Prevalence of urinary incontinence in middle-aged and older women: A survey-based methodological experiment

Objective: To determine whether self-reports of urinary incontinence are influenced by the context in which survey questions are presented. Variation in the extent to which surveys promote complete and accurate reporting may help explain the wide range of published prevalence estimates. Methods: Respondents to a nationally representative telephone survey were randomly assigned one of two questionnaire forms. Both forms contained the same incontinence question. One form included an introduction and follow-up probe, which acknowledged the embarrassment of discussing incontinence and stressed the importance of thorough reporting. Results: Use of the introduction and probe produced a significantly higher prevalence rate, with the effect varying by respondent's age. Discussion: The prevalence of incontinence is likely to be underestimated unless studies attend to the possibility of biased reporting. The finding of a form by age interaction suggests the importance of survey context in studies of risk factors, as well as in prevalence studies.