Partnership Aims to Bolster Kidney Care with Predictive Analytics

Trinity Health and Strive Health announced plans to enhance treatment for chronic kidney disease (CKD) and end-stage kidney disease (ESKD) for Loyola Physician Partners patients using predictive analytics and machine learning.  

One of the most prominent nonprofit Catholic healthcare systems in the country, Trinity Health includes 123,000 employees and almost 27,000 physicians and clinicians. Strive Health partners with payers and providers to support value-based kidney care using a combination of care teams, predictive analytics, advanced technology, and integration with local providers.

Loyola Physician Partners is a clinically integrated regional network. Formed by Loyola Medicine, a part of Trinity Health, it includes more than 2,000 employed and independent primary care physicians and specialists.

In this collaboration, Strive Health will work with Loyola Physician Partners to pilot a clinical care model that aims to help slow kidney disease progression and preserve patients' kidney function.

According to the Centers for Disease Control and Prevention (CDC), 37 million adults in the US are battling kidney disease, of whom over 38 percent of people are older than 65.

Strive Health and Loyola Physician Partners will pilot the model through specialized population health programs to treat CKD and ESKD patients in the Chicago area.

The programs will leverage Strive Health technology and services. The technology uses machine learning and predictive analytics capabilities to provide a comprehensive view of the patient experience. Typically, kidney care focuses on the latter stages of kidney disease, at which point the only options are dialysis or transplant.

Strive's care team, which includes dietitians, pharmacists, and licensed clinical social workers, uses the technology to assess a patient’s risk of hospitalization or the progression of their disease in an effort to reduce hospital stays and readmissions.

“Loyola Physician Partners is proud to add specialized kidney care services to our population health programming, both to improve outcomes for our patients, as well as providing additional support to our providers,” said Keith Veselik, MD, vice president and chief medical officer of population health at Loyola Medicine, in the press release.

Similarly, a partnership that began in January aims to optimize value-based care delivery for kidney disease patients.

Through the new collaboration, Evergreen Nephrology will use an Innovaccer solution suite to enhance kidney care through analytics, care management, physician and patient engagement, and referral management.

In addition, research has shown that predictive risk assessment can improve care in various areas, including kidney care. But hospital-specific data is needed.

A study from June 2022 described how clinical risk prediction models performed well when they were used in live clinical workflows within hospitals for sepsis, delirium, and acute kidney injury.

The study shows that despite the common prior use of clinical risk prediction tools, most are developed using retrospective data. As researchers compared the performances of risk prediction models that used live clinical workflow data and those that used retrospective data, they discovered that though the overall performance was similar, the cross-hospital evaluation showed severely reduced prediction performance.

Thus, they concluded that model calibration using data from the deployment hospital is necessary for prediction model performance.