Covid-19 ADT Notification and Case Management


Personal Experience with COVID-19

  • A couple of us got COVID-19. Our COVID-19 post-discharge experience identified gaps of poor communication between hospital/ED and Primary Care Physician (PCP). This inspired us to solve this broken communication to improve post-COVID care outcomes for the patients.

Prior experience with Healthcare Patient movement and care coordination​

  • Patient movements across healthcare organizations are not well tracked and managed. Use of manual tools like fax, SMS and phone calls to track and communicate patient information.
  • When a patient visits an emergency room or is hospitalized then his Primary Care Provider would not know about it. Less than 10% time, PCP themselves reach out to patient after his/her discharge.
  • Burden is on patient to inform his/her PCP, resulting in poor patient satisfaction. Reliance on patient call can be unreliable ​
  • Medication and equipment needs are not communicated timely to Nursing homes more than 50% of the time
  • Increased healthcare cost due to adverse events, re-admission) and provider burden (locate, reconcile, and coordinate information)​

Centers for Medicare & Medicaid Services (CMS) requires notifications across organizations- effective 5/1/2021​

  • A new CMS rule requires hospitals send event notifications regarding admission, discharge, and transfer to other providers.
  • Real-time, using Secure eMail or through Health Information Exchanges.
  • CMS $ = 50% of a hospital payer mix. Given the large revenue potential that CMS requirement represent for all hospitals, there is a strong business driver for us to solve this problem!

What our solution does

  • Allows to identify COVID-19 cases early during patient’s ER or hospital admission, discharge
  • Easily manage the arrival of patients’ COVID-19 discharge status to post-acute providers such as Primary Care Providers.
  • Allows Case worker to review the COVID-19 case, notify and route to PCP and scheduler
  • Allows PCP to review and confirm the COVID-19 case and notify scheduler
  • Allows Scheduler to review the case and complete scheduling using a separate scheduling application.
  • Allows Case Manager to do workflow management and review KPI metrics using a Business Intelligence dashboard.
  • Supports patient's case management across different departments and organizations - Hospital, ED, Care Team, Scheduler and Primary Care Provider

How we built it

We followed an Agile process as 1-week Sprint over 6 weeks. For each sprint, we set the scope at the beginning and demo'ed the MLP capabilities to the business users at the end of each Sprint. The implementation was refined based on this feedback.

We worked with the Business Owner and Product Owner to elaborate and prioritize the requirements as user stories for each sprint.

We maintained a backlog. This determined the initial scope of our MLP.

Team Setup

Nine people team with diverse skills assigned specific roles.

  • Omar - Business Sponsor, End Customer representative
  • Nitin - Product Owner, Decision Maker
  • Nausheen - Scrum Master, Facilitator
  • Tony, Neelima - Architecture
  • David, Elisa - Developer (Lead)
  • Jo - Usability Tester
  • Yiqi - Technical support

Sprint 1 (Week #1: 3/1-3/5)

  • Define team roles and responsibilities
  • Conceptualized MLP scope:
    ❏ User personas,
    ❏ User story boards,
    ❏ Business work flows
    ❏ Screen mockups,
  • Hackathon environment setup, user account setup
  • Develop Test data

Sprint 2 (Week #2: 3/8-3/12)

  • Implement and demo Tech Story - System simulates the consumption of ADT Messages
  • Develop ADT message data model
  • Implement and demo User Story 1.1 - Case Worker assigns ADT Case
  • Conduct user research and interview

Sprint 3 (Week #3: 3/15-3/19)

  • Implement and demo User Story 1.2 - Case Worker notifies PCP
  • Implement and demo User Story 1.2.1 - PCP receives an email notification
  • Refine ADT message data model

Sprint 4 (Week #4: 3/22-3/26)

  • Implement and demo User Story 1.3 - PCP reviews and notifies Scheduler
  • Implement and demo User Story 1.2.2 - Case Worker assign severity to case

Sprint 5 (Week #5: 3/29-4/2)

  • Implement and demo User Story 1.4 - Case Worker notifies Scheduler
  • Implement and demo User Story 1.4.1 - Scheduler receives an email notification
  • Solicit user feedback from the demo

Sprint 6 (Week #6: 4/5-4/9)

  • Refine Look and Feel of UI and the workflow based on the user demo
  • Develop demo scripts
  • 5-minute Video Production
  • Prepare Hackathon Application

Challenges we ran into

Three challenges were encountered:

Learning curve and time available

We needed to get up to speed with the Pega low-code platform and develop all without impacting other project commitments. The training courses were too long to fit into this window.

Difficulty in fully defining user requirements

We took the draft flows and screen views but they were not sufficient to get an accurate real-user requirement. Subsequently we have defined the application with an Enterprise Architecture methodology which could have been used earlier but did not fit with the agile style.

Setting up dummy user accounts

The setting up of dummy accounts was not successful due to the requirements to have real email addresses. Eventually, we had to use real user accounts by assigning them roles.

Accomplishments that we're proud of

Incremental development of MLP

Despite the learn curve and available time, the MLP was developed incrementally as per the sprint plan assigned and was demonstrated successfully.

MLP Refinement response

The speed of code modification was fast enough to show changes right in the sprint review meeting. Patterns emerged which were reused between sprints to work with different personas.

Promise to solve a real-world problem

The demo of our MLP prove that we can solve a real-world problem. The user feedback is promising and we think the solution can be ready to be deployed at a real-customer.

What we learned

  • Once the learn curve has been overcome, development productivity is very high. We have not measured this but in the early stages of learning, productivity will not exceed conventional development techniques.
  • Narrow the scope down to keep it simple - easy to use and at the same time have all the functionality customers required.
  • Knowing the healthcare domain and the data from the inside out helped us to support every assumption with market insights.

What's next for COVID-19 ADT Event Case Management

We are interested to discover the opportunities to engage with the real prospective customers.

  • Replace the simulated ADT messages with the real HL7/FHIR ADT feeds.
  • Demo to our prospective customers
  • Work with Pega in identifying opportunities to pilot MLP with real-world healthcare organizations (federal and private sector)

Built With

  • agile
  • electronic-medical-record
  • github
  • hl7-adt-v2.x
  • infinity
  • pega
  • smtp
  • sql
  • teams
+ 10 more
Share this project: