156 - Disparate Outcomes in Children with Type 1 Diabetes Assigned to Practices Based on Insurance Type: Equality Does Not Lead to Equity

Poster Number: 156
Publication Number: 156.411

Zenna Solomon, NewYork-Presbyterian Komansky Children’s Hospital, Bronx, NY, United States; Zoltan Antal, Weill Cornell Medicine, New York, NY, United States; Charlene Thomas, Weill Cornell Medicine, New York, NY, United States; Linda M. Gerber, Weill Cornell Medicine, NY, NY, United States; Joy Howell, Weill Cornell Medicine, New York, NY, United States

Background: Several studies have documented differences in outcomes based on race, ethnicity, and insurance type in children with Type 1 Diabetes (T1DM). At our center, children with T1DM are treated at different outpatient practices based on insurance (public vs. private). A previous study found lower hemoglobin A1c (HbA1c) values, greater use of insulin pumps (IP) and continuous glucose monitors (CGM), and increased utilization of the online patient portal (OPP) in our private insurance practice (PRV) compared to our public insurance practice (PUB). In 2018, a pediatric endocrine nurse practitioner (NP) was introduced to the PUB, thus matching the panel of providers for the PRV, with the expectation that diabetes outcomes would improve.

Objective: Investigate whether the introduction a NP to the PUB mitigated the disparate outcomes observed between the PRV and PUB.

Design/Methods: Retrospective chart review of all patients with T1DM, birth to 21 years of age, at least 1 year removed from initial diagnosis, with at least one practice visit in 2017 or 2019. We compared the difference between the two practices over time by using descriptive statistics and generalized estimating equations.

Results: There were 119 patients in the 2017 cohort (37 PUB and 82 PRV), and 120 patients in the 2019 cohort (37 PUB and 83 PRV). In both cohorts there was no statistically significant difference between practices with respect to median patient age or sex. In 2017 a greater percentage of patients in the PRV identified as White [50% vs 2.7%, p < 0.001] and a greater a percentage in the PUB identified as Hispanic [37.8% vs 8.5%, p= 0.032]. This distribution persisted in 2019. In both cohorts, PRV patients had lower mean HbA1C [2017: 8.8% vs 10.1%, (p= 0.00033); 2019: 8.5% vs 9.7%, (p= 0.002)]. In 2017 PRV patients were more likely to have used an IP (72.0% vs 45.8%, p= 0.011) or a CGM (52.4% vs 8.1%, p < 0.001). Technology use increased in the PUB in 2019 (58% IP and 42% CGM), however the practices were no less disparate over time. In both cohorts, PRV patients were more likely to have an active OPP [2017: 52.4% vs 5.41%, (p < 0.001); 2019: 65% vs 25%, (p < 0.001)]. However, analysis of access to the OPP during the Covid-19 pandemic (2020) showed no difference between the two practices (PRV 87.9% active vs PUB 83.7% active, p= 0.8).Conclusion(s): The addition of NPs to the PUB did not mitigate disparate outcomes between the two study populations. Changes in OPP access in 2020 demonstrates that patient, provider, and systems level barriers can be surmounted with sufficient motivation.
Average HbA1cAverage hemoglobin A1c for the PRV and PUB in 2017 and 2019
Continuous Glucose Monitor UseContinuous glucose monitor (CGM) use in 2017 and 2019 for the PUB and PRV. Generalized estimating equations showed that the two clinics did not become less disparate over time