BACKGROUND & OBJECTIVE
Persistent shoulder pain is common and the response to physiotherapy can be variable with the best treatment still unknown. Up to 3% of doctors visits in the UK are for shoulder pain. Whilst the majority of patients receiving physiotherapy do completely recover, still around 30% have only slight or no recovery.
A previous study, using a multi-variable general linear model (GLM), identified prognostic factors that were consistently associated with a better outcome at a six month follow up. A limitation of this model was its ability to be used practically in the clinic. This study used a Classification and Regression Tree (CART) analysis that can be easily graphically represented with the prognostic factors shown in hierarchical importance. These prognostic factors can then be targeted through therapeutic interventions.
The aim of the study was “to provide clinicians with a guide to the most influential factors that predict outcome for people undergoing management for non-surgical musculoskeletal shoulder pain”.
1030 people with shoulder pain were recruited between 2011 and 2013 in the UK who were referred to physiotherapy treatment for shoulder pain, with exclusions for shoulder dislocations, fracture and radiculopathy. The validated outcome measures of the Shoulder Pain and Disability Index (SPADI) and Quick Disability of the Arm, Shoulder and Hand (QuickDASH) were collected at baseline and at a six month follow up.
Potential prognostic factors were collected at the initial appointment. These included clinical examination findings, general health, work, patient expectations and beliefs, and shoulder history.
Out of the 1030 participants included in the study, 810 provided complete outcome data at the six month follow up. Those who completed the outcome data at 6 months had greater pain self efficacy, were older by 10 years, and