Patient Preference and Adherence (Oct 2017)
Education and patient preferences for treating type 2 diabetes: a stratified discrete-choice experiment
Abstract
Ellen M Janssen,1 Daniel R Longo,2 Joan K Bardsley,3 John FP Bridges1 1Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, 2Department of Family Medicine and Population Health, Virginia Commonwealth University, Richmond, VA, 3MedStar Health Research Institute and MedStar Nursing, Hyattsville, MD, USA Purpose: Diabetes is a chronic condition that is more prevalent among people with lower educational attainment. This study assessed the treatment preferences of patients with type 2 diabetes by educational attainment. Methods: Patients with type 2 diabetes were recruited from a national online panel in the US. Treatment preferences were assessed using a discrete-choice experiment. Participants completed 16 choice tasks in which they compared pairs of treatment profiles composed of six attributes: A1c decrease, stable blood glucose, low blood glucose, nausea, treatment burden, and out-of-pocket cost. Choice models and willingness-to-pay (WTP) estimates were estimated using a conditional logit model and were stratified by educational status. Results: A total of 231 participants with a high school diploma or less education, 156 participants with some college education, and 165 participants with a college degree or more completed the survey. Participants with a college degree or more education were willing to pay more for A1c decreases ($58.84, standard error [SE]: 10.6) than participants who had completed some college ($28.47, SE: 5.53) or high school or less ($17.56, SE: 3.55) (p≤0.01). People with a college education were willing to pay more than people with high school or less to avoid nausea, low blood glucose events during the day/night, or two pills per day. Conclusion: WTP for aspects of diabetes medication differed for people with a college education or more and a high school education or less. Advanced statistical methods might overcome limitations of stratification and advance understanding of preference heterogeneity for use in patient-centered benefit–risk assessments and personalized care approaches. Keywords: preference heterogeneity, stated-preference methods, preference heterogeneity, willingness-to-pay, choice experiment, educational attainment
