| Urinary incontinence (UI) is a common condition, with as many as 13 million individuals affected in the United States. Further knowledge of the predisposing factors associated with UI would allow health providers to identify individuals who may benefit from interventions to prevent, manage, or improve UI. In this retrospective research design, secondary analysis was used to examine associations between the factors under study and UI severity as well as a change in UI severity. A sample of 218 older women was initially randomized into either a behavioral management or a control group.; The dependent variable, UI, was operationally defined by severity, as measured by grams of urine loss per 24 hours and by a bladder diary used to assess episodes of urine loss. The independent variables were specific demographic factors, physical health, and health perception factors. Regression methods were used to select models that could be used to predict urine severity at 3 time periods, baseline, 6 months, and 12 months. A mixed-model analysis was used to supplement findings and to confirm results over the 3 time periods.; Model selection at baseline indicated that the strongest predictor of urine loss and episodes was self-perceived severity, followed by ADL index, education, and age. For predictors of baseline episodes, the strongest predictor was baseline severity, followed by the psychosocial subscale and the impact of mouth/eating disorders. At the 6- and 12-month periods, baseline severity was the strongest predictor for both urine loss and episodes. For urine loss in grams, inclusion in the treatment group, age, stroke, and self-health report were also significant predictors, and, for episodes, treatment group, age, treatment/hypertension interaction, and BMI were significant. For predictors of change in severity in urine loss at 6 months, baseline urine loss was again the strong predictor, followed by treatment group, age, and self-health report. The strongest predictor of change in episodes at 6 months was baseline episodes, followed by treatment and age.; The strongest predictor of urine loss at 12 months was 6-month urine loss. For predictors of change in urine loss from baseline to 12 months, perceived health, followed by Treatment x Baseline Urine Loss Interaction were significant. For change in urine loss from 6 to 12 months, the significant predictors included 6-month urine loss, followed by parity. For episodes at 12 months, the strongest predictor was 6-month episodes, and for change in episodes from baseline to 12 months, the treatment group, followed by baseline episodes, age, and perceived health were significant. For change in episodes from 6 to 12 months, 6-month episodes were the strongest predictors. There was a significant treatment effect at all time periods, except for the 12-month episodes and change in urine loss and episodes (6–12 months) outcomes. The mixed model confirmed support for all variables, but the Treatment x Time Interaction at 6 and 12 months was the strongest influence on both urine loss and episodes.; From these findings, the researcher recommended that nursing professionals plan interventions around assessments and focus on the identified factors. |
