ClosedLoop Announces Release of Free Open Source AI-based Tool to Identify Individuals Vulnerable to Severe Complications of COVID-19
- The CV19 Vulnerability Index (CV19 Index) is an open source, AI-based predictive model that identifies people with heightened vulnerability to severe complications resulting from COVID-19 (commonly referred to as “The Coronavirus”).
- The CV19 Index helps organizations identify the most vulnerable individuals and protect those individuals from becoming infected. Further, protecting vulnerable individuals helps to protect the healthcare system, healthcare workers, first responders and other vital community resources from becoming overwhelmed.
- The CV19 Index is being released through multiple channels and is available immediately at https://github.com/closedloop-ai/cv19index, on Google Cloud Marketplace & Amazon SageMaker platforms, and via ClosedLoop’s HIPAA compliant hosted platform. The CV19 Index is available for use through any of these methods at no cost.
- ClosedLoop has created a forum at https://community.closedloop.ai for experts to further enhance and extend the CV19 Index. Healthcare leaders in population health, public health, and hospital systems are encouraged to join the discussion and find ways to leverage this free tool and service to reduce the impact of COVID-19 in the United States and abroad.
AUSTIN, TX—March 16, 2020—ClosedLoop.ai, Healthcare’s Data Science Platform, announced the release of the COVID-19 Vulnerability Index (CV19 Index) — a free, open-source tool designed to help healthcare organizations identify and protect individuals that are most vulnerable to COVID-19. By releasing the CV19 Index as free and open-source, ClosedLoop aims to distribute this tool as widely and quickly as possible while leveraging the collective knowledge and experience of the open source community to quickly improve the Index.
The CV19 Index is available at cv19index.com with support for the following deployment models:
- Open source download
- ClosedLoop HIPAA compliant turnkey hosting
- Google Cloud Platform Marketplace and AWS SageMaker hosted models for virtual private cloud deployment
The CV19 Index identifies people likely to have a heightened vulnerability to severe complications from COVID-19. Because data on COVID-19 cases is not readily available, the CV19 Index was developed using similar proxy events (e.g. pneumonia, influenza) and calculates vulnerability in terms of a person’s near-term risk of severe complications from respiratory infections. The CV19 Index does not predict who will become infected with COVID-19 or identify where the virus might spread. It is only meant to identify people with a heightened risk of severe complications should they become infected.
“The CDC and WHO are providing excellent guidance regarding factors that increase a patient’s risk should they contract COVID-19. Unfortunately, roughly 55% of Americans on Medicare meet at least one of the CDC’s increased risk factors,” said ClosedLoop CEO Andrew Eye. “We can only protect people who are vulnerable if we can identify them, prioritize resources, and provide ways to help reduce their exposure. The CV19 Index provides a way for healthcare organizations to target their outreach and interventions to the most vulnerable patients.”
Why a Vulnerability Index is Important
Early studies from other parts of the world have shown that the risk of severe complications and mortality is higher for certain vulnerable populations, particularly people who are elderly, frail, or have multiple chronic conditions. Overall death rates were 1%, but rose to 6% for people with cancer, high blood pressure, and chronic respiratory disease; 7% for people with diabetes; and 10% for people with heart disease. Risk also rose sharply with age, with a death rate of 15% for people 80 or older.
“It isn’t straightforward to identify who is the most vulnerable,” said Carol McCall, Chief Health Analytics Officer, ClosedLoop. “People with the same chronic condition don’t have the same risk, and simple rules can fail to capture complex factors like frailty. Our results revealed that the top 5% of people represent 46% of the risk. Having this kind of precision is extremely useful and helps organizations focus where they’re most needed. If we want to reduce the impact of COVID-19, we must protect the vulnerable. To do that, we need to know who they are. The CV19 Index can help.”
Calculating the CV19 Index
The CV19 Index was built and tested using medical claims data from more than 2 million elderly and disabled individuals, with features that reflect a person’s medical history, chronic conditions, comorbidities, prior adverse events, recent medical treatments, functional limitations, and degree of frailty. Results show it produces accurate, unbiased predictions with initial results showing ROC AUCs equal to .85 and sensitivity of 46% at 5% alert rates.
Using the CV19 Index
As community actions move from containment to aggressive mitigation, healthcare organizations and community health professionals can use the CV19 Index in several ways including:
- Identifying and targeting vulnerable individuals for outreach
- Preparing and sharing resources such as food delivery, virtual worship, medicine deliveries, telehealth, regular check-ins etc.
- Prioritizing resources as healthcare organizations prepare for shifts in demand
- Coordinating efforts with other organizations and agencies
“It’s going to take a community effort to stop the spread,” said Eye. “We just want to do what we can to help make that happen. We want organizations to jump in with us, to help make the CV19 Index better. If that happens, we’ll have a powerful tool to mitigate the impact of COVID-19 to society at-large.”
ClosedLoop is Healthcare’s Data Science Platform. We make it easy and affordable for healthcare organizations to use data science to improve quality and reduce costs.
The ClosedLoop.ai platform is purpose-built to make healthcare-focused data scientists more effective at their jobs. By providing off-the-shelf models for common healthcare use cases and automating many of the manual processes involved in traditional data science tasks, ClosedLoop allows data scientists to focus on the impactful problems that drive real value for organizations. A finalist in the Center for Medicare and Medicaid Services’s AI Health Outcomes Challenge, ClosedLoop is headquartered in Austin, Texas. For more information, visit closedloop.ai and cv19index.com.