Analytical approach:
An economic evaluation was conducted alongside a multicentre randomised controlled trial (RCT) in which 164 patients were randomised to current local practice or FACETS and observed for four months post intervention. The authors stated that the perspective was of the NHS and personal social service provider (third party payer).
Effectiveness data:
The primary effectiveness outcomes were fatigue severity and self-efficacy for managing fatigue. Fatigue severity was measured using the Global Fatigue Severity (GFS) subscale of the Fatigue Assessment Instrument (FAI) questionnaire. Self efficacy was measured using the Multiple Sclerosis Fatigue Self-Efficacy scale questionnaire. Outcomes in both arms were measured at one week (baseline) before the start of FACETS, one month (follow-up one) and four months (follow-up two) after the final FACETS session. All questionnaires were self-reported and administered and returned through the post. Disease-specific quality of life was measured using the total score on the Multiple Sclerosis Impact Scale.
Monetary benefit and utility valuations:
Overall quality of life was measured using the EQ-5D and SF-6D questionnaires. Measurements were taken at the same time as for the effectiveness outcomes. Utility values were derived from the questionnaires. Quality-adjusted life years (QALYs) were calculated from the utility values at different time points using the under the curve method assuming linear extrapolation between points. Differences in QALYs between trial arms were adjusted for baseline values to avoid bias.
Measure of benefit:
The health benefit was measured in terms of the improvement in GFS scores, the number of persons with a clinically significant improvement in GFS (defined as a 0.5 point improvement) and quality-adjusted life-years (QALYs).
Cost data:
The cost categories included intervention training, equipment and delivery costs, a range of health care worker consultation costs, treatment costs, accident and emergency visit costs and hospital admission and stay costs. Costs were calculated for the three-month period prior to follow-up two in each arm. Health care service use data were derived from the trial using sampling forms completed by clinical staff who delivered the programme. Unit costs applied to resource data were derived from national sources (including the Personal Social Services Research Unit, NHS Reference costs and local NHS Trust cost data). Costs were reported in 2010 GBP (£).
Analysis of uncertainty:
Standard deviation and 95% confidence intervals values were reported alongside relevant outcome data. Non-parametric bootstrapping methods were used to estimate confidence intervals around cost estimates for healthcare service use. Sensitivity analyses were conducted to assess the affect of uncertainties around cost estimates on the results.