Is Drug Therapy for Urinary Incontinence Used Optimally in Long-Term Care Facilities?

Abstract

To examine the management of urinary incontinence (UI) among nursing home (NH) residents in the United States, particularly drug therapy for UI in those who may be suitable candidates for such treatment based on their functional status. Retrospective analysis of admissions (between January 2, 2002, and December 31, 2003) to a total of 373 skilled nursing facilities and assisted living centers operated by a single provider of long-term care. Residents identified as incontinent according to at least one Minimum Data Set (MDS) assessment during their NH stay who had adequate mobility and/or cognitive ability to toilet, as determined by a toileting score of < or =2 on the 5-point MDS scale, and/or a score of < or =3 on the 7-point scale, the Cognitive Performance Score (CPS). MDS assessments for individual residents were obtained from a central database linked to a physician order database that captured the dose, frequency, and start and stop dates of all medications prescribed. Residents were stratified into treated or untreated groups according to whether or not they were prescribed medications used to treat UI (including tolterodine, oxybutynin [oral and transdermal patch formulations], desmopressin, and flavoxate). During the study period, there were 58,216 admissions to the 373 participating facilities; 31,219 (54%) were identified as incontinent of urine on the MDS. The study population comprised 25,140 NH residents who met MDS criteria for UI (80.5% of the total identified as incontinent of urine) and who had adequate mobility to toilet and/or did not have severe cognitive impairment. They were typically over 60 years of age (95.2%), female (65.1%), and frequently or completely incontinent (63.1%). Nonpharmacologic treatment (as recorded in the MDS) included pads/briefs (76.8%), scheduled toileting (31.9%), and/or bladder retraining (2.8%). Only 1752 (7.0%) of eligible residents received medication for their UI. Using a multivariate analysis, factors that were significantly associated with drug treatment for UI included female gender, frequent or complete urinary incontinence (MDS category 3-4), constipation, and use of incontinence appliances/programs and walking aids. Older residents and those with severe cognitive impairment were less likely to receive drug therapy. Only a small proportion of incontinent NH residents with mobility and cognitive function potentially suitable for specific treatment for incontinence receives drug therapy for their condition. Further research is needed to determine whether low drug use reflects an unmet need for treating UI, or appropriate prescribing practices based on the multiple and interacting factors that influence decisions on drug therapy in the NH population.