Prevalence of Gestational Diabetes and Preexisting Type 2 Diabetes in a 10-Year Multistate Hospital Discharge Cohort

Abstract

Background: Gestational diabetes (GDM) and preexisting type 2 diabetes (T2DM) in pregnancy are important public health challenges. Our study updates previous literature, using hospital discharge data to define the trends in GDM and preexisting T2DM in multiple states over 10 years. Methods: Utilizing a validated ICD-9 code algorithm, we defined a retrospective cohort of deliveries from the Healthcare Cost and Utilization Project (HCUP) State Inpatient Databases (SID) for FL, NY, and NE from 2005-2014. The HCUP SID contains hospital discharge data for all hospitalizations within a year. Prevalence of GDM and preexisting T2DM were calculated by state and year. Associations with race and socioeconomic status were explored. Results: Of 4,690,551 births in the cohort, 270,623 were complicated by GDM (5.8%) and 26,261 (0.6%) were complicated by preexisting T2DM. The prevalence of GDM increased from 4.8% in 20to 6.7% in 2014. The prevalence of preexisting T2DM increased from 0.5% in 20to 0.7% in 2014. Native Americans had the highest rates of both preexisting T2DM (0.8-1.7%) and GDM (8.0-15.1%). Preexisting T2DM was next most common among African Americans (0.7% in 20to 1.1% in 2014), and GDM was next most common among Asian/Pacific Islanders (8.1% in 20to 13.1% in 2014). Rates of preexisting T2DM were higher in the lowest income compared to the highest income zip codes (0.8 vs. 0.5% FL, 0.6 vs. 0.3% NY, 0.7 vs. 0.3% NE), with rates declining across income quartiles. Women missing income quartile data had higher rates of preexisting T2DM (0.8-0.9%). GDM did not show a similar trend with socioeconomic status. Conclusions: The prevalence of GDM and preexisting T2DM increased between 20and 2014 with rates varying by race. Lower socioeconomic status was associated with higher rates of preexisting T2DM. We will use multivariable analysis to explore associations between GDM and preexisting T2DM and demographic, clinical, and healthcare system access variables. Disclosure C.J. Herrick: Stock/Shareholder; Spouse/Partner; Cardinal Health. M. Keller: None. M.A. Olsen: Research Support; Self; Sanofi Pasteur. Consultant; Self; Pfizer Inc.. Research Support; Self; Pfizer Inc..