Harnessing nontraditional data sources can drive better health outcomes
How does access to a sidewalk correlate to the risk of premature death? An engineer would likely point to transportation research that highlights the benefits sidewalks provide to pedestrian safety. In health care, sidewalks are just one example of an environmental factor that contributes to an individual’s and a community’s health.
It’s been over 10 years since the New England Journal of Medicine published a study that suggested medical care accounts for 10% of a person’s health, with behavior accounting for another 40%, social factors contributing 20%, and genetics the remaining 30%.
The digital revolution has increased the amount of data available to health care professionals. We can now utilize advanced analytics and artificial intelligence to unpack health histories from electronic health records and combine them with health insurance claims data and patient-reported information from clinical research to create a comprehensive picture of health. Equipped with this information, doctors and nurses are able to improve how they care for patients and reinforce positive health behaviors, but they still have limited visibility into what happens when the individual goes home.
But while the health care industry has made great progress in improving patient care, the algorithms that guide informed decisions are only as good as the underlying data. Let’s take a look at how new data sources are revolutionizing how our industry operates and driving better health outcomes.
Better health care data with social determinants of health (SDOH)
As mentioned earlier, 60% of our health outcomes are influenced by behavioral and environmental factors: where we are born, grow, live, work, and age. For example, how can we possibly expect someone to regularly check their blood sugar, adjust their diet, purchase new medication, stand on an internet-connected scale, and communicate through an app back to their doctor if they don’t have stable housing and quality food sources? Addressing these social determinants has the potential to impact our health and better inform medical care, health policy, and patient behavior.
Until recently, it has been difficult to get “out of the system” data into the hands of health care professionals. But technology and rapidly expanding data sources, such as information stored in the cloud from wearable devices, allow us to build more insightful analytics and share pertinent information with health care providers and patients alike.
Expanding digital data through the Internet of Things (IoT)
The Internet of Things (IoT) is a term we use to describe the ecosystem of connected devices, like smart watches, electronic blood glucose monitors, and even pacemakers. For consumers, the benefits of these devices range from convenience to lifesaving and, for providers, they offer access to relevant information about a patient’s health status and activity levels. They are also linked to consumer preference about their health care provider. In fact, one survey found that 62% of patients said they would choose a doctor who used data from wearables over one who did not. And the devices can help doctors and patients focus on developing healthier behaviors. A major health insurer recently integrated a new Apple Watch® feature into a program that rewards participants for meeting health goals tracked through wearables. Early data shows that the employers that offered the program to their employees saw year-to-year savings of $222 in health care costs per person. People with chronic conditions were among the most avid participants. Members with diabetes were 40% more likely to enroll compared with people without a chronic medical condition.
Previously untapped medical data
Advancements in analytics are allowing us to leverage medical information in new ways. For instance, while much of a doctor’s visit gets captured in an electronic health record, the notes the care team enters into free text fields are unstructured, meaning they can’t be recognized or analyzed the same way that numeric values like lab test results can. This means that important clinical observations can slip through the cracks. Clinically intelligent natural language processing (NLP), an emerging form of AI, can parse this data quickly, including diagnoses, procedures, findings, labs and drugs, and outcomes.
Widening the health care data lens
Improving health requires treating people, not just the symptoms that are captured in an electronic health record. Health care needs to widen its lens beyond the four walls of a hospital or a doctor’s office and better understand how consumers behave and what it’s like to live in their home communities. We’re on the cusp of being able to harness nontraditional data sources and build more discerning analytics to guide informed decisions across the industry. With the movement toward incorporating nontraditional data, we are also seeing an increase in collaboration between care providers, health plans, employers, government agencies, and life sciences organizations to improve the consumer experience and health outcomes and better manage cost – something we in health care refer to as the “triple aim.”