Applied to health care, precision engagement means using an individual's unique environmental factors, preferences, and habits to drive adoption of a recommended health approach. As we continue the transition toward value-based care, health systems must use this approach, in concert with the increasing amount of data collected on patients, to improve population health management practices by positively influencing consumer health and wellness behaviors.
Progressive health systems today have successfully employed data science techniques to predict individuals at risk for adverse health outcomes. But effective prevention, which frequently depends on engaged patients, also needs an effective intervention—at the right time, and in the right way.
A vision for the future
In his recent book, "The Internet of Health Things," Dr. Joseph Kvedar paints a vision for a personalized health coach named Sam, along the lines of Siri, Cortana, or Alexa. Sam knows your goals, preferences, and personality type and is a cross between a drill sergeant, personal trainer, and occasional psychologist.
Using technologies similar to those found in a self-driving or semi-autonomous car, Sam monitors your vitals, can respond to your requests, and can also be proactive (see examples in the graphic below). As he collects data on you, he can hit on a motivational toolset perfectly tailored to help you manage your own health and health status.
While the technology exists for this kind of precision engagement, its realization is still in the future—we don't yet have the necessary data, systems integration, and consumer acceptance. However, personalization approaches used by health care organizations today suggest this future may not be so far off.
Data-driven engagement today
Omada is a digital health start up using data science to lead the way on personalized engagement. The company offers a 16-week behavioral change program called "Prevent" that guides participants toward health goals using a wireless scale, mobile app, online tools, health coaches, and a social support network. The coaching platform is the main component of Omada's data product architecture. Through this platform, coaches get recommendations to reach out to specific patients and suggestions for how to do so. Predictive, machine-learning based models that use digitally recorded program behavior data, demographic data, and longitudinal weight data inform these recommendations:
Omada's precision prevention approach has successfully driven engagement and is helping participants achieve impressive health outcomes. Approximately 65% of Prevent participants are still engaged with the program at 12 months, a figure significantly higher than the 6.6% average for leading commercial weight-loss programs.
Traditional provider organizations are in a different position than digital natives such as Omada—as a digital program that has served over 75,000 participants, Omada has the ability to collect billions of data points on participants' interactions with the program and outcomes. However, everyone in the health care delivery system can—and, as the above results prove, should—think about how the ability to collect more data on patient behavior, through digital health tools such as wearables and connected sensors, can power more effective approaches to patient engagement.
A different approach to personalization: An example from TriHealth
There are also ways to get started with precision engagement that don't depend on integration with digital health tools. TriHealth, a health system in Cincinnati, Ohio, employs psychographic segmentation (grouping people based on shared values, interests, and lifestyles) to personalize their health coaching program for patients with chronic conditions.
In a three-month pilot of this approach, TriHealth partnered with c2b solutions, a health care consumer insights and strategy firm, to classify 210 patients who work with health coaches to manage either diabetes or musculoskeletal disorders into five segments. Each of the five segments—direction-takers, balance-seekers, willful endurers, priority jugglers, and self-achievers—has unique motivations, approaches to health care, and communication preferences. To do this, patients participating in the pilot completed an online survey consisting of c2b's 12 segment classifier questions.
Alongside this process, TriHealth's health coaches were trained to understand each of the groups and how to best engage them in-person or over the phone during a coaching session. Individuals in one segment, for example, don't like to be told what to do by a health care professional and prefer to be given options rather than directive guidance, while those in another segment seek this directive guidance.
This personalized engagement approach yielded positive feedback from patients and coaches and resulted in an increase in personal health goal attainment during the pilot. Patients with diabetes, for example, achieved an over 90% increase in mean goals completed. TriHealth has since expanded the pilot to include additional coaches and patients.
What does all this mean for my organization?
A "one size fits all" approach to population health management and disease management that assumes that all patients with a similar health profile share the same needs, attitudes, and behaviors won't maximize patient outcomes. Providers should evaluate ways to leverage new or previously untapped sources of data, including patients themselves, to support more precise—and thus effective—engagement efforts. This personalization will not only improve health outcomes—it will also deliver a consumer-centric experience that drives success in today's competitive health care marketplace.
October 17 webconference: Health Care IT 101
Join our 30-minute session to learn about critical health IT capabilities and the rise of digital-enabled strategies in health care.
Just updated cheat sheets: Digital Health Systems 101
Download our cheat sheets so you can keep track of the fast-changing technologies and capitalize on opportunities for IT-powered innovation. Check out our guides for these topics and more: