Abstract
Introduction: There are currently no validated tools that accurately define overall disease severity in inflammatory bowel disease (IBD). The disease severity index (DSI) for Crohn’s disease (CD) and ulcerative colitis (UC) is a novel tool measuring the disease burden, disease phenotype and impact of the disease on the patient and has previously been associated with psychological distress and a complicated IBD course. This prospective multicentre study aimed to validate the DSI against indices of endoscopic activity, patient reported symptoms, quality of life (QoL) and disability.
Methods: Patients were recruited at the time of routine clinical assessment. Inter-rater reliability was performed on a subset of participants using intraclass correlation coefficients (ICC) and weighted Kappa. Cronbach’s alpha measured the internal consistency of each DSI dimension to the whole index. Construct validity was assessed using Spearman rank correlations against endoscopic indices (Simple endoscopic score for CD (SES-CD), Mayo endoscopic score for UC) performed within 1 month of DSI scoring. Validity was also assessed against patient reported symptoms (CD activity index (CDAI), partial Mayo score), QoL (IBDQ-32), and disability (IBD disability index (IBD-DI)) performed at the time of DSI scoring. Principal components analyses (PCA) reduced DSI dimensions to core domains describing disease severity.
Results: In total 369 participants were included for analysis (CD, n=230; female, n=194; mean =46 years (SD, 15), median disease duration=11 years (IQR, 5-21)). The DSI showed good inter-rater reliability for CD (ICC 0.93, =65) and UC (ICC 0.97, =33). DSI components demonstrated fair to near perfect inter-rater agreement and showed internal consistency with the total index (Table 1). The DSI was significantly associated with endoscopic activity (CDn=141, r=0.67, p<0.001; UCn=105, =0.80, p<0.001), symptoms (CDn=159, =0.66, <0.001; UCn=132, r=0.58, p<0.001), QoL (CDn=198, r=-0.55, p<0.001; UCn=128, r=-0.68, p<0.001) and disability (CDn=83, r=-0.67, p<0.001; UCn=52, r=-0.74, p<0.001). CD-DSI dimensions reduced to 3 components representing the inflammatory burden, impact of disease on the patient and disease phenotype (Figure 1). For UC, DSI dimensions reflecting the inflammatory burden and patient reported outcome subscales were significantly correlated (r=0.58, p<0.001) and so the UC-DSI was reduced to 2 components - the disease burden and disease phenotype (Figure 1).
Conclusions: The DSI is a valid and reliable instrument that reflects the total burden of disease in IBD. Patient reported outcome measures in the DSI are strongly associated with inflammatory activity in UC but not CD. The DSI can be calculated during IBD assessment to reflect overall disease severity.