As state governments look to slow the spread of COVID-19, they are having to stand up massive contact tracing efforts practically overnight. When deciding how to best tackle the new project, the state of Maryland turned to a cloud-based call center.
During an Aug. 11 MeriTalk webinar, Lance Schine, deputy secretary of information technology for Maryland, shared how the state worked with Amazon Web Services (AWS) to leverage cloud-based call center solutions to rapidly set up a contact tracing program.
The webinar, sponsored by AWS, delved into why cloud-based solutions were critical to rolling out contact tracing, how Maryland set up its program, and best practices for other states looking to set up their own contact tracing programs.
Schine explained that contact tracing isn’t new; it’s been used to track E. coli and other disease outbreaks for years. However, the coronavirus pandemic has massively increased how much contact tracing has to be undertaken. Schine said that while counties and cities across the state had contact tracing programs in place, there were significant discrepancies in what those programs looked like. Smaller counties were doing things by pen and paper, with no technology in place. Obviously, that wouldn’t work for the scale needed.
Other issue states faced when trying to ramp up contact tracing were shutdowns and stay-at-home orders. Schine said that with the stay-at-home orders in place, call centers weren’t open. Reopening the call center – and having 200 people all together in a building – wasn’t an option. So, the state had to adapt to a virtual call center, with hundreds of contact tracers working from home. Amazon Connect, the program Maryland selected, handled all the telephony aspects of contact tracing, managing in-bound and out-bound calling, as well as automated outreach to citizens, explained Dominic Catalano, WWPS specialist solutions architect, AWS.
Schine credited Maryland’s success to a decision by Governor Larry Hogan. He required all counties and the city of Baltimore to use the state program for contact tracing. Schine said that consolidating data across 24 different systems would “have been impossible.”
When it came to choosing what to deploy, Schine reiterated that his team had to make the decision quickly.
“This happened very rapidly,” Schine said. “We looked at high level requirements. We needed to use products that were as close to off the shelf as possible, but there wasn’t anything that did what we needed to do, so we had to build something.” On top of that Schine said his team “made the decision to work with products that are very low code, but also incredibly powerful and dynamic in their abilities to handle any intake or workflow requirement.”
With those requirements in mind, the state ended up opting to work with Salesforce and AWS. This wasn’t the first time Maryland had to turned to AWS to meet its technology needs.
“The state has already chosen AWS as its cloud provider, so we already have a shared services platform that has already been developed,” he explained. “That shared services platform was set up with the necessary security, governance, and controls to handle the most sensitive information. So, we knew we had a platform that was already built that we put dozens of applications on already. It was clear that we were going to move forward with that platform since we were doing everything we could to launch in weeks.”
The decision to go with AWS and Salesforce has allowed Maryland to adapt its contact tracing program as more knowledge about how COVID-19 spreads comes to light. Schine said the flexibility of the AWS platform enables the state to improve and bring in new features and functionalities as needed.
Schine underscored the importance of rapid development and deployment of a contact tracing program.
“We knew the sooner it was launched, the sooner it would save lives,” Schine said. “It was effective because just within the first 10 days or so, we kept over 2,000 people off the streets that were exposed [to COVID-19] and ended up having symptoms.”
To learn more, listen to the full webinar.
