OBJECTIVES: To determine the important attributes of continuous glucose monitoring (CGM) devices for patients with type 2 diabetes (T2D).

METHODS: Seven CGM-specific attributes were identified through a literature review and consultation with five clinical experts. Preliminary data were collected from 30 adult Americans living with T2D who were proficient in English via a cross-sectional, web-based, discrete choice experiment (DCE) survey. A D-efficient design was used to generate DCE choice sets. A multinomial logit model was used to determine the preference weights of CGM attributes. Conditional relative importance was calculated by differentiating the highest and lowest level of preference weights of each attribute.

RESULTS: Twenty-six participants passed the validity tests. The preference weights of all attributes, except measurement error, were in the expected direction. The preference weights of five attributes, including alarm function, information provided, calibration frequency, sensor lifespan, and out-of-pocket cost, were statistically significant. The conditional relative importance of out-of-pocket cost was the highest (1.12), followed by sensor lifespan (0.72), calibration frequency required (0.71), information provided (0.65), alarm function (0.56), methods to access information (0.24), and measurement error (0.11).

CONCLUSIONS: The majority of selected CGM attributes in this preliminary study are potentially important to patients living with T2D. Specifically, the patients identified the alarm function, information provided, calibration frequency, sensor lifespan, and out-of-pocket cost as significantly important attributes when selecting preferred CGM devices. However, other attributes, especially measurement errors, still require attention before conducting the main survey.