Validation of a brief inventory for diagnosis and monitoring of symptomatic gastro-oesophageal reflux.

摘要

BACKGROUND: This article presents a brief inventory for the diagnosis and monitoring of GORD symptoms. METHODS: The inventory consists of five items pertaining to different aspects of GORD to be graded for frequency on a 5-point Likert scale. It was validated on a consecutive group of GORD patients diagnosed either by endoscopy (n = 25) or by 24-h ambulatory pH monitoring (acid exposure time > or = 5%; n = 233) and control subjects (n = 300). RESULTS: Each of the inventory items was significantly associated with GORD (P < 0.001). Factor analysis indicated that all items loaded on a single scale with a high reliability (Cronbach alpha 0.88). Each item was weighted by its respective odds ratio in favour of GORD and a total score for the scale was then calculated as the sum of weighted scores on the five individual items. Receiver-operator (ROC) curve analysis implemented on a random sample comprising 67% of the group indicated that a total weighted score of 31.6 (percentage of the maximal possible weighted score) was 91% sensitive and 92% specific in the diagnosis of GORD. This score was then validated on the rest of the sample, where it correctly classified patients and controls with a sensitivity of 89% and specificity of 94%. The score proved stable on repeated administration in controls and in patients with stable symptoms, and decreased by 66% (P < 0.001) after 1 month of treatment with omeprazole in patients who reported symptom relief. CONCLUSIONS: The scale described in this article is a brief, simple and accurate measure, for the diagnosis of GORD as well as for monitoring its symptoms.