Analytical approach:
This economic evaluation was based on a single randomised controlled trial (RCT) and had a time horizon of one year. A subgroup analysis on patients' preference was conducted to assess its impact on the outcomes. The authors did not report the study perspective.
Effectiveness data:
: The RCT included patients aged over 18 years, with neck pain of musculoskeletal origin and lasting at least two weeks. A sample of 268 patients was enrolled, with 129 in the usual physiotherapy group and 139 in the brief intervention group. Questionnaires were completed by 78% of the usual group and 81% of the brief group at three- and 12-month follow-ups. The baseline characteristics of the two groups appeared to be different, but confounding due to these baseline differences was minimised using statistical adjustments.
Monetary benefit and utility valuations:
The utility valuations were obtained at baseline, and the third and twelfth months, from the sample of patients enrolled in the RCT, using the European Quality of life (EQ-5D) questionnaire.
Measure of benefit:
The summary benefit measure was quality-adjusted life-years (QALY) and discounting was not required, given the short time horizon. A secondary outcome measure was the Northwick Park Questionnaire (NPQ), which measured the level of neck pain and resulting disability.
Cost data:
: National Health Service (NHS) costs, patient costs and productivity costs were reported, but only UK NHS costs were analysed. These included the costs of the general practitioner, physiotherapist, hospital specialist, and other out-patient visits. The resource use was derived from data obtained from patients in the RCT using their diaries and case record forms, which were sent as postal questionnaires. The unit costs were based on data from UK national sources (e.g. British National Formulary, Department of Health, and National Statistics). All costs were expressed in UK pounds sterling (£) at 2002 prices and discounting was not relevant and was not applied.
Analysis of uncertainty:
A bootstrapping approach was used to obtain the 95% confidence interval (CI) for the differential costs and QALYs. Cost-effectiveness acceptability was presented to reflect the uncertainty in the incremental cost-effectiveness ratios. Statistical methods were applied to adjust for censored data and differences in the patient baseline characteristics.