A retrospective analysis: the development of patient reported outcome measures for the assessment of Crohn's disease activity.

Abstract

The Crohn's Disease Activity Index (CDAI) is a measure of disease activity based on symptoms, signs and a laboratory test. The US Food and Drug Administration has indicated that patient reported outcomes (PROs) should be the primary outcome in randomised controlled trials for Crohn's disease (CD). As no validated PRO exists for CD, to investigate whether CDAI diary card items could be modified for this purpose. Data from a trial of rifaximin-extended intestinal release were used to identify cut-points for stool frequency, pain and general well-being using receiver operating characteristic curves with CDAI <150 as criterion. The operating properties of 2- and 3-item PRO were evaluated using data from a trial of methotrexate in CD. Regression analysis determined PRO2 and PRO3 scores that correspond to CDAI-defined thresholds of 150, 220 and 450 and changes of 50, 70 and 100 points. Optimum cut-points for CDAI remission were mean daily stool frequency ≤1.5, abdominal pain ≤1, and general well-being score of ≤1 (areas under the ROC curve 0.79, 0.91 and 0.89, respectively). The effect estimates were similar using 2- and 3-item PROs or CDAI. PRO2 and PRO3 values corresponding to CDAI scores of 150, 220 and 450 points were 8, 14, 34 and 13, 22, 53. The corresponding values for CDAI changes of 50, 70 and 100, were 2, 5, 8 and 5, 9, 14. Responsiveness to change was similar for both PROs. Patient reported outcomes derived from CDAI diary items may be appropriate for use in clinical trials for CD.