Strengthening Quality Improvement in Rehabilitation by Standardized Reporting of Functioning Outcomes

Background: The primary goal of rehabilitation is to optimise functioning in everyday life. Therefore, functioning is a main quality indicator in the field of rehabilitation. Functioning is defined and classified by the World Health Organization's International Classification of Functioning, Disability and Health (ICF). In clinical practice, functioning outcomes are collected with a variety of ordinal-scaled assessment tools that are often only applicable within individual clinical encounters. A Standardized Assessment and Reporting System (StARS) can support the comparison and aggregation of functioning outcomes assessed with different tools, building a basis for continuous learning and quality improvement. This doctoral thesis takes Switzerland as an example, where musculoskeletal and neurological rehabilitation clinics use different assessment tools, the Functional Independence Measure (FIM™) or the Extended Barthel Index (EBI), to report functioning outcomes for national rehabilitation quality reviews, what limits the comparability across clinics.

Objective: To examine how an ICF-based and interval-scaled StARS for functioning outcomes can be created and implemented in Swiss rehabilitation quality reviews.

Methods: Four quantitative studies and stakeholder involvement activities were conducted. The basis of the studies were datasets with 18'000 cases collected for Swiss rehabilitation quality reviews in 2016 from 29 musculoskeletal or neurological rehabilitation clinics. In studies 1 and 2, the Rasch model was applied to define whether FIM™ and EBI can be reported as unidimensional interval-scaled metrics. In study 3, an ICF-based interval-scaled common metric encompassing FIM™ and EBI was created as core of a StARS by applying ICF Linking Rules and the Rasch model. In study 4, the impact and added value of the developed StARS for Swiss rehabilitation quality reviews was examined in comparison to the current ordinal-scaled system, including descriptive statistical methods, regression models and the exploration of further development areas of the reviews. Stakeholder involvement activities comprised a stakeholder brief and a stakeholder dialogue conducted to support the development of strategies for implementing a StARS for functioning outcomes in Switzerland.

Results: Studies 1 and 2 showed that FIM™ and EBI can be reported as unidimensional and interval-scaled metrics for musculoskeletal and neurological rehabilitation, when Rasch-based transformation is applied. In study 3, concept and score equivalence of FIM™ and EBI could be established, resulting in an ICF-based interval-scaled common metric. In study 4, the comparison between the interval-scaled and the ordinal-scaled reporting systems showed that the achieved outcomes reported on an interval-scaled StARS tended to be smaller but more precise. Furthermore, study 4 demonstrated the added value of employing the ICF as the basis of a StARS, showing concrete functioning categories that can inform further development of national rehabilitation quality reviews. The output of the stakeholder brief and stakeholder dialogue was an implementation agenda in which the stakeholders decided on the next steps for implementing the developed StARS.

Conclusion: This thesis shows how a StARS for functioning outcomes can be created for national quality reviews in rehabilitation and that it can have a positive influence and added value in comparison to the currently applied ordinal-scaled reporting system. Moreover, the thesis illustrates the potential of involving stakeholders in developing implementation strategies to implement a StARS. An ICF-based and interval-scaled StARS enables valid aggregation and comparison of functioning outcomes assessed with different assessment tools resulting in richer data. This, in turn, enables learning from functioning outcomes and has the potential to inform and strengthen quality improvement, in the sense of a learning health system.