A quantitative assessment of patient barriers to insulin

Abstract

To assess diabetes treatment preferences with a focus on patient barriers to insulin treatment. A questionnaire using indirect and direct methods was administered as part of the International Diabetes Management Practices Study (IDMPS). Discrete choice modelling was used to assess how product attributes influence patients' preferences for diabetes treatment. A multinomial logit model was used to find the odds ratio for each parameter, representing the probability of selecting a chosen alternative given a choice set. This allowed for the derivation of relative attribute importance, an indication of how influential product attributes are in the respondents' choices. The IDMPS questionnaire was administered to 14,033 individuals with diabetes in 18 countries. The majority of respondents were women (53%) and had Type 2 diabetes mellitus (T2DM; 85%). Across subgroups, administration (i.e. oral vs. injection) was a driver of preference. Patient preferences varied according to diabetes type; individuals with T2DM assigned much higher relative importance to administration than those with Type 1 diabetes mellitus (T1DM; 30.86% vs. 4.99%; p<0.0001). Individuals with T2DM treated with insulin placed less importance on administration than insulin-naïve T2DM patients (3.09% vs. 47.48%; p<0.0001). Diabetes education also had a significant effect on the priority given to administration between T2DM patients who received diabetes training and those who did not (28.21% vs. 33.68%, respectively; p<0.0001). The insulin barriers perceived by patients with diabetes evolved with their disease experience. While administration was the primary preference driver for insulin-naïve patients, patients were increasingly concerned with more clinically relevant barriers as they gained experience with insulin. This finding suggests that patients using insulin understand the importance of achieving an optimal balance between safety and efficacy.