DOP065 Long-term cost-effectiveness of tight control for Crohn’s disease with adalimumab-based treatment: economic evaluation beyond 48 weeks of CALM trial

Abstract

Background: Tight control (TC) for Crohn's disease (CD), using biomarkers (faecal calprotectin and C‐reactive protein) to direct adalimumab (ADA) treatment, was associated with improved outcomes and considered cost‐effective compared with clinical management (CM) in a 48‐week (week) multicentre, randomised controlled trial (CALM).1,2 In this analysis, long‐term cost‐effectiveness (CE) was assessed beyond 48 week. Methods: The CE of TC vs. CM was assessed in a UK setting. An ordered probit regression estimated CDAI‐based health state (remission: CDAI <150, moderate: CDAI ≥150 to <300, severe: CDAI ≥300 to <450, very severe: CDAI ≥450) transition matrices for TC and CM patients; the matrices predicted patient health states weekly over 5 years. A probit predicted likelihood of hospitalisation as a function of health state and randomisation to TC or CM. Observed ADA 40 mg injections were summed over 48 week, average injections over week 23 to 48 were assumed to continue from week 49 to 260. ADA costs were based on ex‐factory prices. Health states were associated with health utility and costs from prior UK analyses. 3,4 Remission rate, CD‐related hospitalisations, ADA injections, other direct medical costs, quality‐adjusted life‐years (QALYs), and incremental cost‐effectiveness ratio (ICER) were calculated at 2 and 5 years. Work Productivity and Activity Impairment was converted into productivity measures using UK average weekly earnings, and probabilistic sensitivity analyses (PSA) were conducted. Results: Over 2 years, TC was associated with a higher remission rate (65.3% vs. 50.7%), fewer CD‐related hospitalisations (0.275 vs. 0.720/person year) and more ADA injections (mean 61.34 vs. 46.17) than CM. Total medical costs over 2 years were £25 808 and £24 939 for TC and CM; TC had 0.09 higher QALYs (95% confidence interval [CI]: 0.16 to 0.03) and £868 higher total medical costs (95% CI: £4551 to ‐£34 094). The ICER was £10 102 per QALY (95% CI: £119 525 to ‐£405 732). PSA simulations indicated 67.1% of the time an ICER was below a cost‐per‐QALY threshold of £30 000. TC became dominant (i.e., ICER<0) when including the costs associated with work productivity gained (‐£9328 in TC and ‐£6206 in CM). The results for 5‐year timeframe were similar, with an ICER of £9030 in base case and a negative ICER when including work productivity costs. Conclusions: Cost‐effectiveness of TC improved over time when extrapolating outcomes from the CALM trial. Incorporating costs related to work productivity further strengthened the economic value of TC.