Empowering informed decisions with predictive health data and real-time insights during COVID-19.
Designed a data-mapping platform allowing government and medical leaders to intelligently assess community risk and readiness in response to the COVID-19 pandemic.
As COVID-19 began to accelerate its spread in Spring 2020, Anthem tapped our team at TM to lead an urgent project: figure out how to make the best use of health and economic data to serve timely community needs. In conversations Anthem was having with large employer groups and government officials, executives were increasingly aware of the need to help those groups understand how to make informed decisions around risk and readiness—and the tools they were painfully lacking. At the same time, Anthem partnered with leading data science providers CloudMedx and XY.ai, which supplied valuable data. TM stepped in as the organizing product strategy force, leading the charge in figuring out how to organize, prioritize, and design this partner data into a platform that could support government officials, local leaders, employers, business leaders, and individuals in understanding health risks posed to specific communities—and how medical systems could be impacted by changes in the health profile of their population as well as human behavior.
The result: C19 Explorer, a platform intelligently mapping health and economic data and offering predictive insights about pandemic risk and business readiness across the United States.
Problems
▸ Working at hyper speed
Information about COVID-19 and government guidance was shifting daily (if not hourly).Time was of the essence; the team had to move fast, balancing a rigorous design process and expediency. Phase 1 involved launching a resource hub within 2 weeks.
▸ Consistently mapping incompatible government, state, and local mandates
Guidelines at the state and country level were coming in different forms from different sources. TM had to devise a common pattern and rules for reading data from local, state, and federal, agencies, including the CDC.
▸ Designing predictive models that accommodate situational variables
From the start, C19 Explorer was meant to enable officials to take as many inputs from the past and present, and predict what the future could look like for a population. However, the model also wanted to be flexible enough to incorporate potential human behaviors. For example, what might ICU bed utilization look like over the next 30 days if schools and airports reopen, but keep people wearing masks? How about if people don’t wear masks?
▸ Normalizing disparate data sets into a common display pattern
There were two main data science partners (CloudMedx and XY.ai) and the data from these two sources wasn’t structured in the same way. TM had to standardize the inputs to design the most intuitive display patterns.
Maecenas sed diam eget risus varius blandit sit amet non magna. Aenean lacinia bibendum nulla sed consectetur. Vestibulum id ligula porta felis euismod semper. Fusce dapibus, tellus ac cursus commodo, tortor mauris condimentum nibh, ut fermentum massa justo sit amet risus. Morbi leo risus, porta ac consectetur ac, vestibulum at eros.
Federal, State, and Local policy makers
Policy makers are interested in learning about comparative health risks between communities so they can allocate resources, emergency and economic relief, as well as institute policy changes to mitigate any negative impacts on the health of the general population.
Hospital system administrators
Looking to model the impact of community health changes on hospitals in order to understand and manage the load on hospital systems under pressure. Administrators are interested in sharing information and reallocating resources (ventilators, beds), based on expected peaks and valleys.
Private sector business leaders
Looking to understand the status and risk of the communities in which they do business. Business leaders are interested in learning about how and when to safely re-open, protecting employees and assets.
Duis mollis, est non commodo luctus, nisi erat porttitor ligula, eget lacinia odio sem nec elit. Maecenas sed diam eget risus varius blandit sit amet non magna. Lorem ipsum dolor sit amet, consectetur adipiscing elit. Vestibulum id ligula porta felis euismod semper. Cras mattis consectetur purus sit amet fermentum. Morbi leo risus, porta ac consectetur ac, vestibulum at eros. Curabitur blandit tempus porttitor.
A rapid-launch landing page
The first, most urgent order of business was to get a resource center up fast. TM designed and launched an informational single-page site in under two weeks to provide quick access to common questions, reports, and resources, point people to Anthem testing resources, and offer information on how to get involved. The hub served as a home for the initiative as it was being further built out, explaining the goals and linking to news from the CDC and partners' websites that showcased relevant data.
Visualizing inconsistent data with consistency
Once the resource hub was launched, TM was able to turn its attention to Phase 2, an interactive data platform providing insights, trends, risk scores, and predictions based on community-level conditions. At its core, C19 Explorer is meant to provide visibility to risks across the economy, hospitals, and to public health imposed by COVID-19 at nationwide and community levels based on a range of variables. Someone can use C19 Explorer to easily understand how a community’s confirmed number of cases might impact anything from its food stability to the percentage of doctors without symptoms. While dropdown menus allow for seemingly limitless numbers of options in customizing the data, the hard work was under the hood -- normalizing all the data from multiple sources and creating a standard taxonomy that makes logical sense and allows for correlations and predictions. While the data is robust, the information is conveyed with clarity.
Looking back—as well as forward
The C19 Explorer platform is as much about historical data as it is a predictive look at the future. Government officials and hospital leaders can use the platform to model scenarios in order to predict the impact on human health and medical systems based on changes to community behavior and disease spread. This was only made possible by TM’s ability to design a one-size-fits-all approach to information analysis, spanning a geographic plane and translated over time. For example, in the above view, a community official could reliably predict what next month’s positive COVID-19 case count would look like based on historical trends across the US. They could then click into their local or neighboring states and counties to drill down for more localized data, and even model out the impact of potential human scenarios, such as an increased percentage of people returning to work.
Defining a community risk level
The biggest challenge with this project was figuring out how to make sense of masses of data. How could TM, Anthem, and the data science partners not just provide people with numbers, but actually support and help them? The answer came through applying a score to a community’s risk level. “We wanted to help people see at a glance, are things moving in a positive direction?” notes TM Design Director John Ashenden. The team designed an indexing system drawn from CloudMedX’s algorithm that assigns a risk score across a few dimensions: public health, the economy, social conditions, and hospital systems. A heat index—spanning from red (high risk) to orange (moderate risk) to yellow (low risk)—quickly brings visual clarity to tell the story geographically and over time.
Moving to community recovery
Because this platform ultimately aims to provide information as a solution, the team built a simple and clear framework for people to see a community’s “recovery” level. A referenceable system brings uniformity and ease across the country: on one end 1 means the community is in good shape, and 4 indicates that there is a lot of work to do to ensure the health system can handle an outbreak. More subtle shades of blue and violet signal optimism and confidence rather than trying to incite stronger emotions.
Finding quick and relevant insights
From the start, Anthem had visibility into common questions hospital administrators and government officials were facing. At the core, people wanted to understand the impact of the pandemic on their community and businesses. An insight center in C19 Explorer allows people to quickly find answers to the most common questions and supplies helpful analysis on top of data regarding areas such as infection rate and community risk scores. Intelligent AI models take into account relevant user data right from the start, continuously reflect user behavior, and supply the most current and useful recommended information and actions.
We were dealing with so much data, and we needed to find a common language. We figured it out by creating a framework for talking about “risk” and “recovery.” By leveraging the story of recovery, and building scoring mechanisms around risk and recovery, we are giving people tools to get where they want, to get BACK.
Widespread Usage
