Acute kidney injury alert system receives $3 million from NIH for testing

A team of Yale researchers led by Yale nephrologist F. Perry Wilson, M.D., has been awarded $3 million from the National Institute of Diabetes and Digestive and Kidney Diseases at the National Institutes of Health (NIH) for a series of randomized, controlled trials of an automated, electronic alert system for acute kidney injury in inpatient settings.

Overall, about 1% of patients die during a hospitalization of any nature; however, for patients who suffer acute kidney injury (AKI) — the abrupt loss of kidney function due to various causes —  that rate rises to 10%. AKI occurs in about 10%–15% of hospitalized patients and has no symptoms, meaning it may be missed by providers, limiting their ability to take appropriate, mitigating actions (i.e., diagnostic evaluation and monitoring, medication dosage-adjustment, and avoidance of drugs that might be toxic to the kidneys).

While healthcare systems in the U.S. and U.K. have adopted electronic alerts to help identify AKI, according to Wilson’s team, no studies have evaluated AKI alerts, coupled with clinical recommendations, in a randomized fashion across multiple hospitals. In collaboration with the Yale School of Medicine, the Yale New Haven Health System has built a real-time, automated alert system for AKI. With the NIH funding, Wilson’s lab will perform the randomized, controlled trials over five years at six Yale hospitals to test the efficacy, efficiency, and usefulness of their alert system. 

“Our overarching goal is to determine how alerts for AKI can be engineered to provide benefit to patients and whether that benefit can be enhanced with intelligent targeting,” says Wilson. “In brief, what we’ve designed are ‘pop-up’ alerts that occur when a provider tries to access the medical record of a hospitalized individual with AKI, providing them information about the patient and access to an ‘order set’ — a prepackaged group of evidence-based directives that apply to a particular diagnosis — to help with a safe and quick diagnostic workup.”

For example, although it has been well-documented that there’s a causal link between non-steroidal anti-inflammatory drug (NSAID) agents and the development of AKI in hospitalized patients, many providers continue to administer NSAIDs to patients who’ve developed AKI. Therefore, one of the alerts in the system they are testing will give repeated encouragement to the provider to discontinue NSAID use in a patient who’s developed AKI.

Rather than roll out their alert system for immediate hospital-wide use, Wilson’s team is taking a measured approach based on the relative ineffectiveness of other hospital alert programs. “Because alerts have proliferated in medicine, and our everyday lives, more alerts may actually hinder care,” says Wilson. “We want to make sure, in as rigorous a way as possible, that this benefits patients before we insist upon broad adoption.”

“We’re very excited about the NIH funding,” says Wilson, “and think the project will demonstrate that Yale is using the most robust academic tools to ensure the health and safety of our patients.”

This research will be funded by NIDDK grant DK113191.