Objectives: Questions have been raised about the discriminative value of the three laboratory items (hematocrit, erythrocyte sedimentation rate, and albumin) and three physical items (height, perirectal disease, and extraintestinal manifestations) included in the Pediatric Crohn's Disease Activity Index (PCDAI). The aim of this study was to analyze the value of these six "criticized" items to the discriminative properties of the PCDAI.
Methods: Data from 71 children with Crohn's disease visiting an outpatient clinic were analyzed. Physician global assessment of disease activity was used as the gold standard. A "basic index" was calculated by subtracting the score of the six criticized items from the score of the PCDAI calculated in the standard fashion. Multivariate logistic regression procedures identified which items significantly contributed to the "basic index". Receiver operating characteristic curves were produced comparing the standard PCDAI score to the "basic index" and a new "clinical index" which included only the criticized items truly contributing to the discriminatory ability of the "basic index".
Results: Logistic regression models identified only perirectal disease as contributing to the discriminative abilities of the basic index. The clinical index therefore consists of the three history items (abdominal pain, number of liquid stools, and general well-being), three physical examination items (weight loss, abdominal examination, and perirectal disease) and no laboratory tests. The clinical index had an area under the curve not significantly inferior to that of the original PCDAI (0.93 [95% confidence interval, 0.89-0.99] vs. 0.96 [95% confidence interval, 0.92-0.99]).
Conclusions: A clinical index consisting of three history items and three physical examination items has an accuracy equal to the standard PCDAI in distinguishing children with disease in remission from those with a relapse.