Clinician-reported Gloucester Comfort Scale scores underestimate patient discomfort and pain during colonoscopy: insights from comparison with a patient-reported experience measure

Background Patient experience is a fundamental element of colonoscopy. The Gloucester Comfort Scale (GCS) is used by clinicians to report patient comfort. However, insights regarding the extent to which clinician-reported GCS scores represent the patient's experience are lacking. We assessed the level of agreement between clinician-reported GCS scores and patient-reported discomfort and pain. Methods Consecutive patients undergoing colonoscopy at two Dutch endoscopy clinics were included. Patient comfort during colonoscopy was reported using the GCS (1-5 scale). Patients' colonoscopy experiences were assessed using the Newcastle ENDOPREM, a validated endoscopy patient-reported experience measure (PREM). Patients reported both discomfort and pain levels experienced during colonoscopy on a 1-5 scale. Levels of agreement were assessed using Cohen's kappa statistic. Results For 243 included patients, the GCS score was higher than the PREM discomfort score in 52 patients (21%) and lower in 72 (30%). GCS score was higher than the PREM pain score in 39 patients (16%) and lower in 71 (29%). Moderate-to-severe discomfort and pain (scores >= 3) were reported by 53 patients (22%) for discomfort and 60 patients (25%) for pain. For these patients, the GCS underestimated discomfort and pain levels in almost all cases (discomfort 49/53 [92%], pain 54/60 [90%]). Agreement between GCS scores and PREM discomfort and pain scores were minimal (Cohen's kappa 0.34) and weak (Cohen's kappa 0.47), respectively. Conclusions Clinician-reported GCS scores frequently underestimated the level of discomfort and pain reported by patients. For accurate monitoring of patients' colonoscopy experiences, the use of PREMs should be considered.