A Cure For Typos: A Simple Solution For A Serious Matter – Interview With Dr. Gidi Stein, Co-founder & CEO Of MedAware

Dear medical futurists,

MedAware was founded following a tragic story of a primary care physician, treating a 9-year-old boy with asthmatic symptoms. Using his computerised physician order entry system (CPOE), the doctor mistakenly clicked on the next entry of the medication pull-down menu list instead of the medication he wanted to choose. 

The physician, the pharmacist, and the parents all missed this typographical error, and the boy died a week later. This horrible story led to the conception of MedAware, a company focusing on correcting such issues with the help of A.I. and smart thinking. I talked with Dr. Gidi Stein, Co-Founder & CEO of MedAware if there’s a cure for typos.

What did you think when you first heard of this event explained above?

I was baffled by this story. I wasn’t aware of how easily physicians can kill with a typo or a misclick. One would have thought that there would be a smart prescription “spell checker” in place, but apparently, this was not the case. It could have easily happened to me, as a prescriber, or to one of my kids, as a patient. So, I thought I should do something about it. I was convinced that by applying A.I. and outlier detection algorithms on large scale medical records, we could accurately identify these “typos” and save lives. And this is how MedAware was born.

What is the estimate, annually, how many people are affected by medical errors due to spelling?

Medical errors are the third-leading cause of death in the US, with a reported 250,000 annual deaths due to medical mistakes. This results in as many as 4 out of 10 patients harmed in healthcare settings, with up to 80% of those medical errors preventable. Although typographical errors at the CPOE represent a major and unaddressed medication safety issue, in live settings, we find that the majority of the dangerous medication situations we prevent are evolving adverse events that emerge due to changing diagnostic situations, like med-lab or med-vital incompatibility, personalized laboratory trend outliers, and other patient-specific risks at any point following the medication ordering event. Given the multi-faceted nature of the challenge, we set out to provide a comprehensive solution covering medication-related risks from multiple angles.

Rule-based solutions mainly focus on drug interactions or dosage and allergy issues but are not technologically capable of addressing typographical errors (patient or drug mix-up) or evolving adverse events post-prescribing (lab or vital irregularities).

That’s where AI comes in. According to a recent Sheba Medical Center prospective inpatient study, most of these clinical decision support systems hold a 16% or lower accuracy rate, resulting in “alert fatigue” associated with the blanket dismissal of alerts by the providers, even when a warning is warranted.

You also have clinical data backing up your concept – can you please tell us more about it?

MedAware has validated its technology and methodology on more than seven million patients to date in clinical studies, and in multiple live deployments in inpatient and outpatient settings.

Two studies by Harvard Medical School (2017 and 2019) and two studies by Sheba Medical Center (2019 and 2020) have shown the following concerning MedAware’s system:

It protects junior prescribers in intense high-workload settings

  • 85% true alerts with clinical value
  • 70% of interventions are unique to MedAware’s system
  • 60% of interventions are for post-dispensing ADEs
  • 40% of interventions lead to immediate behavioural change
  • Significant direct cost savings, even in an outpatient setting

Besides a spell-checking algorithm, what do you think could help to avoid issues like these in the future – do you have any suggestions? Is there another A.I. to the rescue?

Acting as a smart safety layer, clinician-driven A.I. can be utilized to:

  • Monitor a patient’s medical profile to actively look for emerging risks and adverse drug effects when the patient is at home.
  • Identify personalised risk for future opioid dependency for opioid naïve patients when they are prescribed opioids for the first time.
  • Identify and suggest underdiagnosed clinical conditions and disease states to enhance patient care.

Sounds great.

Thank you,


Become a patron to

Unlock 687 exclusive posts
Get exclusive merch
Listen anywhere
Connect via private message