Don'tGotAlertFatigue

• 

  Patient records with several alerts with each priority shown.

Inspiration

Currently, CDSS(clinical decision support systems) provide help to clinicians by throwing alerts when necessary to avoid or act upon unsafe situations for the patient. In its current state though, CDSS end up throwing far too many alerts, where many of them are inconsequential and should be ignored. A 2018 study showed that clinicians often receive more than 100 alerts a day, and they become exposed to alert fatigue, where they become desensitized to alerts, and then ignore serious safety alerts that can put patient safety at risk (Shah et al., 2018).

What it does

Our iOS app aims to make the process for alerts much more customizable for physicians in a way that cuts down on alert fatigue and allows physicians to attend to the alerts based on severity and reasonable action that can be taken. At the base level of operation, Don'tGotAlertFatigue will be able to analyze sample patient records to generate different severity tiers of alerts based on manipulating aspects of the patient records. Ultimately, Don'tGotAlertFatigue will incorporate high level machine learning techniques to become better at categorizing and pushing out alerts to physicians.

How we built it

We build Don'tGotAlertFatigue using the SwifUI framework inside of XCode.

Challenges we ran into

Going into this hackathon, none of us were familiar with iOS app development so that was an area in which we had to become somewhat proficient in before we could begin the project. In addition, we weren't initially sure how to display the app's long term functionality on a smaller scale. Eventually, we decided this could be best done using sample patient records of four patients under one physician.

Accomplishments that we're proud of

We're proud of understanding Swift and being able to build an app in the short period of time that we were given. Additionally, we're proud of how we connecting the medical field along with computer science, and brought a new perspective on patient safety.

What we learned

We learned about how inefficient algorithms in healthcare can lead to cognitive overload of alerts to healthcare workers, which in turn can lead to serious harm for patients when these alerts are often dismissed. We also learned about how hackathons work, as it was our first one, and we understood a lot more about how to prepare, what's possible inside the given time frame, and what we should be spending our time on.

What's next for Don'tGotAlertFatigue

We plan on expanding the app functionality by building machine learning algorithms to filter out the alerts that are unaccounted for, and to also push low level alerts to the patients.

Built With

• swift
• xcode