Development and validation of a symptom-based activity index for adults with eosinophilic esophagitis

Abstract

Background & aims: Standardized instruments are needed to assess the activity of eosinophilic esophagitis (EoE) and to provide end points for clinical trials and observational studies. We aimed to develop and validate a patient-reported outcome (PRO) instrument and score, based on items that could account for variations in patient assessments of disease severity. We also evaluated relationships between patient assessment of disease severity and EoE-associated endoscopic, histologic, and laboratory findings.

Methods: We collected information from 186 patients with EoE in Switzerland and the United States (69.4% male; median age, 43 y) via surveys (n = 135), focus groups (n = 27), and semistructured interviews (n = 24). Items were generated for the instruments to assess biologic activity based on physician input. Linear regression was used to quantify the extent to which variations in patient-reported disease characteristics could account for variations in patient assessment of EoE severity. The PRO instrument was used prospectively in 153 adult patients with EoE (72.5% male; median age, 38 y), and validated in an independent group of 120 patients with EoE (60.8% male; median age, 40.5 y).

Results: Seven PRO factors that are used to assess characteristics of dysphagia, behavioral adaptations to living with dysphagia, and pain while swallowing accounted for 67% of the variation in patient assessment of disease severity. Based on statistical consideration and patient input, a 7-day recall period was selected. Highly active EoE, based on endoscopic and histologic findings, was associated with an increase in patient-assessed disease severity. In the validation study, the mean difference between patient assessment of EoE severity (range, 0-10) and PRO score (range, 0-8.52) was 0.15.

Conclusions: We developed and validated an EoE scoring system based on 7 PRO items that assess symptoms over a 7-day recall period. Clinicaltrials.gov number: NCT00939263.

Keywords: Disease Activity Measurement; Esophagus; Marker; Patient-Reported Outcome.

Figure 1

Conceptual framework for development of EEsAI instruments. The components of the flow chart outlined with a dashed line, such as EndoFlip or mucosal biomarkers, were not, as of yet, evaluated for the purposes of the EEsAI study. Abbreviations: EndoFlip®, Endolumenal Functional Lumen Imaging Probe.

Figure 2

The relationship between endoscopic / histologic activity and patient-assessed EoE severity. The box contains the 25^th – 75^th percentile of values, the horizontal line in the middle of the box represents the median.

Figure 3

A. The correlation plot between the EEsAI PRO score and the PatGA in the validation group. B. The Bland-Altman plot for the agreement between the EEsAI PRO score and the PatGA in the validation group. The grey box indicates the 95 % limits of agreement. Abbreviation: PatGA, patient global assessment; EEsAI, eosinophilic esophagitis activity index; PRO, patient-reported outcome.