OBJECTIVES: To explore differences in mean costs (from a UK National Health Service perspective) and effects of pharmacist-led management of chronic pain in primary care evaluated in a pilot randomised controlled trial (RCT), and to estimate optimal sample size for a definitive RCT. DESIGN: Regression analysis of costs and effects, using intention-to-treat and expected value of sample information analysis (EVSI). SETTING: Six general practices: Grampian (3); East Anglia (3). PARTICIPANTS: 125 patients with complete resource use and short form-six-dimension questionnaire (SF-6D) data at baseline, 3 months and 6 months. INTERVENTIONS: Patients were randomised to either pharmacist medication review with face-to-face pharmacist prescribing or pharmacist medication review with feedback to general practitioner or treatment as usual (TAU). MAIN OUTCOME MEASURES: Differences in mean total costs and effects measured as quality-adjusted life years (QALYs) at 6 months and EVSI for sample size calculation. RESULTS: Unadjusted total mean costs per patient were £452 for prescribing (SD: £466), £570 for review (SD: £527) and £668 for TAU (SD: £1333). After controlling for baseline costs, the adjusted mean cost differences per patient relative to TAU were £77 for prescribing (95% CI -82 to 237) and £54 for review (95% CI -103 to 212). Unadjusted mean QALYs were 0.3213 for prescribing (SD: 0.0659), 0.3161 for review (SD: 0.0684) and 0.3079 for TAU (SD: 0.0606). Relative to TAU, the adjusted mean differences were 0.0069 for prescribing (95% CI -0.0091 to 0.0229) and 0.0097 for review (95% CI -0.0054 to 0.0248). The EVSI suggested the optimal future trial size was between 460 and 690, and between 540 and 780 patients per arm using a threshold of £30 000 and £20 000 per QALY gained, respectively. CONCLUSIONS: Compared with TAU, pharmacist-led interventions for chronic pain appear more costly and provide similar QALYs. However, these estimates are imprecise due to the small size of the pilot trial. The EVSI indicates that a larger trial is necessary to obtain more precise estimates of differences in mean effects and costs between treatment groups.
TRIAL REGISTRATION NUMBER: ISRCTN06131530.