FEATURE · ANALYTICS/AI
At Spring Conference, NAACOS Quality Award Winners Share Keys to Innovation
By David Raths A
t the National Association of Accountable Care Organizations (NAACOS) spring conference this
week, three ACOs were recognized with Quality Excellence Awards for moving the needle on improving patient care: Delaware Valley ACO, University of California San Francisco, and Texas-based Southwestern Health Resources. Jason Fish, M.D., chief medical officer
with UT Southwestern ACO in the Dallas- Fort Worth area, spoke about working with a machine learning company called ClosedLoop. They are working to reduce unplanned hospital admissions among patients who receive complex care management by using a machine learning platform the ACO developed to help predict patients likely to incur a preventable event such as unplanned hospital admissions, emergency depart- ment high utilization, or death. “When you traditionally look at how people take care management strategies
toward population health, you usually do something called hot spots, right? You use historical data, claims-based data, and you say that's the population I want to focus on,” Fish explained. “The payers bring us lists of high utilizers, but these are yesterday's high utiliz- ers. The question is, what percentage of them will be tomorrow's high utilizers. When we actually ran this, we found it was only 30 percent. And yet here we were focusing 100 percent of our energy on yesterday's risk and not tomorrow's. That made us say that maybe this isn't the right approach to this. How could we incorporate yesterday's risk, but really try to identify tomorrow's risk and spend our energy there?” Working with ClosedLoop, UT
Southwestern ACO, also known as Southwestern Health Resources, devel- oped a risk stratification algorithm using machine learning techniques to match high-risk patients with complex care
8
hcinnovationgroup.com | JULY/AUGUST 2023
management services to prevent poten- tially avoidable hospitalizations and ED visits. Leveraging data from claims and EHRs, the ACO built predictive models and integrated and trained the models with data from publicly available social determinants of health data sets, such as the Social Vulnerability Index and the Area Deprivation Index, to identify socioeconomic barriers to care at the individual level. The ACO then encouraged patients to
enroll in a longitudinal complex care man- agement program. Using personalized care management plans, patients received tailored support and care coordination for clinical, social, pharmaceutical, and behavioral health needs. The high-inten- sity multidisciplinary care management program spans three to four months and then community health workers follow patients for an additional two months, checking in with patients at least every two weeks and escalating issues to the NAACOS continues on page 26
203859613 © Vaeenma |
Dreamstime.com
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36
