Statewide data infrastructure supports population health management: diabetes case study

Area of Science:

• Health Informatics
• Population Health Management
• Diabetes Care

Background:

• Diabetes management is complex, with costs influenced by comorbidities.
• Effective data infrastructure is crucial for understanding healthcare expenditures.
• Previous analyses often lack comprehensive, linked clinical and claims data.

Purpose of the Study:

• To evaluate the impact of comorbidities on healthcare costs for patients with diabetes using a statewide data infrastructure.
• To identify key factors driving medical expenditures and hospital admissions in a diabetic population.
• To assess the potential for cost savings through targeted interventions.

Main Methods:

• Retrospective analysis of linked clinical and multipayer claims data for 6719 patients with diabetes (2014).
• Examined healthcare expenditures against glycated hemoglobin (A1C) control.
• Used multivariable linear and Poisson regression to identify risk factors for expenditures and hospital admissions.

Main Results:

• No direct linear relationship found between A1C levels and same-year medical expenditures.
• Renal failure, congestive heart failure, COPD, and discordant blood pressure were significant cost drivers.
• Diabetes with congestive heart failure had the highest cost per inpatient admission; high BMI (≥35) correlated with highest aggregate costs and potential savings.

Conclusions:

• Statewide data infrastructure enables identification of high-risk diabetes patients for targeted outreach and management.
• Comorbidities are more impactful on near-term expenditures than recent A1C control.
• Aggregated population data provides actionable insights for alternative payment models and collaborative care.