The prevalence of potentially remediable urinary incontinence in frail older people: a study using the Minimum Data Set.

Objectives: To use the Minimum Data Set (MDS) to describe the frequency and correlates of potentially treatable causes of urinary incontinence among a representative sample of American nursing home residents. To describe current management practices of urinary incontinence in the same population.
Design: Cross-sectional study using the dataset that was part of the Health Care Financing Administration (HCFA) evaluation of the MDS.
Setting: 270 Medicaid-certified nursing homes in 10 states.
Participants: A total of 2014 nursing home residents 60 years or older (mean = 84.3 +/- 8.7), 75.5% women, 81.9% white, who lived in a nursing home during the fall of 1990 were randomly selected to sample a fixed number of residents for each facility based on facility size.
Measurements: Incontinence was defined as the presence of at least two episodes of urinary leakage per week in the previous 2 weeks. Management techniques (toileting, pads/briefs, catheters) were those listed in the MDS. Potentially remediable causes of urinary incontinence available in the MDS were: medications (antipsychotics, antidepressants, and antianxiety/hypnotics); congestive heart failure; diabetes mellitus; pedal edema; delirium; depression; and impairments in activities of daily living (ADLs) (transferring, locomotion, dressing, toileting; bedrails; trunk restraints; and chair restraints).
Results: Forty-nine percent of residents were incontinent. Of these, 84.0% were managed by pads/briefs, 38.7% by scheduled toileting, 3.5% by indwelling catheter, and 1.2% by external catheter. Of the potentially reversible causes, bivariate analysis revealed associations (P < .1) with use of antidepressants, antipsychotics, and antianxiety/hypnotics; delirium; bedrails; trunk restraints; chair restraints; and ADL impairment. Dementia was also associated with incontinence (P < .1). Multivariate analysis revealed that urinary incontinence was independently associated with impairment in ADLs (OR = 4.2; CI = 3.2,5.6), dementia (OR = 2.3;CI = 1.8,3.0), restraints-trunk (OR = 1.7; CI = 1.5,2.0), chair (OR = 1.4; CI = 1.2,1.6), bedrails (OR = 1.3; CI = 1.1,1.5), and use of antianxiety/hypnotic medications (OR = .7;CI = .5,1.0) (all P < .04).
Conclusions: Current management practices for urinary incontinence are inconsistent with advocated guidelines. These data also confirm the association between incontinence and several potentially remediable conditions and suggest that, even in the nursing home setting, urinary incontinence may respond to efforts to improve conditions not directly related to bladder function. This study underscores the need to examine the impact on urinary incontinence of strategies to address such conditions.