The quality of rehabilitation services for patients with rheumatic and musculoskeletal diseases

Abstract

Background: Patients with rheumatic and musculoskeletal diseases (RMDs) constitute a large and increasing part of people in need for multidisciplinary rehabilitation services. Rehabilitation interventions are complex and individually adapted, and it is of utmost importance to coordinate all involved services to ensure continuous processes for each patient. However, several reports conclude that there is a gap between recommended and current delivery of rehabilitation services, with little coordination and communication across levels of healthcare, and lack of patient involvement in planning of supported self-management and follow-up interventions. Evaluation and improvement of rehabilitation quality may be guided by the three-fold model of structure, process, and outcomes, and capture the perspectives of both providers and patients. The use of quality indicators (QIs) and quality improvement programs (QIPs) are recognized as promising strategies to ensure better quality in healthcare, but these strategies are scarcely used within team-based rehabilitation for patients with RMDs.

Aim: The overarching aim was to explore and evaluate ways to measure, monitor and improve quality in rehabilitation services over time. The specific objectives were i) to assess the responsiveness of a QI set for use in rehabilitation, which comprises 19 structure, 11 process, and 3 outcome indicators, ii) to examine the associations between patient-reported quality of processes and clinical outcomes of rehabilitation, and iii) to investigate how a teambased QIP was delivered in rehabilitation practices, focusing on the structure dimension of quality and the providers’ fidelity to the planned processes.

Methods: Three different studies were undertaken to address the objectives, all nested within the Norwegian stepped-wedge cluster-randomized BRIDGE trial. The BRIDGE program, developed to improve coordination, continuity and follow-up, was added to the existing programs at eight rehabilitation centres in secondary care. The program components were motivational interviewing, patient-specific goal setting, written plans for rehabilitation and self-management, digital self-monitoring of progress on outcomes, and tailored follow-up. Data were collected from the provider teams and 293 patients with various RMDs admitted to rehabilitation at the participating centres. The first aim was examined in a longitudinal prepost study, using a construct approach to evaluate the responsiveness of the QI set by testing 62 hypotheses of expected changes in structure, process, and outcomes after adding the BRIDGE program. In the second study, using a longitudinal cohort design, linear and logistic mixed models were used to examine associations between the pass rates of QIs and outcomes (goal attainment, physical function, and health-related quality of life). The third aim was explored in a mixed methods (MMs) study, using a convergent approach to combine and compare quantitative (questionnaires) and qualitative (focus groups) data about program delivery.

Results: Analyses of responsiveness showed that ≤ 25% of the hypotheses were rejected, confirming the QI set’s ability to detect changes in quality of delivered, team-based rehabilitation. In the cohort study, no associations were found between patient-reported pass rates of process indicators and the outcome variables. The MMs study indicated that structural improvements do not necessarily lead to better quality of rehabilitation processes, in terms of the interactions between providers and patients. The results further support that providers’ program fidelity depends on both the rehabilitation content and on how this content is delivered. Potentials for improvements concerned follow-up and supported self-management, as well as the providers’ skills, knowledge and development as specialized rehabilitation workers.

Conclusions: The QI set showed satisfactory responsiveness when applied in team-based rehabilitation for adults with various RMDs, and can be used as a tool to capture changes and monitor maintenance of rehabilitation quality. The set can also be used to establish benchmarks for good quality in rehabilitation, and to evaluate effectiveness of quality initiatives. Based on the results from the second study, we suggest that inferences about quality of rehabilitation should be drawn from complementary information about both structures, processes, and outcomes. Lastly, quality in rehabilitation depends on several contextual factors, which exist at the level of institutions, teams, and individual providers. It seems particularly important to support rehabilitation providers’ confidence in delivering all parts of the intended care, and to develop a culture of continuous improvement within institutions and teams, and across sectors and levels of healthcare.