Author information

1 Division of Gastroenterology and Hepatology, University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA.

2 Center for Clinical Epidemiology and Biostatistics, University of Pennsylvania, Philadelphia, Pennsylvania, USA.

3 Duke Clinical Research Institute, Duke University, Durham North Carolina, USA.

4 Division of Gastroenterology, University of Pennsylvania, Philadelphia, Pennsylvania, USA.

5 Department of Biostatistics and Epidemiology, University of Pennsylvania, Philadelphia, Pennsylvania, USA.

6 Abramson Cancer Center, University of Pennsylvania, Philadelphia, Pennsylvania, USA.

7 Division of Clinical Immunology and Rheumatology, University of Alabama at Birmingham, Birmingham, Alabama, USA.

8 Department of Epidemiology, University of Alabama at Birmingham, Birmingham, Alabama, USA.

Abstract

BACKGROUND: Corticosteroids (CS) and anti-TNF drugs are used to treat Crohn's disease (CD). In this study, we assessed the net health benefit of initiating anti-TNF therapy relative to additional CS use in CD using a novel combination of a retrospective cohort study and a simulation model.

METHODS: Using Medicaid data from 2001 to 2005 and Medicare data from 2006 to 2013, beneficiaries were identified with CD and CS use who subsequently received either an anti-TNF or reached a cumulative dose of >3000 mg CS during the year. By using overall and latent class-specific remission-time equivalent (RTE) estimates derived from discrete-choice experiments, mean 12-month cumulative RTEs were calculated after propensity score adjustment for baseline characteristics. A Markov model was constructed using transition probabilities derived from the retrospective cohort to perform additional sensitivity analyses of RTE estimates, analytic assumptions, and transition probabilities. Cumulative RTEs were calculated via Monte Carlo simulation in this model.

RESULTS: In the retrospective cohort, 1563 new anti-TNF initiators and 1563 propensity score-matched prolonged CS users were identified. Anti-TNF use was associated with greater mean RTEs at the end of 1 year (5.34 vs 4.54, incremental benefit: 0.79; 95% CI, 0.53-1.07). This benefit persisted in all latent classes. In the Markov model, anti-TNF therapy was the preferred strategy, and the results were robust in multiple sensitivity analysis and latent class analysis.

CONCLUSIONS: In both a retrospective cohort study and a simulation model, anti-TNF use was associated with improved quality of life, measured as RTEs, when compared with continued CS utilization for CD.