Validation of an obstetric fistula screening questionnaire: a case–control study with clinical examination

Background: Obstetric fistula (OF) is a significant cause of maternal morbidity in lower resource settings where women experience obstructed labor without timely access to skilled obstetric care. The true prevalence of OF is unknown; however, it is estimated to affect 2 to 3.5 million women globally. The Demographic and Health Surveys’ (DHS) Fistula Module includes the OF symptom questions most frequently used for prevalence estimates, but these questions have not been validated. The aim of this study is to validate a symptom-based screening questionnaire for OF, including a question in the DHS’ Fistula Module. Methods: With an international panel of fistula surgeons, we developed and face-validated a screening questionnaire that assessed for symptoms of lower urinary tract fistula (LUTF) and lower gastrointestinal tract fistula (LGTF), as well as urinary and fecal incontinence (UI, FI). We evaluated the discriminative ability of the questionnaire using a case–control study design in a 1:2:2 ratio: cases were parous women with fistula confirmed on examination, controls included parous women without fistula on examination, with and without UI symptoms. All women underwent screening for fistula symptoms and a physical examination, with examiners blinded to screening results. Results: Of the 367 Rwandan women who completed the questionnaires and underwent clinical examination, 59 women had LUTFs and 34 had LGTFs, 274 women were classified as controls with and without symptoms of UI. All LUTF screening questions performed well, including the DHS fistula question. The combination of two LUTF screening questions had the highest sensitivity (100%; 95% CI 94%, 100%), specificity (96%; 95% CI 93%, 98%), and area under the curve (AUC) (0.98). The combination of a LGTF screening question and FI question demonstrated the highest sensitivity (97%; 95% CI 85%, 100%), specificity (98%; 95% CI 95%, 99%) and AUC (0.98). Conclusions: Our OF screening questionnaire, including the DHS fistula question, demonstrated high sensitivities, specificities, and AUC.