ABSTRACT
INTRODUCTION AND HYPOTHESIS: The urogynecology subspecialty relies on appropriate referrals from their referral base. We sought to provide guidance for optimizing appropriate referrals to urogynecology by comparing pre-referral characteristics between appropriate and inappropriate referrals. METHODS: This retrospective cohort study examined predictors of appropriate urogynecology referrals. Appropriateness categorization was based upon pelvic floor disorder (PFD) symptoms and signs provided by the referring provider. Patients with both a PFD symptom and sign were considered "appropriate." Patients with neither a PFD symptom nor sign were considered "inappropriate." PFD symptoms were: vaginal bulge, voiding or defecatory dysfunction. PFD signs were: vaginal vault prolapse, urethral hypermobility, mesh/sling exposure, elevated post-void residual, positive standing stress test, abnormal urinalysis or urine culture-proven infection. Continuous and categorical data were analyzed with ANOVA and chi-square test, respectively. A logistic regression model to predict appropriateness was developed from variables identified from the bivariate analysis. RESULTS: Bivariate predictors of an appropriate referral for 1716 study subjects were older age, prior overactive bladder medication use, MD/DO referrer source and OBGYN, urogynecology or urology referrer specialty. Our logistic regression model correctly classified referrals as appropriate in 93.6% of cases. CONCLUSIONS: Age, anti-cholinergic medication use, referrer source and specialty are pre-initial visit predictors of urogynecology referral appropriateness. The predictor-generated model was successful in predicting referral appropriateness. Potential bias from information transfer issues, lack of pre-referral evaluation and referring provider unfamiliarity with urogynecology are possible reasons for inappropriate referrals and potential areas for improvement.
