Search in the Abstract Database

Abstracts Search 2017

* = Presenting author

P359 Correlation between physician and patient disease assessments in ulcerative colitis: baseline data from the ICONIC study of 1816 patients in 33 countries

Ghosh S.*1, Peyrin-Biroulet L.2, Sensky T.3, Casellas F.4, O'Shea C.5, Pappalardo B.6

1University of Birmingham, Birmingham, United Kingdom 2University of Lorraine, Nancy, France 3Imperial College, London, United Kingdom 4Crohn-Colitis Care Unit (UACC), Hospital Universitari Vall d'Hebron, Barcelona, Spain 5AbbVie Ltd, Dublin, Ireland 6AbbVie Inc., North Chicago, IL, United States

Background

ICONIC is the largest prospective multi-country (n=33) observational study, assessing burden in adult ulcerative colitis (UC) patients (pts) under routine care. Both pt & physician assessments of disease severity, activity & life impact will be captured at 6 month intervals through 2 years. This analysis will evaluate baseline (BL) demographics, clinical characteristics & the extent of agreement between pts & physicians in measures of disease activity.

Methods

Adult pts with early UC (diagnosed ≤36 months) were enrolled irrespective of disease severity or treatment. Pt self-assessments include: disease severity, Pictorial Representation of Illness & Self-Measure (PRISM, [a tool assessing perception of disease-associated suffering]), Patient Health Questionnaire-9 (PHQ-9), Short Inflammatory Bowel Disease Questionnaire (SIBDQ) & pt-modified Simple Clinical Colitis Activity Index (P-SCCAI). Physician assessments include: clinical parameters, PRISM, & SCCAI. Correlation between PRISM & SIBDQ, PHQ-9, & SCCAI were evaluated. BL characteristics were obtained from observed data by calculating means. Mean differences between pt & physican measures were calculated using a one-sample t-test. Correlation analyses were performed by kappa statistics & Pearson correlation.

Results

BL demographic & clinical characteristics of 1816 UC pts enrolled in ICONIC are summarized in Table 1. At BL, pt/physician PRISM was moderately correlated with SIBDQ, PHQ-9, P-SCCAI, or SCCAI (Table 2). For 1796 pts with self & physician assessments, most showed agreement on disease severity (concordant pairs: mild 60.3%, moderate 60.6%, severe 56.0%) (κ=0.43, 95% CI 0.40–0.47, p<0.0001). Although P-SCCAI & physician SCCAI mean values±SD differed (4.2±3.6 & 3.0±3.0 [p<0.0001]), the measures were highly correlated (Table 2). For pt/physician PRISM assessments, mean scores (4.0±2.5 & 4.3±2.4, respectively) differed (p<0.0001) & were moderately correlated (Table 2).

Conclusion

Although the majority of pts received therapy & reported improved disease activity since diagnosis, nearly half had moderate-to-severe UC. PRISM, used for the first time in UC, was moderately correlated with disease-specific measures (SIBDQ & SCCAI) & a general depression assessment (PHQ-9). Alignment between pts & physicians on disease activity/severity may depend on the instrument used.