Use of patient-reported outcome measures in physiotherapy clinical trials in six major physiotherapy journals 2 decades apart (2000–2018): a meta-research design

Aim: Using patient-reported outcomes in research has been incentivised to encourage patient-centred care and ensure patient views are considered. We compared the use of patient-reported outcome measures (PROMs) in trials published in physiotherapy journals in 2000 and 2018, and evaluated whether the number of PROMs used differed between musculoskeletal, neurological, and cardiopulmonary subdisciplines. Design: Meta-research. Methods: Six major physiotherapy journals were searched for trials published in 2000 and 2018. Two independent reviewers extracted data on study characteristics and reporting of PROMs. PROMs were classified according to their outcome domains. Descriptive statistics and inferences were made based on proportions. A 20% difference between 2000 and 2018 was regarded as meaningful. Results: A total of 140 trials were included, 39 were published in 2000 and 101 in 2018. Eighty-four percent (n = 118/140) of trials reported ≥1 PROM, while 89% (n = 125/140) included ≥1 non-PROM. We found no meaningful differences on the average use of PROMs in 2000 and 2018: 74% (29/39) of trials in 2000 versus 88% (89/101) in 2018. PROM use in 2000 and 2018 was 88.5% and 84.4% in musculoskeletal physiotherapy, 57.2% and 86.1% in neurological physiotherapy and 0% and 88% in cardiopulmonary physiotherapy. The most used PROM outcome domains were symptoms and symptom burden (75%) and functional status (65%). Conclusion: Most trials from the six major physiotherapy journals sampled in 2000 and 2018 used PROMs, with no meaningful differences when comparing years. Fewer publications in 2000 than 2018 may account for the differences seen in neurological and cardiopulmonary physiotherapy.