Electronic Health Record Alerts Fail To Improve Acute Kidney Injury Outcomes: Study

While there has been an increased push in recent years to increase the use of electronic health records among medical providers, the findings of a new study suggest that digital alerts generated from electronic health records do not appear to reduce the risk of death among patients with acute kidney injury.

In a report published this week in the medical journal The BMJ, Yale researchers conducted a review to evaluate whether impact electronic medical record alerts here helpful in providing an early warning of acute kidney injury among hospitalized patients.

Researchers conducted a double blind randomized controlled trial among six hospitals, including four teaching and two non-teaching facilities, in the Yale New Haven Health System in Connecticut and Rhode Island. The study included 6,000 adult inpatients with acute kidney injury randomized over 22 months, and found that the alerts were not only unhelpful, but may increase the risk of certain side effects.

Acute kidney injury can lead to an abrupt decline of the kidney’s filtration function. Roughly 15% of hospitalized patients have this issue and it increases the patient’s risk of death significantly. However, it is not well tracked or treated.

The electronic health record alerts for acute kidney injury were meant to improve patient outcomes, reducing the risk of death, dialysis or injury progression.

For the study, an alert for acute kidney injury with an associated acute kidney injury order was set to pop-up when the doctor opened the patient’s medical record and laboratory results indicated kidney injury. All patients in the study had acute kidney injury, but only half of the patients’ doctors received pop-up alerts. The function was switched off for the other half to determine if the pop-up alerts helped to inform treatment.

The study indicated the alert led to only small changes in doctor behavior. Only some of the time the doctor ordered additional urinalysis tests, creatinine tests, or documented the issues more frequently.

Overall, the alerts and lack of change in doctor treatment behavior resulted in few patients having improved outcomes. Patients who had the pop-up alerts were still more likely to die within 14 days, need dialysis or have their symptoms worsen. 

“Alerts did not reduce the risk of our primary outcome among patients in hospital with acute kidney injury,” the researchers wrote. “The heterogeneity of effect across clinical centers should lead to a re-evaluation of existing alerting systems for acute kidney injury.”

Among patients in the two non-teaching hospitals, outcomes for patients who had pop-up alerts were actually worse than for patients who did not get the pop-up alerts on their records. Patients with medical record pop-up alerts had higher death rates at non-teaching hospitals than those who did not get the alerts.

While researchers had hoped pop-up alerts would lead to improved outcomes for kidney injury patients, the findings indicate other factors need to be explored when determining how to best treat these patients to prevent negative outcomes.