Introduction: Non-invasive uroflowmetry is often used to assess lower urinary tract symptoms (LUTS). However, interpretation and characterization of flow patterns is not well established. Subjective interpretation is the most common approach for flow curve classification, and variable parameters have been reported for uroflowmetry curves. We sought to describe the reliability of uroflowmetry curve interpretation in adult women. Methods: Uroflowmetry studies were obtained in 296 adult women who participated in an observational Prevention of Lower Urinary Tract Symptoms (PLUS) Research Consortium cohort study. Four investigators with expertise in female LUTS and urodynamics reviewed and categorized each tracing for inter-rater reliability and a random subset of 50 tracings were re-reviewed by each investigator for intra-rater reliability. The uroflowmetry tracings were rated using categories of continuous, continuous with fluctuation, interrupted, and prolonged. Tracings could also be categorized as uninterpretable or ambivalent if the voided volume was <150ml or the reviewer could not assign one of the pattern categories. Other parameters included flow rate, voided volume, time to maximum flow, and voiding time. Agreement between raters was summarized with kappa (k) statistics and percentage where at least 3 raters agreed. Results: The mean age of participants was 44.8 ± 18.3years. Participant age categories were: 18-24y 20%; 25-34y 18%; 35-64y 43%; =65y 19%. Nine percent described their race as Asian, 31% Black, 62% White and 95% non-Hispanic origin. The inter-rater reliability was highest for the continuous flow category (k=0.65), 0.41 for continuous with fluctuation, 0.47 for interrupted, and 0.41 for prolonged flow curves. Agreement amongst at least 3 raters occurred in 74.3% uroflow curves (45.9% for continuous, 13.5% for continuous with fluctuation, 3.4% for interrupted, and 4.1% for prolonged). For intra-rater reliability, the mean kappa across the four raters was 0.72 with a range of 0.57-0.85. Conclusions: Currently accepted uroflowmetry curve categories have fair to moderate inter-rater reliability and is lower for flow curves that don’t meet “continuous” criteria. Given the subjective nature of interpreting uroflowmetry data, more consistent and clear parameters may enhance reliability for use as a research parameter. Algorithms that can process numerical and pattern data may also improve the reliability of uroflowmetry studies as a diagnostic and predictive tool for LUTS interventions. SOURCE OF Funding: The content of this [abstract/ presentation/ poster] is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The Prevention of Lower Urinary Tract Symptoms (PLUS) Research Consortium is supported by the National Institutes of Health (NIH) through cooperative agreements (grants U24DK106786, U01DK106853, U01DK106858, U01DK106898, U01DK106893, U01DK106827, U01DK106908, U01DK106892, U01DK126045).
