Skip to main content
. 2022 Nov 25;17(1):108–118. doi: 10.5009/gnl220081

Table 4.

Univariable and Multivariable Analyses of Predictors Associated with Relapse in Patients with Inflammatory Bowel Disease

Variable Univariable analysis Multivariable analysis
HR (95% CI) p-value HR (95% CI) p-value
Age (>40 yr) 0.617 (0.327–1.163) 0.136 0.987 (0.965–1.009) 0.247
Sex (female) 1.065 (0.539–2.102) 0.857
Current smoker 0.982 (0.456–2.112) 0.962
Disease duration (>2 yr) 1.434 (0.768–2.678) 0.257
History of abdominal surgery 1.140 (0.406–3.201) 0.804
Enterotype based on EVs
Type 1 Reference
Type 2 0.600 (0.289–1.247) 0.171 0.772 (0.335–1.777) 0.542
C-reactive protein (>0.5 mg/dL) 2.801 (1.485–5.283) 0.001* 1.379 (0.653–2.910) 0.400
Fecal calprotectin (>250 mg/kg) 3.418 (1.828–6.389) <0.001* 3.147 (1.545–6.408) 0.002*
Medication use
Steroids 3.033 (1.500–6.135) 0.002* 1.827 (0.784–4.260) 0.163
Immune modulators 2.227 (1.208–4.106) 0.010* 0.892 (0.413–1.927) 0.772
TNF-α inhibitors 2.226 (1.123–4.414) 0.022* 1.777 (0.786–4.015) 0.167

HR, hazard ratio; CI, confidence interval; EV, extracellular vesicle; TNF, tumor necrosis factor.

p-values were calculated using Cox regression. Relapse was defined as a composite outcome of (1) new use of steroids, immunomodulators, and biologics; (2) a visit to an emergency department; (3) hospitalization; or (4) abdominal surgery.*p<0.05.