Study: Computerized clinical decision support system helps reduce systolic BP

Key takeaways:

• The intervention reduced patients’ mean systolic BP by 14.6 mm HG.
• However, there was no difference between the intervention and usual care groups in the percentage of patients who achieved BP control.

A computerized clinical decision support system intervention reduced patients’ mean systolic BP and increased prescriptions of multiple medications, a study in JAMA Internal Medicine showed.

“It was a small change, but many other studies that have been done in the same population have not been able to show a difference, so we were glad to see that,” Lipika Samal, MD, MPH, an associate professor at Harvard Medical School, told Healio.

According to Samal and colleagues, research has shown that primary care providers “are not always aware of [chronic kidney disease (CKD)] management guidelines and face barriers to implementing them.”

The researchers developed a CDS intervention based in the electronic health record that included a set of five Best Practice Advisories (BPA) embedded into the Epic Systems software, and it also featured several external and internal behavioral economic elements.

The internal elements included preselected orders, like medication and basic metabolic panel orders, “that nudged PCPs toward recommended actions and a required accountable justification if the PCP did not place the orders that were recommended,” they explained.

“We were really trying to build up this literature in nudges that have been used for other types of clinical domains,” Samal said. “It’s kind of a new clinical area.”

The external elements included an email asking PCPs to pledge to follow recommendations for BP management and requiring an accountable justification if they did not.

In the randomized clinical trial, 87 PCPs and 1,029 patients were assigned to the intervention, whereas 87 PCPs and 997 patients were randomly assigned to usual care, where the CDS operated in silent mode.

The primary outcome was change in systolic BP between baseline and 180 days, while other outcomes included change in controlled BP and initiation of angiotensin-converting enzyme inhibitors (ACE), angiotensin receptor blockers (ARB) or hydrochlorothiazide.

Samal and colleagues found there was a statistically significant difference in mean systolic BP change in the intervention group (change = 14.6 mm HG; 95% CI, 13.1 to 16) vs. the usual care group (change = 11.7 mm HG; 95% CI, 10.2 to 13.1).

There was no difference in the percentage of patients who achieved BP control in the intervention group compared with those of the control group (50.4% vs. 47.1%).

However, more patients in the intervention group received an action that aligned with the CDS recommendations compared with their counterparts (49.9% vs. 34.6%).

The intervention also increased prescriptions of ACEs and ARBs, “which is an important step toward improving long-term clinical outcomes,” the researchers noted.

“The positive results of this study could be related to the behavioral economic elements of the intervention, which may be generalizable to other conditions,” they wrote. “Similar CDS tools are likely to be effective in chronic conditions, like diabetes, because most diabetes is managed by PCPs who are motivated to provide evidence-based care but often experience clinical inertia during busy, time-pressured office visits.”

For future research, Samal suggested building the same kind of algorithm into a population health report.

“You could just do the same thing on a whole population of patients, create a list of patients that would be recommended for each of the five [BPAs] that we gave, and instead of having a PCP need to deal with this right in the middle of a visit, that could go to a pharmacist,” she explained. “I think it would be even more powerful if it was used in a population health framework, it’s just something the health system would need to invest in."