Retrospective evaluation of operator bias in the performance of high‐resolution anorectal manometry

BACKGROUND: Measurements obtained during high-resolution anorectal manometry (HRM) are subject to operator-patient interactions. For example, standardized enhanced instruction delivered by a single operator in a test-retest fashion did not consistently increase pressures generated during dynamic maneuvers. It is probable that factors other than verbal instruction effect communication during the procedure. To investigate this hypothesis, we retrospectively examined inter-operator variance in HRM results. METHODS: The electronic health records at Mayo Clinic, Rochester, were used to identify patients who had undergone HRM in 2019 and 2020. The analysis focused on constipated patients. The instructions given to the patients they had examined, and the pressure measurements obtained during HRM, by 6 different nurse operators were compared. KEY RESULTS: When performing HRM on their individual patients (range 126-673), the 6 nurses used similar instructions for each of the maneuvers and sensory testing thresholds. The proportion of patients with prolonged balloon expulsion tests and the rectal sensory thresholds were similar among operators. Significant variance was seen in the mean rectoanal pressures at rest, during squeeze, and during dynamic maneuvers. The proportion of patients with manometry results suggestive of a defecatory disorder differed between operators by 18% and 28% in women <50 and >50 years old, respectively. CONCLUSIONS & INFERENCES: Operators obtain significantly different results during HRM despite using similar instructions to patients. Substantial differences in the proportion of patients with manometry findings suggestive of a defecatory disorder among operators may have a significant impact on the diagnoses and therapies offered to constipated patients.