Community-based genetics study uses SAS® machine learning and artificial intelligence to improve population health in Nevada
Heather Allen’s health was deteriorating by the day. A young mother and high school math teacher, Allen was experiencing inexplicable symptoms, including fatigue and hard-to-fight infections. Doctors searched for answers, but nothing seemed to work, leaving them perplexed by her changing symptoms and conditions.
Then one night everything changed.
The previous year, Allen was one of 10,000 Nevadans to sign up for a groundbreaking health and genetics project called the Healthy Nevada Project. A recent breakthrough in the project allowed for further DNA testing, and soon after, her genetic results came in. Allen learned she has a high propensity for Alpha-1, a genetic condition that causes lung and liver disease. Tests confirmed the DNA results and made an official diagnosis, and soon the mother was on a path to monthly treatments.
“If not for the Healthy Nevada Project, I believe I’d be dead,” Allen says.
Reversing poor health outcomes with analytics
The Healthy Nevada Project, developed by Renown Institute for Health Innovation (Renown IHI), is one of the first community-based population health studies in the United States. By combining genetic data, environmental data and individual health information, researchers and physicians are gaining new insight into population health, enabling personalized health care while improving the health and well-being of entire communities in Nevada.
The Project comes at a time when the state continues to struggle with poor health outcomes and excess costs. Nevada ranks near the bottom of overall health rankings in the U.S. and suffers from high mortality rates for chronic conditions like heart disease, cancer and chronic respiratory disease.
“This was our call to action,” says Dr. Anthony Slonim, President and CEO of Renown Health.
To address Nevada’s most complex population health problems, investigators started by collecting data. Working in tandem with experts in environmental data at the Desert Research Institute, Renown Health fuels the project with de-identified electronic health care records, and supplements this with data from the Environmental Protection Agency and others.
“Everyone tends to focus on the genetics, but what we’ve created is a large data warehouse that has genetic data, clinical data, environmental data and social data,” Slonim says. “We know that clinical care is responsible for only 20 percent of your overall health status, with those other factors making up the rest.”
To see exactly how those other factors come into play, data scientists apply machine learning and artificial intelligence capabilities to the DNA results generated by personal genomics partner, Helix. This forms connections between participant genetic information and varying environmental factors, such as air and water quality, to see who might be predisposed to certain conditions. This could allow analysts, for example, to identify people prone to breathing problems and notify them to stay indoors when air quality is poor.
“We’re working to understand how environmental factors can help predict who may be at risk, allow for quicker diagnoses, and encourage the development of more precise treatments,” says Jim Metcalf, Chief Data Scientist of the Healthy Nevada Project. “The modern statistical and machine learning methods, along with the intuitive data visualizations made possible by SAS, have been critical elements of our success to date.”
The modern statistical and machine learning methods, along with the intuitive data visualizations made possible by SAS, have been critical elements of our success to date. Jim Metcalf Chief Data Scientist Healthy Nevada Project
An ounce of prevention
The Healthy Nevada Project benefits citizens and researchers alike.
For citizens, the project is one of the first broad genetics studies in the nation to return clinical results. Consenting study volunteers are learning of their risks for many serious genomic conditions including:
- Familial hypercholesterolemia, which has a genetic tie to high cholesterol.
- BRCA ½, a hereditary breast and ovarian cancer syndrome.
- Lynch syndrome, which is tied to endometrial and colon cancers.
For researchers, the project offers a unique opportunity to study pathogenic mutations and alert the people who have them. Recently, Project staff identified 90 people with FH, hereditary high cholesterol that’s related to increased risk for heart disease and not always diagnosed with cholesterol tests. With this information, administrators can contact these people and suggest they speak with a doctor.
“If we understand why people are at risk for certain diseases because of their DNA, we actually might be able to do something about it,” Dr. Slonim says. “We can give them the information, help them change their behavior and understand how to address those preconceived risk factors.”
Insight from the Healthy Nevada Project also helps Slonim with resource allocation. By knowing that a significant percentage of the population has a certain condition, he can implement prevention programs and hire doctors with the skills to treat those patients in five to 10 years.
“This is the ultimate strategic planning process for our community,” Slonim says. “If we can uncover things that put people’s health at risk, our health care providers can do appropriate screening and, ultimately, take better care of our people.”
Moving from patient health to community health is a large step that takes a lot of analytical power. “The flexibility in SAS allows us to ask for something in these reams of data and get an answer within 15 minutes,” says Slonim.
Metcalf agrees. “We’re living inside virtually all of the machine learning and AI procedures that SAS has. We are going to be solving some big problems in models with hundreds of variables. And so the ability for SAS to scale is crucial.”
Healthy Nevada Project – Facts & Figures
participants (and growing)
connections with machine learning
Aspirations for a healthier America
Slonim aspires to make Nevada the country’s healthiest state, and one day use the Healthy Nevada Project to power a Healthy USA Project. More than 35,000 northern Nevadans have already donated their DNA, and soon the project will expand into southern Nevada and other health systems around the country.
“If we bring on just one major health system per state, this could really advance the work needed from a national perspective to get a grip on chronic disease,” Slonim says.
The health system President and CEO welcomes those seeking advice on similar projects. “Come to us,” he says. “We’ve acquired a lot of knowledge over the years. We’ve hardwired the consents. We’ve hardwired the institutional review board approvals. We’ve hardwired the reports. Don’t recreate the wheel. Let us help you do it. And let us help you evolve.”
The mission is personal for Slonim, a cancer survivor who credits early intervention for saving his life. “I’m a believer in this and this is what personally for me drives the project. I believe we can make large communities, large states and perhaps even America healthier if we use these tools for the benefit of good.”
The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.