How Can Population Health Management Improve Chronic Kidney Disease Outcomes?

Gaps between medical guidelines for managing chronic kidney disease (CKD) and how it is managed in practice have persisted for a decade, from poor patient education and awareness to inadequate diagnostic evaluation and delayed nephrology referral by primary care physicians, who manage most patients, particularly in the early stages. The result is additional disease progression, hospitalizations and deaths.

An estimated 15% of American adults (33.5 million people) have CKD, which consumes about $137 billion in Medicare dollars, a third of which iss spent on the 5% of patients with end-stage kidney disease. Filling in the gaps would make a big difference for improving patient outcomes and reducing healthcare costs.

The University of Pittsburgh Medical Center (UPMC) is one of the few fully integrated health systems in the U.S. with its own insurance plan, which covers more than 4 million members, “thus providing an ideal environment for innovative models of care,” a team of researchers, all at Pitt, writes in a detailed perspective article published online March 17, 2025, in the American Journal of Kidney Diseases.

The not-for-profit University of Pittsburgh Health System has a network of 40 tertiary, specialty and community hospitals and 800 doctors' offices and outpatient sites in Pennsylvania, New York and Maryland.

Manisha Jhamb, M.D., M.P.H.

“The diverse patient population, heterogenous provider settings,” and mature electronic health record with a strong clinical informatics infrastructure “has allowed us to develop and test” multiple population health management strategies for CKD, writes first and corresponding author Manisha Jhamb, M.D., M.P.H., and colleagues. Jhamb is an associate professor of medicine and director of the Center for Population Health Management at the University of Pittsburgh School of Medicine and a practicing nephrologist at UPMC Kidney Clinic.

They go on to describe several strategies:

Building a strong PHM infrastructure to identify care gaps. Based on similar initiatives elsewhere, they created a CKD registry within the EHR that includes all outpatients (currently 157,000) with a diagnosis and/or two specific most recent eGFR measures at least 90 days apart. It identifies patients, phenotypically characterizes them, stores and calculates certain information and updates automatically. Provider-facing dashboards for primary care physicians and nephrologists facilitate monitoring and run in real time to track guideline-based metrics. Actionable links from the dashboard help address gaps.

Leveraging EHR to facilitate CKD recognition and management. After two tools failed to improve certain outcomes – clinical use was limited, perhaps due to “alert fatigue” and physician burnout – they developed what is referred to as a “novel PHM program for co-management with primary care” using some of the same tools, called K-CHAMP. The provider-centric intervention included a remotely delivered, multifaceted bundle with timely nephrology guidance and evidence-based recommendations in several areas. A pragmatic randomized controlled trial with 1,596 patients at high risk of progression who were not seeing a nephrologist showed improved quality of care on some measures but did not reduce disease progression, perhaps due to pandemic timing. Areas for future enhancements were identified.

UPMC CKD PHM program — translating research into clinical practice — learning health system approach. What is planned as a systemwide education initiative to improve CKD outcomes, care quality and costs in UPMC’s integrated health system was launched with funding from the UPMC Health Insurance Plan. The Kidney-Care (K-Care) program was built on K-CHAMP and uses a learning health system. It includes additional primary care provider education on new guidelines, individualized CKD education for patients, including a renal dietitian, a social worker to address social needs, and referral to renal palliative care clinicians as needed. Systematic screening and EHR documentation of social needs (medication assistance, financial assistance, transportation needs and food insecurity) is integrated during each education encounter, with a trigger for renal social worker referral as needed. The program has provided thousands of e-consults, medication reviews and education sessions over three years, but its effectiveness has not yet been evaluated.

PHM strategy for kidney palliative care and dialysis decision-making. Patients with advanced CKD have high palliative care needs that often are not identified or addressed, in part due to a shortage of palliative specialists, with the burden falling on primary care physicians or nephrologists who have inadequate training and limited time. Patients report high numbers of symptoms and inadequate support choosing end-stage kidney disease treatment options. UPMC developed a systematic approach within the K-Care program of identifying and initiating referrals to renal palliative care clinicians. Patients were triaged to a treatment decision-making education session based on age (over 80), CKD stage 4-5 and a “surprise question” asked of physicians: “Would you be surprised if the patient died in the next 12 months?” Clinical and cost outcomes have not yet been evaluated.

PHM implementation strategy for kidney transplantation. With transplantation the treatment of choice for patients with end-stage kidney disease, UPMC is developing a PHM-based approach to automate transplant referrals in the CKD clinic for all patients under age 80 or with a specific eGFR measure. The goal of automated referrals is to ensure that all patients are given equal opportunity and timely access to mitigate disparities and delays in kidney transplant referrals.

PHM implementation strategy for optimal transition to kidney replacement therapy. This means receiving a preemptive kidney transplant, initiating home dialysis or outpatient center dialysis. Optimal start is linked with better patient outcomes and lower health-care costs. UPMC plans to use EHR PHM to identify all eligible patients in the CKD clinic and use a multidisciplinary team model to ensure informed patient discussion of treatment options, timely referral for vascular or peritoneal dialysis access and transplant, care coordination, and identification and management of social risk factors.

“PHM can be an impactful strategy to improve value-based kidney care, but there are several financial and operational challenges to deploying this model of care in a fee-for-service environment,” write Jhamb and her colleagues. There needs to be strong infrastructure to support extensive EHR builds, data analytics and clinical services, and a strong commitment and upfront investment from health system leadership, a huge challenge, they note, with many health systems struggling financially. Moreover, the value-based care focus of PHM is unlikely to succeed without policy changes that incentivize and promote value-based care for upstream CKD care, with buy-in from payers, physicians and health systems. But combining individualized plans of care with data-driven enhancements to PHM, incorporating enhanced predictive modelling of risk with additional data, identifying and addressing social determinants of health as part of clinical pathways, and prediction modelling of therapeutic response together have significant potential, they write.