Screening patients for social needs in electronic medical records yielded a 9% improvement in blood pressure control over time, according to a study published in the Annals of Family Medicine.

The findings add to mounting evidence showing the effect of these social risk factors on health outcomes and come as healthcare organizations increase their documentation of the social determinants of health.

“Social risks such as food and housing insecurity impact health, and healthcare providers are increasingly asking patients about these risks,” said Laura Gottlieb, MD, MPH, professor of family and community medicine at the University of California, San Francisco, who led the study. However, clinicians do not yet know the best ways to leverage social risk information to improve care and patient outcomes, she said.

Gottlieb and colleagues studied the effect of electronic health record (EHR) alerts that instructed clinicians to screen for social risk factors.

If patients had uncontrolled hypertension or diabetes, clinicians were prompted to add relevant Z codes on the basis of screening results. When the record showed a patient had not picked up a previous prescription, clinicians received a notification to discuss reasons for nonadherence and to discuss affordability and other potential barriers. Clinicians also received a checklist of potential topics to discuss based on the patient’s social needs. These components also popped up for patients with hypertension and diabetes who had not shown up to half of their scheduled appointments over the prior year.

Gottlieb and her colleagues reviewed data from six clinics with nearly 10,000 adults that implemented the EHR intervention over a year-long period. Over 40 clinics with more than 69,000 patients served as control clinics. The control clinics did not implement pop-up alerts for social support and used usual care protocols. The intervention group included more Hispanic patients and patients who preferred to speak another language than English (39% vs 31% and 37% vs 33%, respectively).

The primary outcome was control of blood pressure in patients with hypertension, defined as < 140 mm Hg systolic and < 90 mm Hg diastolic. Control of A1c in patients with diabetes was also measured.

Patients were more likely to achieve good control of blood pressure in the intervention clinics compared with control clinics (odds ratio, 1.09; 95% CI, 1.01-1.19; P = .04). No statistically significant differences were observed between the groups for control of blood sugar.

Clinicians in intervention clinics were also significantly more likely than clinicians in control clinics to complete social risk screenings (odds ratio, 7.3; 95% CI, 1.5-36; P = .01) and to document social risks using Z codes (odds ratio, 11.3; 95% CI, 3.1-40.7).

However, the researchers noted that uptake of the tools to document care adjustments was low overall, with 21.3% of visits at intervention clinics adding or removing a Z code to a diagnosis or problem list. In comparison, this action was taken during 8.5% of visits at control clinics, where clinicians were not prompted by the electronic record to do so.

“We were not surprised to see that clinic teams liked and responded to reminders to screen patients for social risks, but we were surprised that they rarely used the tool elements to actually document care plan adaptations, particularly since we co-designed the tools with future users who said the documentation support would be helpful,” Gottlieb said. “The documentation demands on already overburdened community health center staff might have been too much.”

Tim Joos, MD, internist and pediatrician at Neighborcare Health in Seattle, said the social risk data seemed to be easily collected and presented in a brief format.

“Common sense tells us that the more we know about the challenges our patients are facing, the better we can tailor our treatment, and I fully support this intervention,” Joos said.

However, “when making the decision to screen for these risk factors, it is always important to have a remedy or response available,” he added. “Unfortunately, with the recent short-sighted cuts to Medicaid and to federal support of community health centers, our ability to respond to social risk factors with support staff such as nursing and social workers is extremely limited.”

The study was supported by the National Institute on Minority Health and Health Disparities. The researchers and Joos disclosed having no financial conflicts.