Reducing COVID-19 Vaccine Waste: A California Case Study

A Serverless Web App tool for Vaccination Providers to optimize the amount of COVID vaccines needed to solve the "pooling" dilemma and ultimately minimize leftover vaccines.

Comment 1

Inspiration

  1. Initially, we came into the hackathon not really sure what to do--this is our first time attending a hackathon creating our first, fleshed-out programming project.
  2. After brainstorming of problems to solve, we decided to solve a problem related to COVID: there are many leftover doses, and if these collective doses were "pooled," [1] and if it were allowed by the FDA, these leftover doses could be used to vaccinate thousands of doses.
  3. However, pooling isn't allowed. Therefore, we decided to tackle a better design question around this: "how might we reduce the amount of leftover doses?"
  4. Our team discussed this problem with GCP Mentors Jon and Terrance. We discussed an optimization algorithm approach, but given our circumstances, we chose to create something simpler.

What is our product?

Our product is a serverless web app (powered by GCP Firebase) that allows COVID-19 vaccine providers to receive a recommended number of vials (with a margin of error) to prepare based on input patient information data of all the people who have signed up to get vaccinated.

In the current implementation, only California is considered as a case study/proof of concept with the potential to scale with GCP Firebase.

This recommendation is based on a couple of assumptions:

  1. Vaccination dosage depends on BMI.
  2. Average BMI can be found and aggregated by Zip Code.
  3. Vaccination providers have this patient data of people who have appointments.
  4. Patients reside at their stated Zip Code.

How it works

  1. Using Visual Studio Code and Git, we created and edited a simple web application with HTML and Javascript. We programmed a button on this web application to read and Zip Codes from a .csv file of patient data.
  2. These Zip Codes are then aggregated based on count. From here, we created a Submit button to send an API call containing the distribution of patients from Zip Codes to the GCP Firebase server. The API call is expected to receive a recommended number of vials to prepare. This concludes the front-end.
  3. On the backend, a collection of Zip Codes with the average BMI of those respective Zip Codes are stored on GCP Firebase. From this API Call, a distribution of patients aggregated by Zip Code is received.
  4. A classification is assessed based on the Zip Code's average BMI (ie underweight, normal weight, etc). Based on this classification, a recommended dosage is assigned. From this, a proportional amount of dosages is returned to the API Call, and sent to the front-end for the vaccination provider to peruse.

Challenges we ran into

  1. Identifying what GCP Services would best fit our use cases
  2. Figuring out how to get started as first time hackathoners was very daunting
  3. Javascript was difficult to learn

Accomplishments that we're proud of

  1. Accomplished and submitted something to our first hackathon
  2. Learned A LOT
  3. The learning inspires us to continue to learn and attend events like these

What we learned

  1. Setting up how to install and use Git/GitBash
  2. 1/2 of the team are first time coders
    1. What a for and if statement are, and how to implement logic
    2. How to index and iterate through a list
    3. How higher order functions work in Javascript
  3. Learned how to setup a serverless database through GCP
  4. How to use Google Cloud Platforms and utilize their data in our program

What's next for Reducing COVID-19 Vaccine Waste: A California Case Study

  1. Receive feedback from the community on what they think of the project idea
  2. Figure out different ways to solve this problem of how we might reduce the amount of leftover doses
  3. Understand how vaccines are administered to improve recommended dosages, and whether or not our assumptions are correct
  4. Learn more code and different coding language!

[1] "It's called "pooling" — and it's not a new concept. Pharmacists have been doing it for years with everything from flu vaccines to some chemotherapy medications to antibiotics. It involves taking what's left over in a drug vial and combining it with what's left in another vial to create a full dose."
link

Built With

Share this project:

Updates