Physicians engage more with EHR prompts when feedback, incentives added

Physicians are more responsive to EHR quality improvement prompts when the reminders are offered in combination with individual feedback and financial incentives, according to a study published in the October issue of the American Journal of Managed Care.

Lead author Stephen D. Persell, MD, MPH, of the Feinberg School of Medicine at Northwestern University in Chicago, and colleagues attempted to measure physician responsiveness to EHR-based strategies to improve adherence to identified quality measures for patients with coronary artery disease (CAD). The researchers selected CAD, because it is a common chronic diagnosis among the outpatient population and because implementing CAD quality improvement measures was feasible in the practice settings studied.

They conducted the study at four private practices at separate locations in the Chicago area. Two family practices and two internal medicine practices participated. The practices were all part of a larger multispecialty group. The four practices used the same EHR system and all had been using it for at least five years.

The researchers designed an electronic point-of-care reminder system that flagged CAD patients as potential recipients of antiplatelet therapy and/or statin treatment, and patients with prior MI as potential recipients of beta-blockers and/or angiotensin-converting enzyme (ACE) inhibitors or angiotensin receptor blockers (ARBs). They considered a physician to have performed according to these measures if the patient was receiving the indicated drug or the physician documented an exception (a medical contraindication or patient refusal) in the EHR.

Beginning in July 2008, when a participating physician had a patient encounter with an apparently eligible patient who did not have an indicated medication on his or her medication list, the physician received an alert, in the form of a yellow tab on the side of his or her screen. These electronic alerts included standardized methods of either prescribing the medication or documenting an exception.

In September 2009, the researchers began sending physicians monthly feedback in reports that assessed their performance on each on the four measures, and included lists of patients who were apparently appropriate candidates for one of the four measures, but who were not receiving the medication and had no exception recorded.

In October and November 2009, the medical group announced that the four quality measures would be included in quality metrics upon which a small portion (1.5 percent) of total compensation would be based.

The authors found that the electronic reminder system alone had little impact on the physicians' behavior. There was no statistically significant change in the mean percentage of encounters in which the physicians either prescribed the indicated medication or entered the reason for the exception after the electronic reminder was introduced in September 2008. 

Physician performance improved measurably after the introduction of individual feedback and financial incentives, largely attributable to an increase in physician documentation of exceptions. The only category in which medication prescribing increased was antiplatelet therapy for CAD: at baseline, 78.9 percent of patients deemed eligible for antiplatelet therapy were receiving it, and at the end of the study that number had increased to 90.4 percent. The prescription rate for statins and beta-blockers showed no statistically significant differences and the prescription rate for ACE/ARB decreased.

However, documentation of exceptions increased significantly in all categories. Peer review found that 87.3 percent of the exceptions were medically appropriate, 11.4 percent had an inappropriate reason for the exception or no reason listed, and 1.2 percent were "of uncertain appropriateness."

“This observed change suggests that linking CDS [clinical decision support] tools to local accountability systems like incentives or providing feedback on performance can lead to greater physician engagement with EHR quality improvement tools than would occur with CDS alone,” they wrote.

Because financial incentives were introduced near the time of the individual feedback, the researchers could not determine the relative impact of these two individual factors on physician behavior. They suggested further studies looking at this particular question.

"This distinction is important since the long-term sustainability of financial incentives requires that the organization commit some amount of financial resources to prioritizing quality over other behaviors such as clinical volume," they wrote.

The researchers noted other limitations of the study. They did not review the charts of patients who had apparent quality deficits (not receiving indicated medication and no exception recorded); it was not a controlled trial and other factors may have influenced the physicians' behavior; and the results may not be generally applicable to other settings or provider groups.