Stroke Patients' Outpatient Rehabilitation Therapy (SPORT)

Background Stroke is the leading cause of long-term disability in Singapore. Stroke accounts for 9.1% of years of life lost (YLL) and 6.1% of disability-adjusted life years (DALYs). With a rapidly aging population, the burden of stroke is anticipated to grow correspondingly in the years to come. Stroke survivors with moderate or severe functional disabilities require long-term medical and social care, imposing financial and psychological burdens on the patients and their family members. Research indicates that intensive specialized stroke rehabilitation can improve post-stroke functional outcomes for a substantial portion of stroke patients. Extant studies also indicate that rehabilitation is cost-effective in both inpatient and outpatient settings, and can potentially lead to reductions in both sub-acute and long-term post-stroke healthcare costs. In addition, a longitudinal study on post-stroke patients in Singapore found that receiving outpatient rehabilitation service at day rehabilitation centers resulted in earlier functional recovery. However, despite the benefits associated with adherence to prescribed rehabilitation sessions, the uptake rate of and adherence to outpatient rehabilitation service in Singapore is low. Only 33% of stroke survivors who were recommended for post-discharge community-based rehabilitation actually took up the service. The low uptake rate also implies substantial potential avoidable disability among stroke patients in Singapore.

Significance According to a longitudinal survey on patients who were discharged from community hospital and recommended for post-discharge outpatient rehabilitation in Singapore, financial burden of rehabilitation services and the inconveniences of transportation are associated with the low attendance rates to outpatient rehabilitation.

Behavioral economic theory and empirical studies also suggest that the low rate of uptake and adherence, at least in part, is attributable to stroke patients' perceptions or uncertainty about the health benefits of rehabilitation. Stroke patients may make a decision to utilize ORS or not, based on incomplete information about the service attributes (e.g. service quality, social connectedness, and perceived benefits). Over time, patients and caregivers may update their perceived service attributes through information and education on stroke management or actual experiences. Hence, providing incentives to take up ORS and educational materials on post-stroke management will elevate the current uptake rate of ORS and long-term adherence.

The proposed study will be the first attempt in Singapore to test innovative ways of encouraging uptake of stroke ORS. It will systematically test promising economic strategies that might be effective and cost-effective in increasing the uptake of and adherence to the prescribed ORS. Given the high bed occupancy rate in public acute hospitals in Singapore, outpatient center-based rehabilitation service could be a more viable option than inpatient rehabilitation and more cost-effective in Singapore. It will provide evidence for future policy change in financing outpatient rehabilitation service and other long-term care services in Singapore and other countries experiencing similar problems.

Specific Aims \& Hypothesis Stroke is the leading cause of long-term disability in Singapore, and the burden of stroke is posing challenges to the healthcare system and society. Previous international and regional research studies suggest that a substantial proportion of stroke patients can benefit from post-stroke rehabilitation services and achieve improved functional outcomes. However, despite government subsidies and Medisave coverage for outpatient rehabilitation, the uptake rate of outpatient rehabilitation service (ORS) among stroke patients is low.

The investigators propose a randomized controlled trial (RCT) to test different interventions aimed at improving the uptake of and adherence to outpatient rehabilitation services: a rehabilitation educational program (Group 1, Education), free transportation only (Group 2, Transportation Incentives) and free transportation and rehabilitation sessions (Group 3, Transport \& Sessions Incentives) for the first 3 months from the date of discharge.

The study aims to:

1. Examine whether incentives for outpatient rehabilitation services (Group 2 and Group 3) are more effective at improving the uptake of rehabilitation services compared with education program only (Group 1);
2. Determine whether stroke patients in Group 3 Transport \& Sessions Incentives are more likely to take up ORS than those in the Group 2 Transport Incentives only;
3. Assess the effectiveness of interventions (Group 2 and Group 3) on patient health outcomes; and
4. Evaluate the cost-effectiveness of the interventions (Group 2 and Group 3) at the 4th month.

At each recruitment site, the Research Coordinator will assess the patient's eligibility before approaching them for recruitment. The patient will be given an Abbreviated Mental Test (AMT) for cognitive assessment. If the patient is cognitively unfit, then his/her caregiver/proxy will be required to sign the informed consent form on behalf of the patient. The enrolment period will last for approximately 14-16 months or when the recruitment target of 300 participants has been achieved.

Baseline Assessment: After signing the informed consent form, the Research Coordinator will proceed with the baseline assessment which consists of participant's socioeconomic characteristics, health status, health and medical services utilization, financial status, stroke-related disability measures, quality of life, and self-efficacy.

The Research Coordinator will educate all participants on their conditions and how to recover after stroke. A pamphlet with information on post-stroke rehabilitation will also be distributed to all participants.

Patient-level Intervention: The intervention phase is for a period of three months post-discharge. The Study Team at Duke-NUS will be coordinating all the incentivised services for the participants

Measures and Analysis The primary study outcome is uptake of outpatient rehabilitation services. Uptake is defined as a patient attending at least one rehabilitation session after discharge. A key secondary outcome will be adherence, defined as the proportion of times the participant attended prescribed outpatient rehabilitation sessions during the study period. Additional secondary outcomes of the study include measures of participants' stroke-related functional status (Modified Rankin Scale, mRS, and modified Barthel Index for Activities of Daily Living (ADL) limitation), quality of life (EQ-5D), resilience (CD-RISC 10), and participants' inpatient and other medical services utilization. The investigators will also obtain participants' baseline socio-demographic data and health status from electronic medical records and self-reported questionnaire as control variables. In addition, the program costs of all incentives provided to those in the interventions arms will be tracked systematically.

The investigators will conduct bivariate analyses and employ generalized linear model to examine the differences among three study arms in the outcomes of interest, as identified above. For all the health outcome variables such as, functional status, and will employ a difference-in-differences (DID) design to assess the impact of incentives. This method controls for secular trends in outcomes by comparing the longitudinal change in the controls from the concurrent change in cases between the baseline and follow-up period. Specifically, the covariates for the interaction term in the regression model (i.e. the difference-in-differences estimates) represent the difference in the changes over time. In addition, the incremental cost-effectiveness ratios (ICER) for interventions will be calculated by dividing the difference in costs of control arm and that of intervention arms by the incremental effectiveness of both intervention arms compared with control arm.

Power calculation Assuming the uptake rate for control arm is 33% and the detectable difference for the intervention arm is 10% (that is, the intervention arm that is most different from control has an uptake rate of 43%), the desired sample size is a total of 300 subjects (100 per arm) to achieve 80% power and to accommodate 10% attrition.