Background: Arm pain is common, costly to health services and society. Physiotherapy referral is standard management, and while awaiting treatment, advice is often given to rest, but the evidence base is weak.
Objective: To assess the cost-effectiveness of advice to remain active (AA) versus advice to rest (AR); and immediate physiotherapy (IP) versus usual care (waiting list) physiotherapy (UCP).
Methods: Twenty-six-week within-trial economic evaluation (538 participants aged ≥18 years randomized to usual care, i.e. AA (n = 178), AR (n = 182) or IP (n = 178). Regression analysis estimated differences in mean costs and Quality-Adjusted Life Years (QALYs). Incremental cost-effectiveness ratios (ICERs) and cost-effectiveness acceptability curves were generated. Primary analysis comprised the 193 patients with complete resource use (UK NHS perspective) and EQ-5D data. Sensitivity analysis investigated uncertainty.
Results: Baseline-adjusted cost differences were £88 [95% confidence interval (CI): -14, 201) AA versus AR; -£14 (95% CI: -87, 66) IP versus UCP. Baseline-adjusted QALY differences were 0.0095 (95% CI: -0.0140, 0.0344) AA versus AR; 0.0143 (95% CI: -0.0077, 0.0354) IP versus UCP. There was a 71 and 89% probability that AA (versus AR) and IP (versus UCP) were the most cost-effective option using a threshold of £20,000 per additional QALY. The results were robust in the sensitivity analysis.
Conclusion: The difference in mean costs and mean QALYs between the competing strategies was small and not statistically significant. However, decision-makers may judge that IP was not shown to be any more effective than delayed treatment, and was no more costly than delayed physiotherapy. AA is preferable to one that encourages AR, as it is more effective and more likely to be cost-effective than AR.
Keywords: Arm pain; QALYs; cost-effectiveness; economic evaluation; physiotherapy; trial.
© The Author(s) 2018. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: firstname.lastname@example.org.