Building on innovations to inform kidney disease policy and practice

Over the last several decades, researchers affiliated with the University of Michigan (U-M) have developed expertise across a range of disciplines with the aim of bettering the health and quality of life of those living with kidney disease, while improving their access to care, the safety and effectiveness of clinical practice, and reducing costs associated with this illness. Their work has contributed to evidence-based practice and policy changes in the U.S. and globally that have likely helped save countless lives and hundreds of millions of tax-payer dollars. Their efforts are poised to continue to contribute toward stemming the challenges of this persistent public health issue in the future.

The U-M Institute for Healthcare Policy & Innovation (IHPI)’s expertise in kidney disease health services research and policy encompasses the work of dozens of faculty and staff across several partner units and organizations that are dedicated to these fields. Major collaborators, whose work is described in this brief, include the University of Michigan Kidney Epidemiology and Cost Center (U-M KECC), formed in 1993, and Arbor Research Collaborative for Health, founded in 1997.

To appreciate why research on kidney disease is so important, it’s helpful to understand the crucial functions of the kidneys, as well as what happens when these powerhouse organs no longer work the way they’re supposed to. The primary job of the kidneys is to maintain both the volume and chemical composition of the body’s fluids, especially with respect to electrolytes such as sodium and potassium, remove wastes and fluid by filtering the blood, and produce urine. The kidneys also produce hormones that help regulate blood pressure, create red blood cells, and keep bones strong. When kidneys fail to perform at their full capacity, the results can be devastating.

Chronic kidney disease (CKD), which is the gradual loss of kidney function over time, can go undetected for years or even decades, since people may not experience symptoms until they reach the point where their kidneys have nearly ceased working. Although about 14 percent of people in the U.S. are living with CKD, only a small fraction – about 1 in 10 of them – is aware of it.

As CKD progresses into its most severe phase, known as end stage renal disease (ESRD), or permanent kidney failure, the kidneys can no longer support the body’s needs, and a transplant or dialysis treatment (using a machine that filters wastes from the blood) is required to replace the work the kidneys would normally perform. 

Common risk factors for CKD include older age, diabetes, high blood pressure, heart disease, obesity and acute kidney injury. Minorities tend to be affected more. CKD frequently overlaps with other chronic conditions, most commonly diabetes, hypertension and heart disease. Kidney disease often has the effect of intensifying the health burden of other co-existing chronic conditions (co-morbidities), such that most people with CKD will die prematurely of other, often related illnesses – most commonly cardiovascular disease – well before ESRD has a chance to develop.

Increasing awareness about CKD and efforts to reduce its risk factors are essential to prevent new cases, decrease its rate of progression, and mitigate symptoms.

Yet another reality is that millions worldwide live with ESRD, the irreversible end-stage form of the illness, some waiting for an organ transplant, and often dependent upon dialysis, a life-sustaining yet also life-consuming treatment in terms of its effect on quality of life for patients and their families. How can the quality of dialysis care be improved for these individuals? And what about reducing the resource use and expenses that the treatment of all stages of kidney disease – estimated at more than $100 billion for the U.S., taking Medicare, Veterans Administration (VA), and private healthcare costs into account – poses to our healthcare system?

Without a comprehensive understanding of the numbers of people living with kidney disease, how they are faring under various treatments, as well as the cost of their care, it’s impossible to monitor the effectiveness of prevention and treatment efforts, to make sense of regional variations in care and outcomes, or strategize the use of healthcare resources in dealing with this illness.

The United States Renal Data System (USRDS) is the largest national registry in the world for monitoring kidney disease in general, and end-stage disease in particular. In 2014, Rajiv Saran, MD, M.R.C.P., M.S., an expert in renal epidemiology and an associate director of U-M KECC, in partnership with Arbor Research and several U-M colleagues, was awarded the USRDS coordinating center contract by the NIH, thus helping to bring back the USRDS to U-M after a gap of 14 years.

Based on prior experience in analyzing ESRD data from the Centers for Medicare & Medicaid Services (CMS) and the Michigan Kidney Registry, Philip J. Held, Ph.D., and Fritz Port, M.D., M.S., were jointly awarded the first USRDS contract from the National Institutes of Health (NIH) in 1988, and U-M served as the coordinating center for the USRDS until 1999.

One of the USRDS’s major products is a web-based Annual Data Report that serves as an authentic source of information on kidney disease by researchers around the world. The registry collects, analyzes, and distributes information related to incidence, prevalence, treatment, costs, and outcomes of kidney disease in the U.S. The USRDS has especially thorough data on the ESRD population because these individuals are required to register for benefits through Medicare, which has provided coverage to all people with permanent kidney failure, regardless of age, since 1972.

U-M KECC investigators, in collaboration with University of California, San Francisco, and Arbor Research, have also established the first-of-its-kind CKD surveillance system for the United States (Saran is U-M’s Principal Investigator). This project is funded by the Centers for Disease Control and Prevention (CDC) and systematically tracks and reports information on risk factors and disease burden, disease awareness, quality and processes of care, CKD-associated health consequences, and health system capacity available to deal with CKD.

Saran, who is also a professor of internal medicine and epidemiology, and U-M colleagues have also recently laid the foundation for a novel, kidney disease data/information system for the Veterans Administration (VA) called the VA Renal Information System (VA-REINS). The VA will use this system to monitor the disease in all stages among the millions of veterans it cares for, with the goals of improving early identification and access to care, optimizing disease management at all stages of CKD, reducing care-related costs, and monitoring prevention efforts and health outcomes.

As a group, U.S. veterans bear a much higher burden of chronic kidney disease and have more risk factors for it than the rest of the country’s population. Over 52,000 veterans developed ESRD over the period 2007-2011, and the cost to the VA for care of patients who have CKD has recently been estimated in recent years to be more than $16 billion per year (preliminary, unpublished results).

Additionally, with the burden of kidney disease much greater in other parts of the world, U-M KECC is assisting other countries such as China and India in developing surveillance systems to monitor risk factors and early-stage CKD in their populations.

The most recent USRDS data indicate that more than 600,000 people across the U.S. are being treated for ESRD, roughly two-thirds by dialysis. While transplantation is clearly the preferred therapy when indicated, the widespread shortage of transplantable organs makes it available to far fewer than the number who could benefit from it; the active waiting list for kidney transplants is 2.7 times larger than the annual supply of donor kidneys, and many waitlisted patients die or become too ill to benefit from a transplant.

There are some encouraging trends: after a sustained rise in new ESRD cases for nearly three decades, this rate appears to be levelling off in recent years, according to the 2015 USRDS Annual Data Report. While mortality rates from kidney disease remain high (approaching 10 percent per year among patients with advanced CKD, and nearly 20 percent among patients on dialysis, with a five-year survival rate of only 40 percent), these rates have also been declining, dropping by 28 percent and 40 percent among dialysis and kidney transplant patients, respectively, since 1996.

Together, these statistics mean that more people will be living with end-stage kidney disease; the number of people on dialysis increased 4 percent in 2013, and is 63.2 percent larger than in 2000. Meanwhile, researchers and clinicians focus on continuing to extend the lifespan for these individuals while improving the quality of their lives as they receive therapy. 

While dialysis offers a life-sustaining intervention, its effect on quality of life for individuals and their families can be dramatic. Most dialysis patients require three in-center, 3.5-to-4-hour-long hemodialysis treatment sessions per week, and may also need the second half of those days to recover from each session, representing a major loss of freedom and autonomy. Focusing on patient-centered outcomes and treatment preferences around dialysis is a major priority for U-M and Arbor Research’s researchers.

A patient-centered project spearheaded by Francesca Tentori, M.D., M.S., senior research scientist at Arbor Research, seeks to identify the factors most important to patients who are transitioning to dialysis, and to develop a decision aid tool to help them identify the treatment options that would best fit their preferences.

Patient safety remains a major concern. In-center hemodialysis typically involves the removal of a high volume of fluid in a relatively short time frame, which can dramatically affect blood pressure and cardiovascular stability. This can lead to potential organ damage and severe complications, even death, especially if dialysis sessions become unstable following rapid fluid removal.

With the aim of reducing the instability of dialysis sessions and improving patient satisfaction and quality of life, Saran and co-principal investigator Tiffany Veinot, Ph.D., an associate professor of information and public health, will be developing educational interventions that will involve patients and providers in creating a culture of safety, by focusing on dialysis facility-wide policy changes related to improving health literacy, behavior modification, and ensuring stability of the dialysis procedure.  

When any of us visits a physician, the decisions made around the care we receive are influenced by a combination of factors: clinical guidelines, best practices, the provider’s own clinical judgment, and our own preferences. Care in practice, as a result, often differs markedly between providers. This is also true for the treatment provided at dialysis facilities, and discretionary practices can vary greatly between units, as each develops its own ‘culture’ around a particular way of doing things.

Since 1996, the Dialysis Outcomes and Practice Patterns Study (DOPPS) has collected observational data on patient outcomes and practice patterns among dialysis facilities around the world, with the goal of identifying optimal treatment practices that extend survival and improve health-related quality of life of patients with kidney failure. The DOPPS is led at Arbor Research by Principal Investigator Bruce Robinson, M.D., M.S., FACP, with contributions from multiple Arbor Research and U-M collaborators.

The motivations driving the DOPPS, borne out through the study’s findings over two decades, is that measurable differences in dialysis facility practices influence patient longevity, morbidity, and health-related quality of life, and that disseminating findings on best practices among healthcare providers, researchers, and policymakers can have a meaningful impact on policy and patient outcomes.

By shedding light on the effects of dialysis practices on a wide range of patient outcomes, the DOPPS has uncovered many other discretionary practices among facilities that are likely modifiable for the betterment of patient care, and has also helped identify and promote optimal practices in areas where no clear clinical guidelines exist. These have included findings relevant to:

• Dialysis treatment session length: evidence shows that slowing the rate of fluid removal through longer and gentler dialysis sessions results in improved long-term outcomes and patient safety, but trends in the U.S. are towards substantially shorter treatment times than elsewhere (by contrast, both Germany and Japan have implemented payment policy changes to favor treatment time of 4+ hours, based in part on international DOPPS data).
• Timing of dialysis initiation: some patients are starting dialysis earlier than necessary, and before they are fully prepared for the treatment, which has tremendous implications not only for health outcomes and costs, but also for a patient’s quality of life; DOPPS data indicate much better preparedness for dialysis, and better patient outcomes, in some countries than others.
• Catheter use: the United Kingdom now imposes a tariff on the use of chronic central venous catheters for dialysis, based in large part on international DOPPS data demonstrating past substandard performance measures for the country in this area; performance in the UK is now much better.

The DOPPS family of studies has now expanded to more than 20 countries, and has tracked over 70,000 patients worldwide, with international comparisons continuing to offer some of the more significant opportunities to identify potential practice improvements and influence on policy. The DOPPS Practice Monitor identifies and publicizes trends in dialysis practice, providing feedback to physicians, researchers, regulators, and industry.

Hemodialysis requires access to a patient’s bloodstream, and there are a number of ways to connect a patient’s blood supply to the dialysis machine that filters the blood during the procedure. The most effective means of engineering this vascular access, often referred to as the “lifeline” for patients on dialysis, is usually through an “AV fistula,” which involves forming a new access point by surgically connecting a patient’s artery and vein together. Compared to other types of access, AV fistulas are more durable and usually less expensive to maintain. They also tend to have the fewest problems with infection, blood flow, and other important outcomes, including lower risks of death and hospitalization.

In the U.S., however, convenience and reimbursement rates had for decades skewed preferences toward other types of access, such as catheters and artificial grafts. A national study of patients starting hemodialysis published in the Journal of the American Medical Association in 1996 by U-M’s Richard Hirth, Ph.D., M.A., a healthcare economist and an associate director of U-M KECC, and colleagues was essential in documenting large variations in the relative use of fistulas and grafts, and a trend away from fistulas. By comparing data between the U.S. and Japan and several European countries in the late 1990s and early 2000s, DOPPS researchers found that fistula use was tied to far better outcomes among patients, and that differences in survival among dialysis patients between countries were largely attributable to differences in fistula use among providers in these different parts of the world.

Together, these findings, generated by researchers at Arbor Research and U-M KECC working on the DOPPS study, including Drs. Port, Saran, and Ron Pisoni, PhD., M.S., spurred the CMS in 2003 to launch a major continuous quality improvement initiative (the “Fistula First Breakthrough Initiative”) to promote the use of fistulas for vascular access in the U.S. Since then, fistula use has risen substantially among U.S. patients, and dialysis survival rates have meaningfully improved in large part because of this development, with mortality among this population falling by 23 percent over the last decade.

Still, there is still much room for improvement, as the U.S. lags behind many other countries when it comes to fistula use, particularly around the transition period when dialysis is initiated. Odds are, if someone begins dialysis with a fistula, they’re likely to continue with it. “Upstream care” for kidney disease, meaning those whose disease has not reached end-stage, is also sub-optimal, with nearly 40 percent of the patients who reach dialysis never having seen a nephrologist before, representing a major practice gap to be overcome. DOPPS researchers, led by Robinson, and USRDS and National CKD Surveillance teams’ researchers, led by Saran, continue to monitor these trends and seek innovative ways to improve patient outcomes through better preparation for dialysis.  

In 2013, total Medicare expenditures for all stages of kidney disease was $81 billion, more than $30 billion of which was spent on those with end stage renal disease, according to USRDS estimates. ESRD patients make up less than one percent of all Medicare recipients, but account for seven percent of all Medicare costs.

ESRD is one of only a few diagnoses that qualifies individuals of any age for Medicare, which is typically reserved for people age 65 and older, and as a result the treatments for most ESRD patients are paid through Medicare. These high costs are often due to the underlying disease complications and multiple co-morbidities that plague those with kidney disease, which can lead to high rates of hospital admission and readmissions.  

For years, Medicare paid for injectable dialysis medications on a fee-for-service basis, which encouraged overuse (in the form of more expensive drugs as well as inappropriately high doses) and discouraged home dialysis. Congress and the CMS sought to reform dialysis payment by developing a bundled, prospective payment system (PPS) to incorporate the dialysis treatment and all related medications and laboratory tests.

Through a series of contracts with CMS beginning in 2005, U-M’s Richard Hirth, also a professor of health management and policy at the U-M School of Public Health, and colleagues at U-M KECC designed a new payment system that would fundamentally alter the incentive structure for the way that Medicare pays dialysis centers for treatment, by offering a flat rate for patients whether the therapy is done at home or in a facility. Since its implementation by Medicare in 2011, the PPS system for ESRD treatment has improved quality of care, efficiency, and patient choice, and contributed to significant cost savings for the government.

Initially, the system was designed to save Medicare around $200 million per year in payments from the change in practice patterns, and the reduced payment rate that Medicare is able to pay facilities because of that change. When PPS changes are fully phased in, annual savings will amount to an estimated $500-600 million per year, or about five to six percent of total dialysis costs.

Just as importantly, in monitoring key quality measures throughout the dialysis industry, such as mortality and hospitalization rates, evidence points to continued improved outcomes for dialysis patients since the implementation of the PPS.

Meanwhile, the practice of home dialysis, which uses far fewer injectable medications, has risen substantially in the last five years (and is 52 percent higher than a decade ago), particularly among patients just starting out with dialysis therapy. The flexibility that this treatment modality offers makes it a welcome alternative option for eligible patients, and represents a positive development for patient access and choice. The PPS has also prevented the overuse of certain services and potentially harmful medications, promoting more appropriate levels of care within systems of dialysis delivery.

To protect against any unintended consequences or a decline in the quality of care under the PPS, CMS developed the ESRD Quality Incentive Program (QIP), the agency’s first national value-based purchasing program, linking a portion of payment directly to dialysis facilities’ performance on quality of care measures (also known as “pay-for-performance”). Arbor Research has supported CMS since the program’s implementation in 2010, providing CMS with objective data analyses to inform policy decisions that achieve the goals of continuing to improve the quality and efficiency of dialysis care.

In addition, as the CMS contractor for ESRD quality measure support, U-M KECC is responsible for developing, revising, and maintaining various measures related to clinical quality of care, care coordination, population and community health, safety, person- and caregiver-centered experience and outcomes, and efficiency and cost reduction, many of which are available for patients and families to evaluate themselves in comparing dialysis facilities. Review of these measures helps ensure that dialysis care quality keeps pace with new clinical evidence or changes in guidelines, for example, and that quality responds to policy changes such as those implemented with the new PPS.    

The complex health needs of people with kidney disease often require them to visit multiple providers and follow multiple care plans, which can be challenging if care is not coordinated. Hirth and U-M KECC colleagues are also involved in a demonstration project funded by the Center for Medicare & Medicaid Innovation (CMMI) to look at whether the concept of accountable care, which has generally centered on primary care practices and physicians, could be applied to specialty care settings – in this case, dialysis facilities – to improve coordination of these patients’ care.

In accountable care generally, groups of healthcare providers bear responsibility for the costs and outcomes of their Medicare patients, with the goal of ensuring that patients get the right care at the right time, while preventing medical errors and unnecessary duplication of services. If they achieve savings while meeting quality targets, they get to share in those savings, but if targets are not reached, they share in the risk of the higher spending.

If this first-of-its-type project is successful, CMS could extend the model to other chronic conditions, which could have enormous advantages for Medicare beneficiaries and the efficiency of their systems of care, and big potential cost savings.

IHPI-affiliated researchers have made important headway in staunching the impact of kidney disease by monitoring trends in prevention and treatment, addressing quality of life, access to treatment, and patient choice, and helping design pioneering new models to optimize resource use and care coordination. For the significant challenges that remain in prevention and disease management, ongoing collaborations at U-M and among its research partners will continue to develop the evidence and novel approaches to policy and practice that can benefit all who are touched by this disease. 

April 2016