ORF10-07 - Breaking Health Care Disparities in Access to Advanced Diabetes Technologies in Children with Type 1 Diabetes (T1D)

Data on automated insulin delivery (AID) technology use support their benefit in children. However, there are striking socio-economic status (SES) and racial/ethnic disparities in outcomes, with worse metabolic control and lower use in those of lower SES and racial/ethnic minorities. Reasons are multifactorial, including more limited access to care, insurance challenges, and providers’ biases in prescribing these.

Aims: To investigate in children with T1D of lower SES and suboptimal diabetes control, if when consistently offered AID closed-loop technology without fixed pre-conditions, and families better assisted with insurance approval:
can disparities decrease by improving diabetes control including time-in-range (TIR) sensor glucose (70-180mg/dl) (primary) and HbA1c;
can clinical benefit be sustained for 6 months in a real-life clinic setting.
We aimed to recruit 55 children with T1D, ages 6 to < 18yrs, with HbA1c≥8.0% and of lower SES (income < 2x federal poverty level for household size), targeting enrollment of ethnic/racial minorities. Experienced African American (AA) and Spanish-speaking coordinators assisted in recruitment. Participants were on multiple daily injections and chose from among FDA-approved AID pumps; devices were prescribed clinically with assistance provided via a clinic insurance navigator. Families were instructed on use per clinical routine. Diabetes history, pump and CGM downloads, and HbA1c were obtained. Primary outcome (%TIR) was assessed at 3-months and a 6-month visit assessed persistence of observed trends. Data were analyzed using a repeated measures linear regression model.

Results: 48 of 56 children recruited initiated closed-loop, 27% Non-Hispanic Whites, 27% Hispanic/Latino, 33% AA, 13% other; mean (SD) age 12.9 (3.3) years (range 6.2-17.6), 48%M. All met SES criteria, 90% had Medicaid. Pumps used were: 25% Tandem Control IQ, 38% Omnipod, 35% Bionic Pancreas. Compared to baseline, at 3-months TIR improved from 30 (12)% to 48 (13)% [p < 0.001], nighttime/daytime changes were comparable. HbA1c decreased from 9.8 (1.6)% to 8.4 (1.4)% [p < 0.001] with no significant hypoglycemia. There were no differences among racial/ethnic groups. Those with worse TIR and HbA1c at baseline had the greatest improvement at 3-months. There were 11 participants who had a ketoacidosis event within 12-months prior to baseline, 6 at 3-months, only 1 SAE was device-related. At 6-months 38 participants showed persistent benefit on TIR and HbA1c. Perceived obstacles for pump initiation were insurance delays, maintaining diabetes supplies, and technology difficulties.

Summary: Use of AID technology in children with poor metabolic control and lower SES in a real-life clinic setting can significantly improve TIR and HbA1c, with a strong safety profile; results were sustained at 6-months. Children with worse control appear to show greater benefit.