Risks and benefits of bladder antimuscarinics among elderly residents of Veterans Affairs Community Living Centers.

OBJECTIVES: To evaluate risks and benefits of bladder antimuscarinics (BAMs) among elderly long term care nursing home residents. DESIGN: Retrospective cohort study using a new user design and propensity score matching. SETTING: Veterans Affairs (VA) Community Living Centers (Nursing Homes). PARTICIPANTS: Older adults (65 and older) admitted for long term care between October 1, 2002, and September 30, 2009. MEASUREMENTS: The study used multiple VA data sources (Minimum Data Set [MDS], inpatient, outpatient, and pharmacy prescriptions administrative files). The following outcomes were evaluated: (1) fractures (hip fracture and "any" fracture) identified from inpatient and/or outpatient data (ICD-9-CM codes) and from MDS; (2) cognitive performance measured using the validated MDS Cognitive Performance Scale; (3) improvement in urinary incontinence measured from MDS; (4) quality of life measured from MDS using 2 validated instruments: Index of Social Engagement and Health Status Index. Covariates included demographic characteristics, baseline continence status (bladder and bowel) and continence management, preexistent urinary tract infections, body mass index, comorbidities, other medication use, cognitive status, and mobility at baseline. These variables were used to calculate the predicted probability (propensity score) of being initiated on a BAM; the resulting propensity scores were used to match new users and nonusers. Outcomes were compared with Cox proportional hazards regression and generalized estimating equations methodology. RESULTS: BAMs were used by 9.8% of the residents 65 years and older admitted for long term care; 44% (1195) were new users. Of these, all but 53 received nonselective immediate release preparations, predominantly oxybutynin chloride (75%). BAM initiation resulted in improved urinary continence status (odds ratio = 1.27, 95% confidence interval [CI] 1.07-1.5) and better social engagement (difference in mean index of social engagement score = 0.2074, 95% CI 0.055-0.3598). The risk of fractures was significantly increased in new users as compared to nonusers (hip fracture: hazard ratio [HR] = 3.67, 95% CI 1.46-9.34; "any" fracture: HR = 2.64, 95% CI 1.37-5.10). The number needed to treat (NNT) to obtain improvement in urinary incontinence after 90 days of treatment (NNT = 32, 95% CI 17-125) was similar to the number needed to harm (NNH) at 90 days in the hip fracture analysis (NNH = 36, 95% CI 12-209). There were no differences in cognitive performance or overall quality of life scores associated with BAM use. CONCLUSION: These results question the continued use of BAMs, particularly immediate-release oxybutynin chloride in elderly nursing home residents.

Duke Scholars

Author Jane Frances Pendergast Biostatistics & Bioinformatics, Division of Biostatistics