Evaluating Linear Parametric Poisson Regression vs Nonparametric Unsupervised Learning in Ulcerative Colitis Data
Li Wai Suen et al1 conducted the PREDICT-UC study to investigate whether early infliximab levels and clearance could predict outcomes following infliximab rescue therapy in patients with acute severe ulcerative colitis. Risk ratios (RR) and confidence intervals (CI) were estimated using modified Poisson regression with cluster-robust standard errors. Multivariable analysis identified independent predictors of day 7 nonresponse, including greater early infliximab clearance (RR, 3.96 per L/d; 95% CI, 1.83–8.58; P = .0005), higher Lichtiger score at infliximab administration (RR, 1.08 per point; 95% CI, 1.03–1.14; P = .0038), and baseline thiopurine use, which reduced nonresponse risk (RR, 0.54; 95% CI, 0.35–0.82; P = .0043).
