If We Don't Ask, They Won't Tell

Screening for Urinary and Fecal Incontinence by Primary Care Providers

Heidi W. Brown, MD, MAS; Wen Guan, BA; Nicholas B. Schmuhl, PhD; Paul D. Smith, MD; William E. Whitehead, PhD; Rebecca G. Rogers, MD


J Am Board Fam Med. 2018;31(5):774-782. 

In This Article


Study Design, Setting, and Population

We conducted an electronic survey of PCPs within our Midwestern academic medical center. All clinicians (physicians, nurse practitioners, and physician assistants) in the Departments of Internal Medicine, Family Medicine, Obstetrics and Gynecology, and Geriatrics were invited to participate. Automated invitational emails were sent every other week during February to March 2015. Up to 3 emails were sent to each provider to maximize participation. This study was deemed exempt by the University of Wisconsin–Madison Health Sciences Minimal Risk Institutional Review Board (2014-1302).

Survey Instrument

The 20-item anonymous online survey assessed clinicians' knowledge, attitudes, and beliefs related to screening and treatment for UI and FI, as well as information about their level of confidence to treat these conditions, and barriers to and facilitators of screening and treatment. The instrument was pilot tested by 6 physicians and revised before data collection based on their feedback. Minimal demographic data were collected. Most questions were multiple choice, often with a 5-point Likert scale for response options, and some allowed for free-text responses.


Descriptive analyses characterized the sample. Providers who reported screening some, most, or every patient for incontinence were considered screeners; those who reported screening few or no patients were considered nonscreeners. χ 2 testing was used to compare providers' screening rates, beliefs about importance of screening, and confidence to treat UI versus FI. Preferences for screening methods and terminology and populations considered to be at high risk for FI were described. Logistic regression was used to identify factors associated with screening for FI. χ 2 testing was used to compare the proportions of providers who reported various barriers to screening for UI versus FI and to compare their preferences for additional tools to facilitate screening for these conditions. We used STATA 14 (StataCorp, College Station, TX) for all statistical analyses and considered P < .05 as statistically significant.