Study: EHR Phenotype Data Can Improve Detection of Type 2 Diabetes | Healthcare Informatics Magazine | Health IT | Information Technology Skip to content Skip to navigation

Study: EHR Phenotype Data Can Improve Detection of Type 2 Diabetes

February 17, 2016
by Heather Landi
| Reprints
Click To View Gallery

Researchers from the University of California Los Angeles’ Department of Psychiatry and Biobehavioral Sciences found that the use of electronic health record (EHR) phenotyping to screen for Type 2 diabetes out-performed conventional screening methods, according to a new study published in the Journal of Biomedical Informatics.

An estimated 25 percent of type two diabetes mellitus (DM2) patients in the United States are undiagnosed due to inadequate screening, because it is prohibitive to administer laboratory tests to everyone, the researchers wrote in the study report.

The research team examined the use of EHR phenotyping to improve DM2 screening compared to conventional models. EHRs contain many phenotypes, such as specific traits or the presence of disease.

In the cross-sectional, retrospective study, EHR data from 9,948 U.S. patients were used to develop a pre-screening tool to predict current DM2. The researchers compared a full EHR model containing commonly prescribed medications, diagnoses (as ICD9 categories), and conventional predictors, and also compared a restricted EHR DX model which excluded medications, as well as a conventional model containing basic predictors and their interactions (body mass index, age, sex, smoking status, hypertension).

The research team found that a diagnosis of “sexual and gender identity disorders” and “viral infections” increased the risk of the disease.

According to the study report, researchers found that EHR phenotyping “resulted in markedly superior detection of DM2, even in the face of missing and unsystematically recorded data, based on the receiver operating characteristic (ROC) curves."

“EHR phenotypes could more efficiently identify which patients do require, and don’t require, further laboratory screening. When applied to the current number of undiagnosed individuals in the United States, we predict that incorporating EHR phenotype screening would identify an additional 400,000 patients with active, untreated diabetes compared to the conventional pre-screening models,” the researchers wrote.


Get the latest information on Health IT and attend other valuable sessions at this two-day Summit providing healthcare leaders with educational content, insightful debate and dialogue on the future of healthcare and technology.

Learn More



Saving 400,000 patients who don't even realize they have diabetes is great and I believe many find this study invaluable but researchers believe 400,000 is unrealistic number unless you can obtain medical records of all citizens.


ONC Roundup: Senior Leadership Changes Spark Questions

The Office of the National Coordinator for Health IT (ONC) has continued to experience changes within its upper leadership, leading some folks to again ponder what the health IT agency’s role will be moving forward.

Media Report: Walmart Hires Former Humana Executive to Run Health Unit

Reigniting speculation that Walmart and insurer Humana are exploring ways to forge a closer partnership, Walmart Inc. has hired a Humana veteran to run its health care business, according to a report from Bloomberg.

Value-Based Care Shift Has Halted, Study Finds

A new study of 451 physicians and health plan executives suggests that progress toward value-based care has stalled. In fact, it may have even taken a step backward over the past year, the research revealed.

Study: EHRs Tied with Lower Hospital Mortality, But Only After Systems Have Matured

Over the past decade, there has been significant national investment in electronic health record (EHR) systems at U.S. hospitals, which was expected to result in improved quality and efficiency of care. However, evidence linking EHR adoption to better care is mixed, according to medical researchers.

Nursing Notes Can Help Predict ICU Survival, Study Finds

Researchers at the University of Waterloo in Ontario have found that sentiments in healthcare providers’ nursing notes can be good indicators of whether intensive care unit (ICU) patients will survive.

Health Catalyst Completes Acquisition of HIE Technology Company Medicity

Salt Lake City-based Health Catalyst, a data analytics company, has completed its acquisition of Medicity, a developer of health information exchange (HIE) technology, and the deal adds data exchange capabilities to Health Catalyst’s data, analytics and decision support solutions.