IDF21-0101 Blood Glucose Control using a Mobile Health Application in Asia – A Retrospective Real-World Data Analysis

Abstract

Background: Mobile health (mHealth) solutions have the potential to reduce the clinical and psychological burden of diabetes and enhance patient self-management. Previous real world data analyses showed significant improvement in blood glucose (BG) control with the use of a mHealth application for people with diabetes after 6 months. However, there is under-representation of Asian data while user behaviours and outcomes remain little known. Aim: We aimed to investigate the potential impact of mySugr Logbook app usage on parameters of BG control in Asia – Singapore, Philippines and Hong Kong. Method: We analysed changes in BG control [mean BG, tests in range (TeIR), estimated HbA1c (eHbA1c)] in a sample of highly engaged users, defined as ≥2 logs on at least 14 out of 30 days, over a 6-month period. We also analysed measures of glycaemic variability [high blood glucose index (HBGI) and Low blood glucose index (LBGI)]. Data from the first month (t0) of use was compared to month 6 (t1) using paired t-tests in individuals who logged continuously for 6 months. Subgroup analysis was performed on those with suboptimal glycaemic control (eHbA1c >7.0%). Results: A total of 1761 users satisfying criteria above have been using mySugr app over a 6-month period. Of this, 63.7% had Type 2 diabetes (T2D), 14.4% had Type 1 diabetes (T1D), 21.9% was others/unknown. Baseline eHBA1c and mean BG were 6.6% and 142.6 ± 36.4 mg/dl respectively. In 357 individuals who logged continuously, t0 eHbA1c 6.7±0.9% fell 0.2% (p<0.01) at t1. In those with eHbA1c>7% (n=107) at baseline, mean t0 eHbA1c 8.1 ± 1.4%, reduced by 0.5% (p<0.01) with TeIR improving by 10.9% (p<0.01) at t1. Likewise, HBGI decreased from 11 (high risk) to 9 (moderate risk) (p<0.01) whilst the LBGI did not significantly change and remained at low risk level (<2.5) (p=0.300). Discussion: This real-world data shows improvements in BG control in those with higher eHbA1c at baseline over a 6-month period in highly engaged users. A clinically relevant decrease in eHbA1c by 0.5% for those with baseline eHbA1c>7% at 6 months of mHealth application usage was shown to be statistically significant. Although mySugr app supports manual logging of BG data, it was previously observed that users who manually logged data had lower retention rates compared to those who used connected BG meters that enable automatic transfer of data. This preliminary work suggests a positive impact of connected BG meters and mobile health applications in supporting diabetes self-management.