Understanding the Conditions When EHRs Work for Patients With Diabetes

Published on: 
Evidence-Based Diabetes Management, Patient Centered Diabetes Care 2016, Volume 22, Issue SP9

Coverage from Patient-Centered Diabetes Care, April 7-8, 2016. Presented by The American Journal of Managed Care and Joslin Diabetes Center.

Broader use of electronic health records (EHRs) is among the legacies of the Affordable Care Act (ACA). Yet studies on their clinical impact in diabetes care have produced mixed results. Ilana Graetz, PhD, an assistant professor at the University of Tennessee Health Sciences Center, is interested in why that happens—and designed a study, she said, to “understand the conditions necessary to really make the most out of electronic health records.”

Graetz presented research that first appeared in The American Journal of Managed Care1 to attendees at Patient-Centered Diabetes Care. Funded by the Agency for Healthcare Research and Quality, Graetz’s study involved a research tool to understand how office dynamics affect acceptance—and success—of EHRs in primary care, where much of diabetes treatment occurs.

“There’s been very little research that looked at how the organizational environment can change how EHRs get adopted and used, and what impact they have on outcomes,” Graetz said. Her point, if understudied, is obvious: simply plunking expensive technology into the workflow does little if users are not ready to embrace it.

Her research has both quantitative and qualitative elements. When Graetz spoke with primary care physicians (PCPs) about their EHR experiences, she heard that the sheer quantity of information can undermine its purpose. “One physician that I talked to said that ‘the quantity of unnecessary information adversely impacts the care quality because it’s overwhelming.’” Another told her, “There’s so much information and repetition in the system that it’s easy to miss the important points.”


Graetz’s evaluation of PCP team dynamics builds on earlier work, which found that just having an EHR in a practice increased the likelihood that persons with diabetes would get annual screenings for glycated hemoglobin (A1C) and low-density lipoprotein (LDL) cholesterol.2 “If they did have a lab value that was above their target, they were more likely to get the appropriate treatment intensification,” she said. “And they were more likely to have reductions in A1C and LDL values.”

Examining data from Kaiser Permanente as it implemented its Epic system from 2005 to 2010, Graetz sent surveys to primary care teams to understand what set apart the primary care practices with the best clinical results with EHR. She received 780 responses, about 50% of those she sent.

Then, Graetz assigned scores for “team cohesion” among the respondents, using an instrument developed to measure team climate at the primary care level. She then matched those up against A1C and LDL cholesterol results from 2005 to 2009—involving some 80,000 patients, with about half the readings coming before EHR implementation and half afterward. Of the patients, 85% were over age 50 and most had comorbidities.

The data analysis showed that when patients saw primary care teams with high levels of cohesion, the use of EHRs was “associated with significantly greater improvements in A1C levels (0.11 percentage point decrease in A1C) compared with patients who saw teams with low cohesion (0.08 percentage point decrease)”1 (see TABLE).

LDL cholesterol results were similar. “You can see that all patients received a significant improvement in their LDL values, but that patients that were cared for by teams with higher cohesion got a significantly greater reduction in their LDL,” she said. Her data showed high-cohesion teams were associated with a 2.15 mg/dL decrease, while low-cohesion teams saw just 1.42 mg/dL.1

“Some implication of the study is that organizational context is definitely important for understanding the impact that EHRs can have on quality outcomes,” Graetz said. “It might not be enough to just turn on the EHR system if you don’t have the organizational support and environment to support that transition.”

“In the future, research should continue to look into not just, ‘Does EHR improve outcomes?’ but ‘What are the conditions necessary to make the most of this investment?’” 1. Graetz I, Huang J, Brand R, et al. The impact of electronic health records and teamwork on diabetes care quality. Am J Manag Care. 2015;21(12):878- 884.

2. Reed M, Huang J, Graetz I, et al. Outpatient electronic health records and the clinical care and outcomes of patients with diabetes mellitus. Ann Intern Med. 2012;157(7):482-489. doi: 10.7326/0003-4819-157-7-201210020- 00004.