Research shows Israel's MedAware can reduce prescription errors, save lives

Adverse drug events (ADEs) – harmful effects resulting from taking a medication – are one of the three most common and harmful categories of medical errors, causing 100,000 annual deaths in the U.S.

MedAware (photo credit: Courtesy)
(photo credit: Courtesy)
Medication errors account for a $21 billion drain on the American healthcare system alone, not including the cost of legal action taken when such errors occur, explained MedAware CEO and co-founder Dr. Gidi Stein.
And adverse drug events (ADEs) – harmful effects or injuries resulting from taking a medication – are one of the three most common and harmful categories of medical errors. Each year in the United States, approximately two million ADEs cause a staggering 100,000 deaths, according to the NGO Patient Safety Movement.
But now, new research shows that there is an Israeli company that can help.
MedAware and Sheba Medical Center recently revealed new research validating the clinical impact of the company’s machine learning-enabled patient safety platform designed to minimize medication-related risks.
The findings were published in the Journal of American Medical Informatics Association in a study titled, “Reducing drug prescription errors and adverse drug events by application of a probabilistic, machine learning-based clinical decision support system in an inpatient setting.”
Physicians at Sheba analyzed results in a single medical ward, from a hospital-wide live implementation of MedAware, which had been integrated into the center’s existing electronic medical record system. The platform monitored all medical prescriptions issued over 16 months, with the department’s staff assessing all alerts for accuracy, clinical validity and usefulness, recording all real-time responses of physicians to alerts generated.
The results of the study demonstrated a low overall alert burden, with MedAware-generated warnings for only 0.4% of all prescriptions.
“Today’s widely used, rule-based systems for prevention of medication risks, including prescription errors and adverse drug events, are unsuccessful and associated with a substantial false alert burden,” said Dr. Gadi Segal, Head of Internal Medicine “T,” who led the study. “These alerts are ignored in nearly 95% of cases. Our study demonstrates that MedAware’s patient safety platform – which leverages a probabilistic, machine-learning approach based on outlier detection – can significantly minimize such risks, with high physician acceptance of MedAware warnings that result in physician behavior change and increased patient safety.”
ADDITIONAL FINDINGS included: 60% of warnings were generated after a medication had already been dispensed following changes in patient status; 89% of all alerts were considered accurate; 80% of all alerts were considered clinically useful; and 43% of alerts caused changes in subsequent medical orders.
“Given the challenge of medication safety and its significant impact on patient care, we elected to work with MedAware when the company was still proving its concept,” said Dr. Eyal Zimlichman, Sheba’s deputy director and chief medical officer. “After years of partnership, our research team set out to assess the clinical impact of the live implementation of MedAware’s platform, and the results speak for themselves.”
Stein founded MedAware in 2012 after reading about a deadly medication error in Israel, in which a physician prescribed an anticoagulant blood thinner to a nine-year-old boy that killed him a week later. The doctor meant to click the medication above the one he had prescribed in the electronic prescription service and didn’t notice his mistake until it was too late.
“It was a typo,” said Stein. “But typos can kill.”
As a computer scientist and physician, he said: “I could not live with myself if anything happened like this. So, I set forth to try to solve the problem.”
MedAware uses AI methods similar to those used in the finance sector to stop fraud, by identifying “outliers” from a trend or practice in order to recognize suspicious or erroneous transactions. Stein said that when a person owns and uses a credit card, the bank begins to track the individual’s personal spending patterns. When an unusual transaction is made, the bank is alerted and can contact the owner or put a hold on the credit card.
“We are trying similar methodologies with healthcare data,” said Stein. “Prescription patterns of thousands of physicians treating millions of patients are used to determine the ‘normal’ treatment spectrum. A prescription largely deviating from this spectrum is likely to be erroneous.”
MedAware’s system is fully personalized in that its responses are based on each patient’s specific data. The system is likewise self-learning, with no rule set that limits the errors it can capture.
“It tells you that it would not be advisable to give a certain medication because it does not suit the patient’s profile,” explained Segal. “Only machine-learning software that knows the patient’s profile would know to tell me, ‘Hey, doctor, this is a typo, or the wrong medication, or you’re too tired – you cannot give insulin to a patient that doesn’t have diabetes.’”