Comparative cost-effectiveness of robot-assisted and standard laparoscopic prostatectomy as alternatives to open radical prostatectomy for treatment of men with localised prostate cancer: a health technology assessment from the perspective of the UK National Health Service.

BACKGROUND: Robot-assisted laparoscopic prostatectomy is increasingly used compared with a standard laparoscopic technique, but it remains uncertain whether potential benefits offset higher costs.

OBJECTIVE: To determine the cost-effectiveness of robotic prostatectomy.

DESIGN, SETTING, AND PARTICIPANTS: We conducted a care pathway description and model-based cost-utility analysis. We studied men with localised prostate cancer able to undergo either robotic or laparoscopic prostatectomy for cure. We used data from a meta-analysis, other published literature, and costs from the UK National Health Service and commercial sources.

OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: Care received by men for 10 yr following radical prostatectomy was modelled. Clinical events, their effect on quality of life, and associated costs were synthesised assuming 200 procedures were performed annually.

RESULTS AND LIMITATIONS: Over 10 yr, robotic prostatectomy was on average (95% confidence interval [CI]) £1412 (€1595) (£1304 [€1473] to £1516 [€1713]) more costly than laparoscopic prostatectomy but more effective with mean (95% CI) gain in quality-adjusted life-years (QALYs) of 0.08 (0.01-0.15). The incremental cost-effectiveness ratio (ICER) was £18 329 (€20 708) with an 80% probability that robotic prostatectomy was cost effective at a threshold of £30 000 (€33 894)/QALY. The ICER was sensitive to the throughput of cases and the relative positive margin rate favouring robotic prostatectomy.

CONCLUSIONS: Higher costs of robotic prostatectomy may be offset by modest health gain resulting from lower risk of early harms and positive margin, provided >150 cases are performed each year. Considerable uncertainty persists in the absence of directly comparative randomised data.