Payers and providers need to examine all activities, not just medical activities, that impact the care of a population. That’s why population health analytics are so critical.
When managing population health, it’s important to have analytics to harness data to better understand outcomes and identify patterns of causality. This is especially important when evaluating relationships with patients who have complex issues, behavioral health conditions, and/or social determinants that materially impact their health. Can your analytics solution determine how your population health programs are performing and turn that analysis into action to improve care management?
Typical analytics-first tools stratify risk, identify and track cohorts, and assess results—providing information that is important, but insufficient because it doesn’t address what to do if those outcomes are not what you want. The only way to improve is to have visibility into the success of the activities you perform, not just the outcomes of those activities. The insight into those activities—at a level of detail that enables measurement—will help you figure out what actions are the most effective in making your population well.
Put another way, most outcome measures focus on clinical conditions and utilization, relying on clinical or claims data to assess what has happened and give insight to support improvement. Without knowing what those activities are and measuring them, you can’t know what permutations of those activities are the most effective.
Analysis that measures the nonmedical activities that go into delivering care can’t typically be performed based using traditional data sources such as EHRs and claims that provide information about the patient, the patient’s condition, procedures performed, and payment factors. No matter how much number crunching analytics vendors perform, they can’t work with information they don’t have.
That’s why payers and providers need to examine all activities, not just medical activities, that impact the care of a population. These activities yield a different kind of data—data that enables you to assess your work and optimize your processes outside medical treatment so that the outcomes can be the best possible.
This approach is in harmony with the analytics most organizations are doing today—that is, creating and prioritizing cohorts to focus on, and measuring the outcomes for those cohorts based on the medical activities performed. Population health analytics augment this picture by focusing on the activities that remain unanalyzed.
In the arc of value-based care realization, relying on outcomes as proxies for insight on how to improve is a self-limiting strategy. To move beyond score cards into improvement plans, your analytics need to explicitly account for the work being done.