File
Wynne-Jones G.1, Artus M.1, Bishop A.1, Lawton S.A.1, Lewis M.1, Main C.1, Sowden G.1,2, Wathall S.1, Burton K.3, van der Windt D.1, Hay E.M.1, Jowett S.4, Kigozi J.4, Foster N.E.1
1Keele University, Arthritis Research UK Primary Care Centre, Keele, United Kingdom, 2Staffordshire and Stoke-on-Trent Partnership NHS Trust, Stoke on Trent, United Kingdom, 3University of Huddersfield, Institute for Research in Citizenship and Applied Human Sciences, Huddersfield, United Kingdom, 4University of Birmingham, School of Health and Population Sciences, Birmingham, United Kingdom
Background: Does a vocational advice service improve work outcomes in patients with musculoskeletal pain? The SWAP randomised trial.
Purpose: The SWAP trial aimed to investigate the effectiveness and cost-effectiveness of adding a physiotherapist-led vocational advice (VA) service to best current primary care for adults with musculoskeletal pain problems.
Methods: The SWAP trial was a cluster randomised controlled trial in 6 general practices, 3 randomised to best current care (control), 3 randomised to best current care and the VA service (intervention). Patients were eligible if they were ≥18 years old, absent from work because of musculoskeletal pain for ≤6 months or struggling at work. Intervention practices referred patients to the VA service provided by a physiotherapist located in the practice. Primary outcome: number of days off work over 4 months. Secondary outcomes: self-reported time off work, return-to-work self-efficacy (RTW-SE), work presenteeism and performance, pain intensity (0-10 scale), health care utilisation and bothersomeness (1-5 scale) at 4 and 12 months. Analysis was by intention-to-treat adjusted for age, gender and GP practice size. Zero-inflated negative binomial regression clustering of effects at GP level was used to evaluate days off work. Cost-effectiveness and cost-benefit analyses were undertaken to calculate cost per sick day avoided and net societal benefits and return on investment.
Results: 348 participants (162 intervention arm: 186 control arm) were recruited. Baseline characteristics were comparable between arms; 4 month follow-up rates were 72% (intervention) and 79% (control). Participants in the intervention arm had significantly fewer days off work over 4 months (mean 9.3 days, SD 21.7) compared with control (mean 14.4, SD 27.7); adjusted incidence rate ratio (IRR) 0.51 (0.26, 0.99), p=0.048. This difference was largely due to fewer GP certified absent days (mean 8.4 days (SD 21.0) in the intervention arm, compared with 13.5 days (SD 27.5) in control). RTW-SE, work presenteeism and performance also differed significantly at 4 months between arms: mean difference 11.4 (2.97, 19.8), p=0.008, 2.23 (0.35, 4.10), p=0.002, -1.05 (-1.96, -0.14), p=0.023, respectively. At 12 months, days off work differed between arms but not significantly (mean in intervention 20.3 days, compared with control mean 24.3; adjusted incidence rate ratio (IRR) 0.65 (0.34, 1.25), p=0.198. Cost-effectiveness analysis: £7.20 per sick day avoided. Net societal benefit: £763 in favour of the intervention. Return on investment: £51 per £1 invested (without training), £25 per £1 invested (with training).
Conclusion(s): Adding a brief, early access, VA service delivered by physiotherapists, to best current primary care for adults with musculoskeletal pain leads to significantly fewer days absent from work over 4 months, improvements in return-to-work self-efficacy, work presenteeism and performance. The VA service had greater economic benefits compared to current best care.
Implications: A primary care based, brief, vocational advice service can be delivered by physiotherapists and leads to fewer days lost from work due to musculoskeletal pain with societal benefits in terms of return on investment.
Funding acknowledgements: The SWAP trial was funded by a National Institute for Health Research (NIHR) Programme Grant for Applied Research (NIHR-RP-PG-0707-10131).
Topic: Musculoskeletal
Ethics approval:
Ethical approval was obtained from NRES Committee West Midlands Staffordshire in April 2012 (REC reference:12/WM/0020).
All authors, affiliations and abstracts have been published as submitted.
