Patient preferences for surgical versus medical therapy for ulcerative colitis

Abstract

Background: Therapy options for mesalamine-refractory ulcerative colitis (UC) include immunosuppressive medications or surgery. Chronic immunosuppressive therapy increases risks of infection and cancer, whereas surgery produces a permanent change in bowel function. We sought to quantify the willingness of patients with UC to accept the risks of chronic immunosuppression to avoid colectomy.

Methods: We conducted a state-of-the-art discrete-choice experiment among 293 patients with UC who were offered a choice of medication or surgical treatments with different features. Random parameters logit was used to estimate patients' willingness to accept trade-offs among treatment features in selecting surgery versus medical treatment.

Results: A desire to avoid surgery and the surgery type (ostomy versus J-pouch) influenced patients' choices more than a specified range of 10-year mortality risks from lymphoma or infection, or disease activity (mild versus remission). To avoid an ostomy, patients were willing to accept a >5% 10-year risk of dying from lymphoma or infection from medical therapy, regardless of medication efficacy. However, data on patients' stated choice indicated perceived equivalence between J-pouch surgery and incompletely effective medical therapy. Patient characteristics and disease history influenced patients' preferences regarding surgery versus medical therapy.

Conclusions: Patients with UC are willing to accept relatively high risks of fatal complications from medical therapy to avoid a permanent ostomy and to achieve durable clinical remission. However, patients view J-pouch surgery, but not permanent ileostomy, as an acceptable therapy for refractory UC in which medical therapy is unable to induce a durable remission.

FIGURE 1

Example of conjoint scenario comparing medication and surgical therapy for UC flare.

FIGURE 2

Identification of final patient population.

FIGURE 3

A, Preference weights for varying levels of mortality from lymphoma or serious infection over 10 years. B, Relative preference utility for risk attributes and medication efficacy. The vertical axis shows relative utility/satisfaction (scores scaled between 0 and 100, with 0 corresponding to the smallest satisfaction score across treatment attributes and 100 corresponding to the largest satisfaction score) and the horizontal axis the varying levels of the attributes. Illustrated satisfaction scores at each level of risk take into account the significantly increased preference utility associated with selecting medical therapy in preference of any type of surgery (e.g., satisfaction scores for J-pouch without incontinence are conditional on selecting surgery).

FIGURE 4

Subgroup analysis of preference utility ratios. Labels on the horizontal axis indicate group A and group B; ratios are interpreted as (attribute importance of group A)/(attribute importance of group B). Numbers <1 and shaded in blue indicates that group A views the option as less important than group B. Numbers >1 and shaded in green indicate that group A cares relatively more than group B about the attribute. Color saturation indicates the distance from 1 relative to other ratios in the figure. Numbers in red indicate a statistically significant difference (P < 0.05). Thus, an example would be for those patients who have had a history of surgery versus those who have not, these patients are significantly less concerned with having surgery but are over 3 times more concerned with having a disease remission.