Patients’ Preferences for Second-line Pharmacological Agents in Type 2 Diabetes

Abstract

The second-line antihyperglycemic agents (AHAs) for Type 2 Diabetes Mellitus (T2DM) have a wide variety of treatment attributes, including treatment benefits, side effects, and various treatment processes, such as dosage form, mode of administration, etc. Identifying what T2DM patients preferred while selecting the second-line AHAs, including sodium-glucose cotransporter-2 inhibitors (SGLT-2is) and glucagon-like peptide-1 receptor agonists (GLP-1 RAs), is important. The purposes of this thesis were to rank the importance of the attributes of second-line AHAs and determine the patients’ preferences for SGLT-2is and GLP-1 RAs. The thesis was conducted in two steps using two stated preference methods. First, a cross-sectional, web-based survey was used to rank the important attributes of second-line AHAs using the best-worst scaling (BWS) method. A balanced incomplete block design (BIBD) was used to generate choice sets. Patients diagnosed with T2DM, aged 19 years or older, and were proficient in English were recruited through QualtricsXM. The BWS data were analyzed using the count analysis method, and the standardized BWS score was calculated for each attribute. Second, another cross-sectional, web-based survey was used to determine the patients’ preferences for SGLT-2is and GLP-1 RAs using a discrete choice experiment (DCE). Six attributes (i.e., how do you take the medication, the chance of reaching target HbA1c (long-term blood glucose level) in 6 months, % reduction in the risk of major adverse cardiovascular events (i.e., heart attack, stroke, and death due to cardiovascular diseases), the chance of gastrointestinal side effects (i.e., nausea, vomiting, and diarrhea), the chance of genital infection and out-of-pocket cost per month) and their level for SGLT-2is and GLP-1 RAs from the literature review were confirmed and consolidated with the results from in-depth interviews with T2DM patients, and the BWS results. An orthogonal and balanced design was used to draw a subset of all combinations randomly. A total of 36 choice sets were generated and divided into four blocks. Each block comprised nine choice sets that were used to develop a self-administered questionnaire survey. Patients diagnosed with T2DM, 19 years or older, and were proficient in English were recruited through QualtricsXM. Multinomial logit (MNL) and mixed logit (ML) analyses were conducted to examine patients’ preferences and preference heterogeneity. Subgroup analyses were used to explore the preference heterogeneity across three variables, including gender, T2DM experience, and SGLT-2is or GLP-1 RA experience. The patients’ willingness-to-pay (WTP) for all attributes and available SGLT-2is and GLP-1 RAs in the market were calculated from the estimated coefficients in the ML model. A total of 99 T2DM patients completed the BWS survey questionnaire. Among 16 important attributes of second-line AHAs, reducing blood glucose and reducing the risk of cardiovascular diseases had the highest standardized BWS scores of 0.48 and 0.39, respectively. While the reaction at the injection site and route of administration attributes had the lowest standardized BWS scores of -0.52 and -0.31, respectively. Thus, T2DM patients ranked reducing blood glucose and reducing cardiovascular diseases as the most important attributes while selecting second-line medications. A total of 176 T2DM patients completed the DCE survey questionnaire. The MNL analyses showed that reaching target Hb1c, reducing the risk of a major adverse cardiovascular event (MACE), gastrointestinal side-effect, genital infection, out-of-pocket cost, and all levels of route and frequency of administration (except for injectable, once a day) were significant attributes while selecting SGLT-2is and GLP-1 RAs. Similarly, the ML analyses confirmed the significance for all attributes, except for the two levels, injectable, once a day and injectable, once a week, of the route and frequency of administration attribute while selecting SGLT-2is and GLP-1 RAs. The ML model also showed preference heterogeneity for all attributes. Subgroup analyses indicated that treatment-related GI side effects and genital infection were not important for male T2DM patients or patients with lesser experience with T2DM, or patients who had experienced SGLT-2is or GLP-1 RAs. T2DM patients were willing to pay approximately $6 per month and $4 per month for a 1% increase in the chance of reaching the HbA1c target and for a 1% increase in reducing the risk of MACE, respectively. Similarly, they were willing to pay approximately $8 and $12 to avoid 1% of GI side effects and 1% of genital infection, respectively. Among four different ways of taking medications, the T2DM patients were willing to pay the highest of $486 per month for oral, once-a-day medication and the lowest of $176 per month for injectable, twice-a-day medication. For different SGLT-2is and GLP-1 RAs, T2DM patients were willing to pay the highest of $1518 per month for oral, once-a-day GLP-1 RAs and the lowest of $1124 per month injectable, twice-a-day GLP-1 RAs. The result suggested that T2DM patients valued all SGLT-2is and GLP-1 RAs higher than their current wholesale acquisition costs (WAC). The findings of this thesis could be used to inform clinicians about what attributes are important when selecting appropriate second-line medications for T2DM patients.