Demographic, medical, and physiologic predictors of behavioral treatment and
pharmacotherapy success would be useful in selecting treatments for specific patients with
urinary incontinence based on their histories, physical examinations, and urodynamic profiles. The author performed a systematic review of clinical trials of behavioral treatment or
pharmacotherapy for
urinary incontinence. Most postulated predictors (age, type and duration of incontinence, medications including
diuretics and
estrogen, obstetric history, physical examination, and urodynamic findings) were not predictive of treatment outcomes. For behavioral
therapy, male gender predicted worse outcomes in 1 study, but it was not a predictor in 2 other studies. Greater severity of incontinence predicted positive outcomes in 2 studies, negative outcomes in 3 studies, and had no predictive value in 5 studies. Prior treatment for incontinence predicted poorer outcomes in 2 studies of
urge incontinence but was not predictive in a study of stress incontinence. Prior surgical treatment predicted better outcomes in 1 study of
urge incontinence in women but was unrelated in 4 studies. Male gender, depression, or the use of
assistive devices for ambulation predicted poorer outcomes in homebound older persons. For
pharmacotherapy of
urge incontinence, older age, female gender, and greater incontinence severity were associated with poorer outcomes in 1 study. Age was unassociated with outcomes in another study. Thus, the literature on predictors of outcomes of behavioral and
drug treatment for
urinary incontinence is inconsistent and does not provide guidelines for treatment selection. More studies, with large samples, that use multivariate regression analysis to examine predictors of outcomes are needed.