Responding to
COVID-19 Together

This Is What The Flu Looks Like

Nick Stepro, Simon Ioffe. 2020. D3.js, with Illustrator. Data sourced from Commercial, Medicare and Medicaid claims from Arcadia Benchmark Database

The 2017 – 2018 influenza season was the worst on record since the 2009 H1N1 swine flu. The CDC estimates 45 million people nationally caught the flu, with 810,000 hospitalizations and 61,000 deaths.

In 2020, viral outbreaks are on most of our minds. While mortality and transmission rates of the novel coronavirus (COVID-19) far exceeds those of the 2017-2018 flu, a look backwards can still be instructive. Even with its reduced virality, and the presence/distribution of vaccines, the 2017-2018 outbreak was explosive.

Looking at Massachusetts transmissions using medical claims data, we visualize not just the growth rate of total cases but the algae-like “blooms” of infections. A few small clusters at thanksgiving bloom into thousands of cases by new years, each interaction a potential for transmission. For the purposes of illustration, the artist draws a linkage between each new case and the closest person that had a diagnosis four days prior. This simulates the transmission graph as a social network of sorts, and highlights the importance of current calls for “social distancing” as we deal with an ever more deadly virus.

A Population Under A Microscope

Samir Farooq, Michael Simon, Nick Stepro. 2019. Python, D3.js, with Illustrator. Data sourced from Commercial, Medicare and Medicaid claims from Arcadia Benchmark Database

Care management programs are not one-size fits all. Rather, patients have subtle characteristics requiring tailored patient-centered plans of care. Similarly, the extent to which the patient will benefit from the targeted program varies as well. Identifying those who will be most successful in a given care program takes a combination of medical expertise, research, and mathematical estimation. Arcadia quantifies this “impactability” with learning algorithms that look through hundreds of thousands of past cases, finding shared attributes in those that had the greatest success.

This illustration analyzes about 64,000 patients and their predicted impactability (main diagram) along with their characteristics (sub-diagrams) to answer the question: what characteristics are the strongest indicators of success in complex care management?

Taking the population under a microscope, by applying t-SNE (a dimensionality reduction technique) and subsequently studying the resulting clusters, we find that four distinct groups of patients exist that have high impact scores, and many more small groups. While the largest group is composed of the most predictable at-risk patients (high cost and complex conditions; top left), the other groups show: 1) a unique pattern of avoidable and emergent Emergency Department utilization lead to high impact scores (top right), 2) a particular combination of chronic, psychosocial, and frailty conditions lead to high impact scores (bottom right), and 3) certain socioeconomic traits, along with a set of specific conditions, lead to high impact scores (bottom left).

Arcadia’s impactability algorithm consumes more than thirty specific risk factors – nearly impossible for care managers to keep track of over all patients, but easy for the algorithm to combine into one risk score tailored to care management. This algorithmic assist not only finds more actionable patients, but also lets program coordinators repurpose their valuable time from patient identification to patient engagement and program design.

Harmony Of Data Integration

Jeff Soloman. 2018. D3.js SVG, with Illustrator. Data from Arcadia integration logs, with metadata from leading EHRs and health plan claims data.

The complexity of the data integration challenge in healthcare, is often ethereal, or even escaping. It is near impossible to instinctively grasp the sheer magnitude of data, the complexity of connecting it, and the value of simplifying it. Years of experience in the industry make it easier. But it requires a deep understanding of numbers that have been proven outside of human capabilities to understand. In this work the artist takes an emotional and stunningly beautiful approach to helping the reader see how 2 billion health records across 58 data sources, 3,909 tables, across 1 million patients, and 27 different vendor systems can be simplified.

The artist’s use of a chord diagram helps emphasize the connectivity and relationships between the source data and Arcadia data warehouse. The imbalance between the almost 4 thousand source tables and the 27 destinations creates an emotional connection with the movement and flow of data that even those familiar with a chord diagram may find jarring.

For those who want to look further, the artist shares two other dimensions in the data. The size of each data source is represented as an outer ring, and the size of each of the 3,909 tables in depicted in the bar charts surrounding the edge of the chord diagram.

Through this work, the artist hopes the viewer leaves with a clear understanding of what are otherwise staggeringly incomprehensible amounts of data, and an emotional connection to what happens when you simplify and unify healthcare data.