Educational program for physicians to reduce use of non-steroidal anti-inflammatory drugs among community-dwelling elderly persons - A randomized controlled trial

CONTEXT. Nonsteroidal anti-inflammatory drugs (NSAIDs) are among the most f requently prescribed drugs for patients 65 years of age or older, primarily for musculoskeletal symptoms of osteoarthritis. Because NSAIDs frequently cause serious gastrointestinal (GI) and other complications among elderly p atients, expert guidelines for osteoarthritis recommend acetaminophen-based regimens, which are safer and often as effective as NSAIDs. OBJECTIVE, Evaluate a physician education program that communicated guideli nes for management of osteoarthitis in elderly patients that emphasized avo idance of NSAIDs when possible. The program reviewed NSAID risks and benefi ts and recommended: reevaluating continuous NSAID users, considering substi tution of up to 4 gld of acetaminophen for the NSAID, and trying topical ag ents and nonpharmacologic measures. DESIGN AND SETTING. Randomized controlled trial among community-dwelling Te nnessee Medicaid enrollees. SUBJECTS. Study physicians had 5 or more patients who: were community-dwell ing Medicaid enrollees 65 years of age or older; had used NSAIDs regularly for at least 180 days; had had no medical care encounters during this perio d suggesting an indication other than osteoarthritis; and had 1 year of bas eline and follow-up data. The study thus included 209 physicians (103 inter vention/106 control) with 1,566 qualifying regular NSAID users (768/798). INTERVENTIONS. Face-to-face visit to study physicians by another physician, and reminder placements in the charts of patients eligible to have NSAID u se reevaluated. OUTCOMES. Change between baseline and follow-up years in: days of prescribe d NSAIDs, acetaminophen, other drugs for musculoskeletal disorders, and GI drugs; outpatient visits and inpatient days of stay; SF36 measures of gener al health, physical function, and bodily pain (from 40% random patient samp le); and over-the-counter NSAIDs (from the sample). RESULTS. Intervention-attributable reduction of 7% (95% CI, 3% to 11%) in d ays of prescribed NSAIDs use with concomitant increase in acetaminophen use . No significant changes in other study endpoints. The intervention effect was greater among 75 physicians with a completed study visit, whose 564 pat ients had a 10% (95% CI, 6% to 14%) attributable reduction in NSAID use. CONCLUSIONS. The educational program modestly reduced NSAID exposure in com munity-dwelling elderly patients without undesirable substitution of other medications or detectable worsening of musculoskeletal symptoms.