As a data scientist, I focus on making data meaningful and relevant to the business of healthcare. I study and analyze patterns and trends in multiple data sources from multiple systems, and extract insights from that information for solutions to healthcare’s most pressing challenges.
The data scientist is a relatively new role within organizations that have begun working with and using large, complex data sets to identify strategic opportunities. Earlier this spring, you met one of Geneia’s top data scientists, Jin On.
In healthcare, the strategic opportunity is to use data to improve health and satisfy the needs of patients and providers.
I’m often asked how this process works. How do you develop technology solutions that address these healthcare challenges? Put another way, how do you make technology personal? The quick answer is to make healthcare more data driven. Peter Norvig, the director of research at Google once said: "We don’t have better algorithms. We just have more data." This was true in most industries but not healthcare until recently. With the recent changes in regulations that encouraged the adoption of electronic health records (EHR), as well as the increased adoption of devices like smartphones and Fitbits, we’ve seen an explosion in the amount of available healthcare data. We can now use clinical and claims data, coupled with personal-device-generated data, to create a complete profile of a patient’s health.
As healthcare transitions to a value-based model (as opposed to a fee-for-service model) where providers can focus on and are reimbursed for keeping patients healthy, identifying patients at high-risk for adverse outcomes becomes an imperative. Effectively evaluating risk and identifying appropriate interventions are perhaps the most challenging aspects of patient care. This is where the availability of data and data science techniques can have the greatest impact and where technology becomes personal. Using all available sources of data – from claims and EHR to pharmacy, laboratory, psychographic and patient-reported – data scientists can develop sophisticated models that not only identify patients at risk for a particular condition but also learn from the additional information generated following treatment for that condition so the model can continually improve. This is what is known as machine learning.
At Geneia, we focus on understanding our clients, the challenges and barriers they face, and the solutions needed to keep them moving forward.
It is not the job of healthcare professionals to figure out how to make the technology work for them; it is up to us as data scientists to find meaning in the data sets and determine how best to use the insights gained to develop technological solutions that address particular healthcare challenges.
For example, care team members need very specific and detailed information about their population that is easily accessible and actionable at the point of care. They also rely on a variety of data sources and what it means in the aggregate to determine the most appropriate course of action. Technology professionals can use this insight about the needs of care team members to develop specific technology solutions that: 1) make it easy to access information from anywhere at any time (through cloud-based portals, using customizable and easy-to-understand dashboards), 2) build a robust technology infrastructure that can import, normalize and apply learning algorithms to vast amounts of disparate data and 3) use predictive modeling to help care team members make decisions and plan courses of action based on future risk.
Taking it a step further, the future of healthcare technology lies in solutions that can materially impact health.
For example, consider a drug that effectively treats blood clots for 70 percent of the population, but is ineffective for the other 30 percent. The reason it isn’t effective with that 30 percent is due to a combination of various genetic factors, family history and personal characteristics. This is where data science can really make a difference. In this example, it can be used to build a model that would interpret the data to accurately determine which 30 percent of the population would benefit from an alternative clot-busting medication. The healthcare provider would know this up front and could prescribe the appropriate treatment immediately, saving cost and providing better care. This is the essence of personalized medicine.
At the end of the day, the question for data scientists like myself is still the same: does the technology solve the challenges in healthcare?
Does it provide care teams the information and tools they need to make informed decisions that will improve patient health and lower costs in the system? That is our goal as data scientists and technology professionals at Geneia – to make technology personal and improve the lives of the people we serve.