Health IT, Startups

Health IT startup’s approach to prevent prescription errors gets validation in clinical study

Can a machine-learning enabled software platform prevent prescription errors before they impact patients? MedAware’s tool got some positive clinical validation in an efficacy study.

Prescription Drugs, Isolated On White, Clipping Path

An Israeli health IT startup has developed a clinical decision support tool to spot prescription errors before they happen. MedAware, the Israeli company behind the machine learning-enabled software platform, received clinical validation in a study to assess efficacy published in the Journal of American Medical Informatics Association this month.

MedAware’s tool analyzed medical records for 747,985 patients and generated more than 15, 690 alerts. A sample of 300 charts was selected for review from the 15 693 alerts generated, according to the JAMIA abstract. A coding system was developed and codes assigned based on chart review to reflect the accuracy, validity, and clinical value of the alerts. Three-quarters of the chart-reviewed alerts generated by the screening system were found to be valid. About 75 percent of these valid alert were clinically useful in flagging potential medication errors or issues, the study noted.

Ronen Rozenblum, the co-investigator of the study and an assistant professor at Harvard Medical School, shared his impressions of the study in a MedAware press release emailed to MedCity News: “There is a need for innovative solutions to address the current prescription errors that result in substantial morbidity, mortality, and wasteful healthcare cost. MedAware’s medication error detection system appears to have the ability to generate novel alerts that might otherwise be missed with existing clinical decision support systems.”

MedAware CEO and cofounder Dr. Gidi Stein referred to its software as a safety net in a phone interview with MedCity News. “We try to identify pathways by which physicians make errors.”

“We try to identify pathways by which physicians make errors.”

Some common errors by doctors, nurses and pharmacists can include clicking on the wrong medication option in a pull-down menu, prescribing medication to the wrong patient, or sticky fingers may lead them to accidentally add an extra zero to a prescription. MedAware’s approach involves looking at the context for a particular prescription and outliers that fall outside typical patterns and call attention to that prescription decision. These outliers are based on the machine learning component of MedAware’s product. The outliers prompt MedAware’s system to kick into action and force physicians to double check what they’re doing. Stein likened MedAware’s system to the financial sector and systems that activate in response to suspected credit card fraud.

Stein likened it to the financial sector and systems that activate in response to suspected credit card fraud.

“It is a continuously learning algorithm and its knowledge is derived from existing data. It can transfer knowledge from one institution to another. Nothing is programmed,” he said,

Although Stein acknowledged that several technology companies are active in the prescription error space, most of the attention has been directed at harmful drug interactions and allergic reactions patients may suffer from certain treatments.

This year, MedAware’s priorities include partnering with U.S. health systems, building on Israeli partnerships with Maccabee Healthcare Services Organization, and Tel Hashomer medical center. MedAware previously took part in accelerators with MassChallenge in Boston and Dreamit Health in Philadelphia.

Last year, MedAware formed a collaboration with medtech business Becton Dickenson for prescription error surveillance, which was funded by the Israel-U.S. Binational Industrial Research and Development Foundation.

Taking steps to reduce prescription errors is a matter of great interest as these errors cost up to $21 billion in medical costs for inpatients and outpatients combined, according to data from the Network for Excellence in Health Innovation.

Photo: Sezeryadigar, Getty Images