Design and validation of a patient-reported outcome measure scale in acute pancreatitis: the PAN-PROMISE study.

OBJECTIVE: This study aimed to develop and validate a patient-reported outcome measure (PROM) in acute pancreatitis (AP) as an endpoint centred on the patient. DESIGN: A PROM instrument (PAtieNt-rePoRted OutcoMe scale in acute pancreatItis, an international proSpEctive cohort study, PAN-PROMISE scale) was designed based on the opinion of patients, professionals and an expert panel. The scale was validated in an international multicentre prospective cohort study, describing the severity of AP and quality of life at 15 days after discharge as the main variables for validation. The COSMIN (COnsensus-based Standards for the selection of health status Measurement INstruments) methodology was applied. Both the design and validation stages considered the content and face validity of this new instrument; the metric properties of the different items, reliability (reproducibility and internal consistence), the construct, structural and criterion validity, responsiveness and interpretability of this scale. RESULTS: PAN-PROMISE consists of a seven-item scale based on the symptoms that cause the most discomfort and concern to patients with AP. The validation cohort involved 15 countries, 524 patients. The intensity of symptoms changed from higher values during the first 24 hours to lower values at discharge and 15 days thereafter. Items converged into a unidimensional ordinal scale with good fit indices. Internal consistency and split-half reliability at discharge were adequate. Reproducibility was confirmed using test-retest reliability and comparing the PAN-PROMISE score at discharge and 15 days after discharge. Evidence is also provided for the convergent-discriminant and empirical validity of the scale. CONCLUSION: The PAN-PROMISE scale is a useful tool to be used as an endpoint in clinical trials, and to quantify patient well-being during the hospital admission and follow-up. TRIAL REGISTRATION NUMBER: NCT03650062.