A regional audit of the use of COX-2 selective non-steroidal anti-inflammatory drugs (NSAIDs) in rheumatology clinics in the West Midlands, in relation to NICE guidelines

Objectives. Whilst all non-steroidal anti-inflammatory drugs (NSAIDs) can cause adverse gastrointestinal events, COX-2-selective inhibitors (COX-2) may have improved gastrointestinal safety compared with non-selective NSAIDs (NSNSAIDs). In 2001, the National Institute for Clinical Excellence (NICE) published guidance on the use of the COX-2 agents celecoxib, rofecoxib, meloxicam and etodolac for rheumatoid arthritis (RA) and osteoarthritis (OA). This study aimed to audit the appropriateness of NSAID use in relation to NICE guidance in rheumatology out-patients. Methods. Questionnaires were completed for all patients attending clinics in 18 rheumatology units in the West Midlands over a 2-week period. Data collected included patient demographics, NSAID type, indications, duration of use (>= 3 months was considered prolonged), and concomitant prescription of corticosteroids, warfarin and gastroprotective agents. Results. Data were collected on 2846 patients; 1164 (41%) were taking NSAIDs (791 NSNSAIDs, 373 COX-2). Of the 1164 NSAID users, 753 (65%) had a diagnosis of RA or OA (483 NSNSAIDs, 270 COX-2). Overall, 37% of NSAID prescriptions were appropriate. Of the NSNSAID users, 92% had at least one risk factor for adverse gastrointestinal events and were therefore inappropriately treated. Prolonged use (in 89%) and age >= 65 yr (in 23%) were the most frequent risk factors identified. Of the COX-2 users, 97% had one or more risk factors and were appropriately treated. Analysis of the RA/OA subgroup revealed similar findings. Thirty-six per cent were taking NSAIDs appropriately; 97% of NSNSAID use was inappropriate and 97% of COX-2 use was appropriate treatment. In the whole cohort, gastroprotective agents were used in 26% of NSNSAID users, 56% of gastroprotective agents being proton pump inhibitors. Conclusions. Ninety-two per cent of patients attending rheumatology clinics who were taking NSNSAIDs should have been prescribed a COX-2-selective agent in relation to NICE guidance. Duration of use and age >= 65 yr emerged numerically as the most important risk factors. Significant numbers of patients taking NSNSAIDs may be at risk from adverse gastrointestinal events and clinicians may wish to review their prescribing patterns. Conversely, 97% of patients taking COX-2 agents were treated appropriately. Although practice overall conformed poorly with NICE guidance, NSAID prescribing also needs to be considered in the context of recent concerns regarding the cardiovascular risks of COX-2 agents.