Korean cost-effectiveness analysis of NSAIDs, NSAIDs with co-treatments to prevent gastrointestinal toxicity, and COX-2 specific inhibitors in the treatment of rheumatoid arthritis

Yun, Hyung Ran; Corzillius, Michael; Kim, Seong Yoon; Bae, Sang Cheol

Abstract

BACKGROUND: Nonsteroidal anti-inflammatory drugs (NSAIDs) are used in nearly every patient with rheumatoid arthritis (RA) but their use can be associated with gastrointestinal (GI) side effects, which may be prevented with prophylactic prescription of misoprostol and omeprazole. Recently marketed COX-2 specific inhibitor (COX-2) affords protection against gastropathy. This study was performed to assess Korean cost-effectiveness of NSAIDs, NSAIDs with co-treatments to prevent GI toxicity, and COX-2 in the treatment of RA, and compare it with American cost-effectiveness analysis.
METHODS
Markov (state-transition) models were used to simulate a cohort of RA patients with approximately 2.5:1 female to male ratio and 50 years, taking disease modifying antirheumatic drugs, low dose steroid (prednisone < or =10 mg/day) and one of the following strategies: 1) NSAIDs without prophylaxis, 2) NSAIDs with misoprostol, 3) NSAIDs with proton pump inhibitor (PPI), or 4) COX-2. Data on incidence, USA cost and consequences of adverse events from treatments were taken from the literature. Treatment costs of adverse events in Korea were calculated based on each disease code. Health effects were expressed as quality-adjusted life years (QALYs). Sensitivity analyses of probability of GI complication and cost were performed. Costs and health outcomes were discounted at a rate of 3% per year.
RESULTS
Among the strategies to prevent GI toxicity, PPI was the most cost-effective strategy in Korea and COX-2 was in USA, respectively. The incremental C/E (cost/effectiveness) ratio between PPI and no prophylaxis was 38,068x103won/QALY (32,044$/QALY) in Korea. The incrementalC/E ratio between COX-2 and no prophylaxis was 53,228$/QALY in USA. The base case analysis results were sensitive to cost of NSAIDs and COX-2 in Korea, and cost of NSAIDs in USA, respectively, and adverse event rates of NSAIDs, misoprostol, and PPI in Korea. The medical cost of NSAID side effects in Korea is 11% of USA, but the sensitivity analyses varying medical costs were robust. The sensitivity analyses using age, discount rate and utility were robust.
CONCLUSIONS
Although PPI in Korea and COX-2 in USA are the best option among the strategies to prevent GI toxicity, the incremental C/E ratios between PPI versus no prophylaxis in Korea and COX-2 versus no prophylaxis in USA are over 30,000 and 50,000$/QALY, respectively. However, it appears that the prescription of COX-2 in the group of higher cost NSAIDs users in Korea was the best option.