Study: Electronic health records can improve care, outcomes in patients with type 2 diabetes

Use of electronic health records shows promise for improving care and outcomes in patients with type 2 diabetes, but still has considerable room for improvement, according to a new study in the journal Health Services Research.

With the U.S Department of Health and Human Services currently offering incentives for medical practices to move their records to electronic systems—and Medicare payment penalties to those who don't convert by 2015—doctors and hospitals are increasingly incorporating these tools into their practices. However, the notion that electronic health records can improve care measurably has been under significant debate, said study co-leader Jeph Herrin, Ph.D., of Yale University.

"What's unknown is whether moving to an electronic health records system would actually improve health outcomes," he said. "So far, different studies have shown different results."

Seeking more evidence, Herrin, co-author David Ballard, M.D., Ph.D., and their colleagues took advantage of two concurrent initiatives at the HealthTexas Provider Network, a northern Texas primary care medical group subsidiary of Baylor Health Care System. The first is a long-term effort made by the group to collect health data on their diabetes patients. The second is the roll out, started in 2006, of an electronic health record system to the Network's nearly 100 clinics.

Herrin and his colleagues followed 14,051 diabetes patients from 34 group clinics over a four-year period. They looked at optimal diabetes care as a whole as well as individually examining each of five measures of diabetes management: HbA1c (a measurement that indicates average blood sugar control over time), LDL cholesterol, diastolic blood pressure, systolic blood pressure, aspirin prescription and non-smoking status.

They report that over the rollout period, the 29 practices that had the electronic health record system saw 6,376 diabetic patients. Comparing those exposed to the system to those that were not, the researchers found a significant improvement in all of the variables except for HbA1c and LDL cholesterol in the electronic health records group.

Herrin explained that certain features of the electronic health record system might be responsible for the better health measures among patients whose doctors used it. For example, even if patients visited their physician for something unrelated to diabetes—such as a flu shot—the system reminded their doctor to ask questions related to diabetes care. The researchers also noted that part of the increase might be due to better documentation in electronic health records than in paper records.

"This study offers evidence and reassurance that switching to electronic medical records isn't going to hurt patients and may in fact help them," Herrin said.

However, despite the significant increase in the number of patients who met some optimal care and outcome standards, Spyros Mezitis, M.D., an endocrinologist at Lenox Hill Hospital in New York City, noted that doctors and electronic health records still have a long way to go.

"This study showed mixed results," he said. "There's much room for improvement here."

Mezitis points out measures for HbA1c and LDL cholesterol, two very important elements in diabetes care, did not improve in this study in the electronic health records group. With the extra initial expense, time and effort it takes to set up electronic health records, he added, more research is needed to develop systems that improve all measures of diabetes care.