthel Index) and cognitive status (measured by the Abbreviated Mental Test) will be extracted from the SCH Management System. The period for extraction will be from the participants' admission to discharge.

2. Qualitative semi-structured interviews to understand the experience of participants (in the intervention group) with the WBCs and Social Prescribing. Following the quantitative survey, purposive sampling will be used, targeting 30 participants including 15 active service users (i.e. patients who have been attending the activities/intervention sessions as per the social prescription plan) and 15 inactive service users, based on the manpower resources available. Semi-structured interviews will be completed through telephone or face-to-face modes after the participants have attended the first activity session in community or received the first service session from community partners.

Both the quantitative and qualitative data will also be used to assess the feasibility, acceptability and adoption of the WBC-driven model. No other retrospective data will be used for this proposed research except the 12-month pre-enrollment healthcare utilisation.