Project Summary
There is no cure for the inflammatory bowel diseases (IBD)—Crohn’s disease (CD) and ulcerative colitis (UC). Intermittent use of corticosteroids (CS) was the main treatment strategy for acute exacerbations of IBD until the introduction of biologic therapies targeting tumor necrosis factor alpha (anti-TNF). There are very limited data directly comparing the safety and effectiveness of CS and anti-TNF therapy, particularly among elderly and disabled patients. A single, small clinical trial in a predominantly younger population demonstrated a therapeutic advantage of anti-TNF therapy plus azathioprine versus CS in the first year. However, fear of complications of chronic immunosuppression often deters patients from pursuing anti-TNF therapies.
In this study, we will test the hypothesis that the greater efficacy of anti-TNF therapy results in reduced need for bowel resection surgery, fewer serious infections, and reduced short-term mortality risks and, therefore, has a more favorable benefit-to-harm profile than CS for IBD. In Aim 1, we will quantify patients’ preferences for relevant treatment outcomes by implementing a discrete choice experiment. Participants will choose between two medical treatment options with various attributes (e.g., the risk of serious infection, treatment failure requiring surgery, death, etc.). Using the methods of conjoint analysis, we will compute mean preference weights for each of these attributes. In Aim 2, we will conduct a comparative effectiveness study among Medicare Parts A, B, and D beneficiaries with IBD. We will compare the incidence of severe infection, bowel resection surgery, and death among new users of anti-TNF therapies and CS. We will compute propensity scores (PS) to describe the propensity for treatment with anti-TNF compared to CS, and we will match CS- and anti-TNF–treated patients on the PS. Cox regression will be employed to assess the hazard ratio for each of the outcomes. In Aim 3, we will combine the results from Aims 1 and 2 to compute the relative net benefit of the two treatment strategies after accounting for patient preferences for each outcome. For each patient, we will compute a preference-weighted value for each month of follow-up using an initial treatment carried forward model that allows patients to switch therapies if the first is not effective. We will retain the matching on PS from Aim 2, and the sum of these values will be compared between the groups.
As a complementary approach, we will also develop a discrete event simulation model that compares the two treatment strategies in terms of benefits and harms, using patient preference data from Aim 1 and drawing transition probabilities from Aim 2. This study will use novel methodology to provide a critically needed assessment of the overall risks and benefits of these two treatment strategies, informed by quantitative patient assessments of therapeutic trade-offs. The results should lead to improved outcomes for patients with IBD.
Medication Choices Based on Data, Not Fear: A narrative on this project, which highlights researchers' work into how patients with inflammatory bowel disease weigh treatment benefits and risks—and then are using that information to consider differences between two types of drugs.
Project Details
Information