In my most recent post, I wrote about usability factors in predictive analytics. In today’s post—the final post in the predictive analytics series—I’ll share an example of a provider that put all the predictive pieces together to transform its population health management program.
Sentara Medical Group uses predictive models to identify high-risk patients, particularly those at risk of hospital admissions. Sentara has 380 primary and specialty care physicians in Virginia and North Carolina. They recently received the NCQA’s highest level of recognition for their Patient Centered Medical Home (PCMH) program because of their commitment to chronic disease management and service excellence. They use Optum One—Optum’s combined analytics and care management platform—to take the PCMH model of care even further.
Sentara uses Optum One’s predictive analytics to stratify patients with congestive heart failure, chronic obstructive pulmonary disease (COPD) and diabetes by risk of future hospitalization. The organization selected 11 pilot sites called “Transformation of Care Sites.” Sentara’s quality team used predictive analytics to identify a small number of high-risk patients for each site, and equipped them with detailed individual patient information via a patient profile. This profile provided a 36-month view of an individual patient’s clinical parameters (e.g., BMI, blood pressure, ejection fraction), utilization parameters (e.g., ER visits), and treatments (e.g., medication changes). This visual picture of what is going on with a patient helps physicians recognize and act on gaps in patient care or changes in health status.
Sentara’s early results from its use of predictive analytics are promising. The analytics information has been well received by physicians and has had a significant impact on high-risk patient lists. In one practice, of the 44 high-risk patients identified, only one had been part of previous high-risk lists. In addition, rates of engagement in care coordination programs have improved, attaining more than 50 percent of eligible patients in some cases. Sentara is now expanding its use of predictive analytics to the remaining PCMH sites and is also introducing the pediatric asthma model as an additional tool.
Sentara is one of many providers beginning to integrate predictive analytics into their organizations. They are using it to help re-balance their care model in favor of more proactive care. By honing in on high-risk patients sooner and with more accuracy, providers can focus their resources where they will have the highest impact, and succeed in an environment rapidly moving toward value-based health care.
For more on how you can use predictive analytics to improve your population programs, download “Predictive analytics: Poised to drive population health.”
–Jeremy Orr, MD, MPH, Chief Medical Officer, Optum Analytics