Patient preferences for the treatment of type 2 diabetes in Australia: a discrete choice experiment

Background Australia has a high proportion of migrants, with an increasing migration rate from India. Type 2 diabetes is a chronic condition common amongst the Indian population. The decision to initiate and continue medication therapy (conventional or ayurvedic medicine) is complex and is influenced by a wide range of factors. Objective To determine preferences for conventional vs. ayurvedic medicines in Indian migrants with diabetes, and to identify the factors that may influence their preferences. Methods A discrete choice experiment was conducted with participants in Australia who were migrants from India with type 2 diabetes (n=141). Each respondent evaluated eight choice tasks consisting of eight attributes describing medicines and outcomes of medication taking; and were asked to choose 'conventional', or 'ayurvedic' medicine. A mixed multinomial logit model was used to estimate preferences. Results Overall, respondents' preference to initiate a medicine was negative for both conventional (beta=-2.33164, p<0.001) and ayurvedic medicines (beta=-3.12181, p<0.001); however, significant heterogenicity was noted in participants' preferences (SD: 2.33122, p<0.001). Six significant attributes were identified to influence preferences. In decreasing rank order: occurrence of hypoglycaemic events (relative importance, RI=24.33%), weight change (RI=20.00%), effectiveness of medicine (RI=17.91%), instructions to take with food (RI=17.05%), side effects (RI=13.20%) and formulation (RI=7.49%). Respondents preferred to initiate a medicine despite potential side effects. Conclusions There was a greater preference for conventional medicine, though neither were preferred. Medicine attributes and medication-taking outcomes influenced people's preferences for an antidiabetic medicine. It is important to identify individual preferences during healthcare consultations to ensure optimal medication-taking.