Recommendations for exercise adherence measures in musculoskeletal settings: a systematic review and consensus meeting

Background: Exercise programmes are frequently advocated for the management of musculoskeletal disorders; however, adherence is an important pre-requisite for their success. The assessment of exercise adherence requires the use of relevant and appropriate measures, but guidance for appropriate assessment does not exist. The aim of this study was to recommend outcome measures of exercise adherence that have clinical and research utility in the musculoskeletal field. Methods: There were two key stages to the research. First, a systematic review of the availability, quality and acceptability of measures used to assess exercise adherence in musculoskeletal disorders; second, a consensus meeting. The systematic review was conducted in two phases and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure a robust methodology. Phase one identified all reproducible measures that have been used to assess exercise adherence in a musculoskeletal setting. Phase two identified published and unpublished evidence of the measurement and practical properties of identified measures. Study quality was assessed against the COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) guidelines. A shortlist of measures was produced for consideration during stage two: a meeting of relevant stakeholders (n=14) in the United Kingdom. During this meeting consensus on the most relevant and appropriate measures of exercise adherence for application in research and/or clinical practice settings was sought. Key findings: Our systematic review identified numerous measures which had been used as measures of exercise adherence within musculoskeletal research. Of these 36 were deemed reproducible. However, only six had been evaluated as specific measures of exercise adherence for musculoskeletal research: evidence of essential measurement and practical properties was mostly limited or not available. Assessment of relevance and comprehensiveness was largely absent and there was no evidence of patient involvement during the development of any measure. During the consensus process the stakeholders reached agreement that none of the measures were relevant, appropriate or acceptable for use in musculoskeletal clinical or research settings. Conclusions and recommendations: Numerous exercise adherence measures are currently used within musculoskeletal research. However, many of these measures were not originally developed for this purpose; many approaches are not reproducible; and evidence of essential measurement and practical properties was only identified for a limited number of measures. Moreover, substantial methodological and quality issues were identified in the development and evaluation of the six short-listed measures which reduces confidence in the ability of these measures to reliably and validly evaluate adherence to exercise. Furthermore key stakeholders unanimously agreed that these measures were not fit for purpose. Measures of exercise adherence must be clearly conceptualised. Future development and evaluation should seek to involve patients, clinicians and researchers as active collaborators and use credible methods to develop and evaluate an appropriate measure of exercise adherence.