Abstract
BACKGROUND: To date, there is a lack of information on the use of flash glucose monitoring system (fCGM) in low-middle income countries, such as Brazil, as well as on digital health platforms most used to calculate the bolus insulin dose. In this study, we aimed to describe the population characteristics, prescription patterns and glycemic control of fCGM users compared to blood glucose monitoring (BGM) system in those who use Glic™, a digital health platform in Brazil, and to assess factors associated with better glycemic control in this population. METHODS: This study is a cross-sectional retrospective study using anonymized aggregated data manually inputted by Glic™ users who self-reported a diagnosis of type 1 diabetes (T1DM), type 2 diabetes (T2DM), gestational diabetes (GDM) and latent autoimmune diabetes in adults (LADA). RESULTS: Of the 12,727 individuals included in this study, 11,007 (86.5%) reported their glucose monitoring method to be BGM, while 1720 (13.5%) reported using fCGM. Most individuals (70.5%) had T1DM. Compared to BGM, fCGM users were significantly younger, had a higher proportion of males, resided more frequently in the Southeast region of Brazil, had a lower BMI, a longer time since diagnosis, and used Glic™ platform more frequently. fCGM users were prescribed significantly more ultra-long and ultra-rapid acting insulins as their basal and bolus insulin, respectively, and less oral anti-diabetics drugs compared to BGM users. Considering only the T1DM and LADA individuals and their manual glucose inputs, fCGM users had non-significant lower glucose levels than BGM. Use of Glic™ platform and a higher percentage of basal insulin dose were associated with a better glycemic control. CONCLUSION: This is the first and largest real-world evidence study that describe and compare fCGM and BGM in users of a digital health patient support platform in Brazil. fCGM users were significantly different from those who perform BGM, in terms of population characteristics and treatment patterns. Glycemic control was better in fCGM users, although not statistically significant due to a restricted sample size. Importantly, a higher frequency of Glic™ use was associated with a higher glucose time in range.