OBJECTIVE - This study aimed to establish the extent to which care-seeking for urinary incontinence is a function of symptom bother; and to identify bother-score thresholds that predict care-seeking in the first 2 decades after delivery.
METHODS - In this longitudinal cohort, women were assessed annually for symptom bother related to stress urinary incontinence (SUI) and overactive bladder (OAB), as well as for recent episodes of care-seeking for urinary symptoms. Because the goal was to model care-seeking as a function of the woman's characteristics at her prior visit, women who completed 2 or more consecutive visits were included. The population was randomly divided into "training" (model development) and "testing" (model validation) sets. The predictive model was developed in the training set. For SUI and OAB bother scores, we identified thresholds to define statistically distinct probabilities of care-seeking. A multivariable model was created, including SUI and OAB bother categories as well as characteristics associated with care seeking at the P < 0.05 level. The resultant prediction model was then applied to the "testing set"; predicted and observed care-seeking frequencies were compared.
RESULTS - Care-seeking was strongly associated with SUI and OAB bother. We defined 3 categories for OAB score and 4 categories for SUI score. The resulting 12 risk categories were then collapsed into 5 distinct risk-groups. These groups accurately predicted care-seeking in the testing set (area under the receiver operating curve, 0.760; 95% confidence interval, 0.713-0.807). Inclusion of other risk factors did not improve the model.
CONCLUSIONS - Symptom bother is a strong determinant of care-seeking in the first 2 decades after delivery. These results define 5 ordinal categories that predict seeking care for urinary symptoms in a community population.
Female pelvic medicine & reconstructive surgery. 2016 Mar 03 [Epub ahead of print]
Jocelyn Fitzgerald, Christopher Pierce, Joann Nugent, Joan Blomquist, Victoria L Handa
From the *Department of Gynecology and Obstetrics, Johns Hopkins School of Medicine; †Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health; and ‡Department of Gynecology, Greater Baltimore Medical Center, Baltimore, MD.