OBJECTIVES: Use of Electronic Health Record (EHR) data for research is promising, but often difficult to analyze at scale. Challenges include harmonizing data into a common data model and mapping clinical concepts to standard ontologies across multiple healthcare organizations; ensuring demographic representation; maintaining recency sufficient to study rapidly emerging public health trends or evolving clinical practices; data quality; and cohort identification. Truveta confronts these challenges through a growing EHR data collective. Here we provide descriptive statistics on Truveta data and the 2023 prevalence for common conditions compared to published sources to validate the Truveta dataset.

METHODS: Using Truveta’s patient population, we examined demographic trends, clinical history, income and payer type to understand dataset representativeness. Prevalence was evaluated by dividing the number of patients with condition by number of patients with ≥1 encounter in 2023.

RESULTS: As of January 2024, Truveta data comprises 110+ million patients, with 54.2% female, 52.2% white followed by 11.3% African American, and 60.7% Non-Hispanic. Most are 18–49 years old (35.1%) and 14.4% are minors (<18 years old). 26.9% patients have >5 years of clinical history. 36.6% of patients carry commercial insurance. 35.9% and 52.4% have individual and family income between $40K and $100K, respectively. Prevalence of common conditions among the patient population compares to CDC or other published resources including type 2 diabetes (Truveta=11.8%, Others=10%-11.6%), asthma (Truveta=6.7%, Others=7%-8%), COPD (Truveta=3.8%, Others=3.3%-6.5%), and heart failure (Truveta=3.7%, Others=1.9%-2.6%).

CONCLUSIONS: The Truveta dataset contains diverse racial, ethnic, age, and social drivers of health patient information. Observed disease prevalence is comparable to government-reported and peer-reviewed resources validating the representativeness of the data in United States.