PROJECT SUMMARY/ABSTRACT
In the United States, the prevalence of youth-onset type 2 diabetes (T2D) diagnosed in adolescents has nearly
doubled in one decade and is projected to increase four-fold by 2050. Young adults less than 40 years old now
represent 15-20% of adults with T2D. Early-onset T2D in adolescents and young adults (AYAs) is associated
with earlier and higher rates of failed glycemic control, cardiovascular disease, microvascular complications, and
mortality compared to T2D that develops after age 40 years old. Accelerated decline in b-cell function is thought
to underlie the aggressive T2D phenotype in AYAs. The T2D glucagon-like peptide 1-receptor agonist therapies
(GLP-1 RAs) achieve glycemic control by stimulating glucose-dependent insulin secretion. However, the real-
world effectiveness and utilization of GLP-1 RAs in young adults with T2D compared to other second-line non-
insulin diabetes therapies are unknown. Furthermore, recent GLP-1 RA shortages, known racial/ethnic
disparities in access to novel diabetes medications, and the disproportionally increased risk of T2D among
minority and socioeconomically disadvantaged AYAs, all raise additional concerns regarding AYA access to
GLP-1 RAs. The overarching goal of this training program is to evaluate the real-world effectiveness and
utilization of GLP-1 RAs in AYAs with T2D using the TriNetX network database, a large-scale electronic health
record database. In Aim 1, a retrospective incident user cohort design will be used to emulate a target trial to
compare the real-world effectiveness of GLP-1 RAs, sulfonylureas, dipeptidyl peptidase-4 inhibitors, and sodium-
glucose cotransporter-2 inhibitors in achieving and maintaining glycemic control in young adults aged 18-30
years. In Aim 2, an interrupted time series will be used to examine changes in utilization of GLP-1 RAs in AYAs
10 to 30 years old since FDA approval of the GLP-1 RA, liraglutide, for adolescents in June 2019 and use
multivariable logistic regression to identify patient characteristics associated with lower utilization of GLP-1 RAs.
The proposed study will provide insights in how to best optimize treatment of T2D in AYAs and lead to future
studies evaluating interventions to increase utilization of these treatments. This grant will provide Dr. Chu with
valuable training in epidemiologic research methods. She will obtain intensive training in complex observational
study design, biostatistics, and database management by accomplishing the above aims and through formal
coursework as part of her Master of Science in Clinical Epidemiology degree program. Her research training will
also involve mentors with complementary expertise in pharmacoepidemiology and patient-oriented research in
diabetes who are focused on her career development into a physician-scientist. The data obtained from this
proposal will inform the design of future studies in optimizing treatment and access of medications in AYAs with
T2D to prevent long-term diabetes-related complications, which will be the focus of her K23 application.
