DHIN’s HCCD Helps the CDC
DHIN offers many services that are useful in improving Delawareans’ health. Recently, the Centers for Disease Control (CDC) used DHIN’s Health Care Claims Database (HCCD) to conduct a study that showed the viability of claims data as a tool for public health.
The study, published in early March 2025, demonstrated the effectiveness of healthcare claims as a source of timely data on health outcomes at geographically refined levels and on health outcomes not available through existing public health surveillance systems. The study also demonstrated the availability of demographic information on all healthcare plan members from all-payer claims databases that can be used as denominator data for public health studies.
In 2014, per- and polyfluoroalkyl substances (PFAS) were detected in five drinking water wells in New Castle City, New Castle County, Delaware. This study used healthcare claims data to show that healthcare plan members living in the ZIP code served by the five wells were at higher risk for type 2 diabetes, hypertension, hypertensive diseases, coronary artery disease, and hyperthyroidism based on new claims compared with healthcare plan members living elsewhere in the county. Healthcare claims data provided timely information on health outcomes not captured by traditional public health surveillance systems and at finer geographic levels.
DHIN provided the Centers for Disease Control and Prevention’s (CDC) National Center for Environmental Health/ATSDR with a de-identified list of healthcare plan members from the HCCD. It listed the healthcare plan members in New Castle County who had a claim consistent with a diagnosis of a health outcome of interest for this study during the years 2013–2019. All data provided was de-identified; a universal unique code across the years was used to represent each member to protect personally identifying information. This file included only information about the first instance of a claim with a principal diagnosis or first secondary diagnosis of the health outcome, including month of diagnosis, year of diagnosis, member’s age at diagnosis, member’s sex, and member’s unique code for this study. DHIN also provided yearly de-identified lists of all healthcare plan members living in New Castle County in the years 2013–2019, regardless of having a health outcome of interest. The information provided included a member’s unique code for this study, sex, age, ZIP code of address, and number of months lived in that ZIP code each year.
The DE HCCD provided information on health outcomes not typically captured in public health surveillance data (e.g., type 2 diabetes and thyroid diseases) and at a fine geographic level (i.e., ZIP code). Healthcare claims data for this study were also timelier than most public health surveillance data, with annual datasets typically available 3 months after the end of the year.
The study demonstrated healthcare claims are a useful source of timely data on health outcomes at geographically refined levels and on health outcomes not available through existing public health surveillance systems. The study also demonstrated the availability of demographic information on all healthcare plan members from all-payer claims databases that can be used as denominator data for public health studies.
Read the full CDC Article: https://pmc.ncbi.nlm.nih.gov/articles/PMC11878993/
