By David Hom, Chief Evangelist, SCIO Health Analytics
Twitter: @SCIOanalytics
SCIO Health Analytics is a sponsor of the #HIMSS17 “Population Care Management Symposium: Clinical & Business Intelligence Road to Value-Based Care Success” program, and will be available for meetings during the Symposium on Sunday, February 19th, from 8 a.m. to 4:30 p.m. in Room W311A at the Orange County Convention Center in Orlando, FL
Under value based care, healthcare organizations are under pressure to improve quality resulting in improved clinical results. The problem: Many organizations are exerting considerable effort without really moving the quality needle in the right direction.
The upshot? Leaders are realizing that they need to understand the root cause of risk therefore leading in using data to drive insights that are meaningful. The key challenge is identifying what data is of value to improve risk and what the sequencing of the data should take to enhance the outputs.
Many healthcare organizations have taken the first steps on this path by using data to identify which patients belong to emerging high risk populations. To do this, they use claims data to determine the likelihood of a specific event such as surgical or hospitalization occurring among specific populations. For example, they might identify that certain groups of patients are likely to return to the hospital via the emergency department within a certain time period.
Such insight helps but it is not quite enough. To truly make a difference, organizations need to leverage data to identify exactly what actions will produce the best results in the most cost efficient manner and then which patients will respond to their efforts. The data needs to help the provider to know which care gaps have the greatest impact hence a prioritization logic needs to be an output.
To start, healthcare organizations can use data to assess impactability – which, according to a definition published in the Dictionary of Disease Management Terminology, “predict[s] which patients will acquire a disease, an adverse event related to a disease, or change from one health (functioning) state to another, where these outcomes are impactable with some specific intervention such as taking or stopping a medication, performing a test, reducing avoidable medical costs, making a behavioral change, or changing the person’s environment.”¹
In essence, when assessing impactability, healthcare providers gain a better understanding of whether or not interventions will result in real improvements – and, therefore, reduce some of the blood, sweat and tears associated with hard work alone. The analysis provides insight into where healthcare providers will get the best return on investment in terms of which diseases to target first and which care gaps to close first and what approaches to take. For example, such analysis could reveal that diabetes is a condition worth targeting. In addition, when assessing impactability, providers could determine if a diabetic patient education program would actually make a difference in the utilization of the emergency department or the number of inpatient admissions – or if it would merely be a “nice to have.”
More specifically, when leveraging data to measure impactability, organizations can work smarter by:
- Excluding the highest risk patients, where future hospitalizations can be very difficult or even impossible to prevent regardless of the intervention.
- Concentrating on patients who are considered to have emerging risk leading to more ambulatory care sensitive conditions. For example, developing interventions for patients with diseases such as heart failure and diabetes – conditions for which prompt, high-quality primary or outpatient care can reduce the risk of hospitalization.
- Focusing on patients who have a number of “gaps” in their care – as healthcare organizations can fill in these holes to produce more optimal care for patients. For example, with rheumatoid arthritis patients, not participating in physical therapy could be considered a gap. And, adding this intervention to the mix represents an easy, tangible step that can make a real difference in outcomes.²
Knowing what could happen by providing particular services to patients is not the same thing as knowing what will happen. To get a better handle on what will actually transpire when particular services are implemented, organizations need to look at intervenability — or the likelihood that the patient will become actively engaged in their own care and take advantage of preventive care interventions. Taking a page out of the retail industry’s handbook, healthcare organizations could segment consumers by type to realize if they are actually going to take action. With insight into a person’s expected behavior, organizations will have a better idea if there is any likelihood for change. As such, an organization would be able to determine if an education program would be likely to result in actual behavior change – or if it would turn into an in-one-ear, out-the-other exercise.
These are just a few ways that data can be used to help organizations work more intelligently as they strive to improve quality. Can you think of others?
¹ Duncan I. Dictionary of Disease Management Terminology. Washington, DC: Disease Management Association of America; 2004
² Lewis, G. Impactability Models: Identifying the Subgroup of High-Risk Patients Most Amenable to Hospital-Avoidance Programs. The Milbank Quarterly, June 2010.
This article was originally published on SCIO Health Analytics and is republished here with permission.