The Disease Severity Index for Inflammatory Bowel Disease Is a Valid Instrument that Predicts Complicated Disease

Integrating the Patient Perspective to Validate a Measure of Disease Severity in Inflammatory Bowel Disease: Online Survey of Patients and Their Physicians

BACKGROUND

The patient perspective is essential for assessing disease severity, but it is not always adequately considered. We describe how a comprehensive clinical disease severity index (DSI) for inflammatory bowel disease (IBD) correlates with patient global self-assessment (PGSA).

METHODS

In an individually linked parallel online survey, physicians provided the DSI, and patients provided self-assessed severity using a global question and visual analog scale (0-100) (PGSA). Mean DSI values by PGSA were calculated with 95% confidence intervals. Pearson correlation (r) and the intraclass correlation coefficient were calculated for PGSA vs DSI. Positive predictive values for identifying severe disease with PGSA categories as a reference were based on a threshold >22 points.

RESULTS

The primary analysis included 89 pairs (46 Crohn's disease [CD], 43 ulcerative colitis [UC]) with strict criteria and 147 pairs when less stringent. Common reasons for exclusion were missing values for albumin or colonoscopy. Mean DSI values showed no clear trend with increasing PGSA in CD but good discrimination between moderate, severe, and very severe PGSA in UC. For PGSA on the visual analog scale, r was 0.54 for CD and 0.59 for UC (difference in means: CD 27.7, UC 13.8; intraclass correlation coefficient: CD 0.48, UC 0.58). A high DSI predicted severe disease in 76.2% of CD and 65.2% of UC.

CONCLUSIONS

The DSI showed good discrimination for patient-reported disease severity in UC but performed unsatisfactorily in CD. Correlations were moderate. Further refinement of the DSI is suggested to better reflect the patient perspective.