View article

Methods: An online discrete choice experiment survey was administered to adults with T2D in the United States. Each respondent completed 8 questions offering a choice between experimentally designed pairs of hypothetical, long-acting insulins that varied by 6 attributes: reduction in A1c level after 6 months, daily time in range, number of serious low blood sugar events, number of nighttime low blood sugar events, weight change over 6 months, and administration frequency. A fixed choice question directly elicited preferences for flexible weekly over daily insulin, holding all other attributes equal. Data were analyzed with random-parameter logits.

Results: A total of 466 people (mean: age= 57; A1c= 7.5%; 59% female) completed the survey. They were on a basal/bolus (33.3%), basal-only (34.3%), or insulin-naïve (32.4%) regimen. Relative to other attributes, people placed most importance on avoiding a 10-pound …