Published in:
Open Access 01-12-2013 | Study protocol
Predicting response to physiotherapy treatment for musculoskeletal shoulder pain: protocol for a longitudinal cohort study
Published in: BMC Musculoskeletal Disorders | Issue 1/2013
Login to get accessAbstract
Background
Shoulder pain affects all ages, with a lifetime prevalence of one in three. The most effective treatment is not known. Physiotherapy is often recommended as the first choice of treatment. At present, it is not possible to identify, from the initial physiotherapy assessment, which factors predict the outcome of physiotherapy for patients with shoulder pain. The primary objective of this study is to identify which patient characteristics and baseline measures, typically assessed at the first physiotherapy appointment, are related to the functional outcome of shoulder pain 6 weeks and 6 months after starting physiotherapy treatment.
Methods/Design
Participants with musculoskeletal shoulder pain of any duration will be recruited from participating physiotherapy departments. For this longitudinal cohort study, the participants care pathway, including physiotherapy treatment will be therapist determined.
Potential prognostic variables will be collected from participants during their first physiotherapy appointment and will include demographic details, lifestyle, psychosocial factors, shoulder symptoms, general health, clinical examination, activity limitations and participation restrictions.
Outcome measures (Shoulder Pain and Disability Index, Quick Disability of the Arm, Shoulder and Hand, and Global Impression of Change) will be collected by postal self-report questionnaires 6 weeks and 6 months after commencing physiotherapy.
Details of attendance and treatment will be collected by the treating physiotherapist. Participants will be asked to complete an exercise dairy.
An initial exploratory analysis will assess the relationship between potential prognostic factors at baseline and outcome using univariate statistical tests. Those factors significant at the 5% level will be further considered as prognostic factors using a general linear model.
It is estimated that 780 subjects will provide more than 90% power to detect an effect size of less than 0.25 adjusted for other variables which have a co-efficient of determination (R-squared) with the outcome of up to 0.5. Assuming a 22% loss to follow up at 6 months, 1000 participants will initially be recruited.
Discussion
This study may offer service users and providers with guidance to help identify whether or not physiotherapy is likely to be of benefit. Clinicians may have some direction as to what key factors indicate a patient’s likely response to physiotherapy.