Choice of NSAID and management strategy in rheumatoid arthritis and osteo-arthritis : the impact on costs and outcomes in the UK.

OBJECTIVE: Although nonsteroidal anti-inflammatory drugs (NSAIDs) are an effective therapy for rheumatoid arthritis, they are associated with significant adverse effects, the management of which imposes additional costs on the healthcare system. Prescribing NSAIDs which have a lower risk of major adverse effects as the first-line NSAID for patients with rheumatoid arthritis and osteoarthritis may be expected to lead to an improvement in clinical outcomes and reduce overall treatment costs. This analysis examines data from a published randomised controlled trial of 5 NSAIDs to explore these hypotheses.

DESIGN AND SETTING: Data from a clinical trial comparing 5 NSAIDs were combined with published cost data to construct 2 clinical decision models, reflecting alternative approaches to the management of major and minor adverse effects in the UK.

INTERVENTIONS: The 5 NSAIDs evaluated in the analysis were nabumetone, diclofenac, ibuprofen, piroxicam and naproxen, although only the results for ibuprofen and nabumetone are reported.

MAIN OUTCOME MEASURES AND RESULTS: The total cost of care per patient receiving nabumetone was estimated to be between 25 pounds sterling (Pound) and 41 Pounds more expensive than ibuprofen. In a hypothetical cohort of 100,000 patients, there were between 690 and 821 more major adverse effects using ibuprofen than nabumetone. The cost per life-year gained (LYG) from using nabumetone rather than ibuprofen ranged between 1880 Pounds and 2517 Pounds (1995 values), depending upon the management of adverse effects.

CONCLUSIONS: These results indicate that: (i) prescribing the newer, currently more expensive, NSAIDs will not necessarily lead to cost savings; (ii) the management of adverse effects can have a significant impact on costs; and (iii) the additional cost may be justifiable in terms of the mortality and morbidity gains associated with the new lower-risk NSAIDs.