Home Dialysis in the United States: A Roadmap for Increasing PD Utilization

Expanding treatment modalities available to patients with kidney failure is now a priority in the United States. On July 10, 2019, the president signed an Executive Order launching the Advancing American Kidney Health initiative, establishing a goal of 80% of end-stage renal disease (ESRD) patients treated with either home dialysis or a kidney transplant by 2025. Following this, the Centers for Medicare and Medicaid Services (CMS) proposed the ESRD Treatment Choices (ETC) model and described several voluntary models to incentivize kidney health, home dialysis and kidney transplantation (1). With transplant currently accounting for ~30% and home hemodialysis (HHD) 2% of patients, peritoneal dialysis (PD) utilization will need to increase markedly from its current share of ~10% to begin to approach this lofty goal (2).

Many reasons exist to promote wider uptake of PD, emulating many nations with developed economies where home dialysis utilization is 2- to 3-fold higher than the US (2). Patient experience and satisfaction with care appears either equivalent or better with PD versus in-center hemodialysis. In the US, nearly 80% of hemodialysis patients initiate with a central venous catheter (2), resulting in poor outcomes that could be abrogated with PD. Comorbid conditions like heart failure and pulmonary hypertension may be better managed with PD than with HD (3), and mortality appears similar or better than that seen with in-center hemodialysis (4). However, PD utilization remains low (supplemental figure 1), despite the modest increase in PD utilization seen after financial incentives were provided by the ESRD Prospective Payment System that launched in 2011 (5, 6). To approach the remarkably ambitious goal in the executive order, patients and providers must overcome multiple social, clinical and financial barriers. Building on findings delineated at a recent National Kidney Foundation conference on home dialysis (7), this editorial highlights several barriers and suggests potential solutions (Box 1).

Box 1. Solutions to Increase PD Utilization in the United States.

• Education and Social Factors

  1. Broaden education to patients and their carepartners about home dialysis

     1. Decision-aids and health literacy appropriate educational tools
     2. Allow separate billing for options discussions in the inpatient setting in addition to a daily nephrology visit
  2. Improve education and comfort with PD among nephrologists, both during fellowship and in practice

     1. Incorporate longitudinal home dialysis exposure in fellowship
     2. Provide resources and opportunities for training and retraining practicing nephrologists in home dialysis
  3. Improve education and comfort with PD among non-nephrology clinicians who commonly provide care for patients with kidney failure, including primary care providers, cardiologists, diabetologists, access surgeons and others

     1. Promote broad outreach to non-nephrology specialists
     2. Introduce PD into the popular culture to improve awareness, supplanting the default hemodialysis facility image
• Clinical Factors

  1. Dialysis facilities

     1. Incentivize growth of the home nephrologist nurse workforce
     2. Incentivize urgent start peritoneal dialysis programs for individuals who ‘crash’ onto dialysis
  2. Non-dialysis facilities

     1. Allow reimbursement for assisted PD at home and for multiple training visits
     2. Promote PD in rehabilitation centers, nursing homes, and skilled nursing facilities to bridge the gap between inpatient and outpatient PD
     3. Reduce regulatory burden for skilled nursing facilities who treat PD patients
     4. Provide resources for respite PD care for patients who rely upon carepartners
  3. Home environment

     1. Provide resources for assisted PD
     2. Increase telehealth resources as currently permitted by regulation
     3. Reduce costs associated with more frequent supply delivery for patients with limited storage space
  4. Increase research into PD, including different PD solutions to improve peritoneal dialysis, technologies for identifying causes of catheter malfunction, and advances to improve catheter function
• Financial Factors*

  1. Modify the existing payment structure to incentivize PD by equalizing physician payments for in-center and home care supervision
  2. Increase reimbursement for home dialysis, incorporating reduced transportation costs and other potential societal savings associated with home dialysis
  3. Revisit the QIP such that facilities providing PD are not more likely to receive financial penalties
  4. Incentivize surgeons for successful PD catheter placement
  5. Ensure adequate availability of fairly priced home dialysis supplies

*Financial factors refer to those potentially modified by US Federal policy

Social and Education Factors

To select their optimal KRT modality, patients and their carepartners must be knowledgeable about treatment options. Patients who receive education prior to dialysis initiation are more likely to choose home dialysis (8) and perceive this education as helping provide them greater control over their lives(9, 10); despite this, as few as one-third of incident dialysis patients are presented with all treatment options available to them (11). Low PD utilization may reflect patients’ lack of early awareness of CKD, failure of timely referral to nephrology, and skepticism (or worse, ignorance) from their nephrologist or other providers about the risks and benefits of home dialysis (12). Regardless, comprehensive education followed by a patient-centric discussion of treatment modalities should occur. Even in cases of “urgent starts,” where dialysis must be initiated within 48 hours of patient presentation, PD is a possibility when the right resources exist and, certainly, an early transition to PD is possible.

Among populations with greater socioeconomic disadvantages, kidney failure remains an outsized burden, and this disparity is further seen in the relative rates of home dialysis uptake (13). Factors such as poverty, housing instability, carepartner limitations, lack of storage space, and low health literacy present real challenges to home dialysis usage (14), and there may also be provider biases around expectations in learning PD. Some of these factors may be overcome by proactively providing assistance, especially in the early period after PD initiation. Supporting caregivers and patients with home visits and assisted PD can safely increase PD utilization (15), while incorporating remote monitoring applications may help maintain PD.

Take home: More patients may be able to achieve home dialysis, particularly with improved education and assistance programs. Education and exposure of medical professionals, including nephrologists, to a high number of diverse PD patients is critical to overcome preconceptions and improve comfort with PD.

Clinical Factors

The only absolute clinical barrier to PD is the lack of a functioning peritoneal membrane. Despite this, our ability to predict long-term PD utilization is limited. In the US, 17% of incident PD patients transfer to hemodialysis within a year, and fewer than half of patients who start PD are using this modality two years later (16). On a population level, precise causes for transfer to HD remain uncertain and likely heterogeneous, but the high transfer rate may incorporate a reflexive transition to hemodialysis when problems with PD arise. Further work to determine the causes of transfer to HD will be helpful to both reduce the need for transition through earlier interventions and ensure a smooth transition from PD to hemodialysis should it occur.

Currently, the average patient treated with PD is healthier than the average patient treated with hemodialysis. As PD expands, the PD population will include more individuals with greater comorbidity burden. These individuals will have acute illnesses requiring acute-care and rehabilitation hospitalizations as well as chronic conditions limiting physical and cognitive function. The lack of clinical expertise in many hospitals and the paucity of rehabilitation facilities, long-term acute care hospitals and nursing homes with the willingness or ability to care for PD patients limits PD as an option for many patients. Additionally, patients with greater comorbidity are more reliant on family support, potentially resulting in greater need for respite and assisted-PD options.

Initiating PD is far more time intensive than hemodialysis. PD initiation requires more extensive discussions with patients and their families, identifying and interfacing with other providers who are comfortable with PD including access surgeons, primary care providers, cardiologists and others, and empowering patients and their families to take a leadership role in their own care. Recognizing this, modality education must begin earlier in the disease course and be facilitated by timelier referral to nephrologists and better decision aids for patients.

Take home: Transfer from PD to hemodialysis is common. Early troubleshooting, maximizing patient flexibility to avoid burnout, and interfacing with other providers, including primary care providers, subspecialists and non-home medical facilities, to improve awareness of PD, knowledge of potential benefits of PD, and availability of PD, should improve PD utilization.

Financial Factors

Multiple incentives currently predispose to greater hemodialysis than PD utilization (17, 18). For nephrologists, assuming four encounters with hemodialysis patients per month, the monthly capitated payment is significantly less for PD than for hemodialysis, even though clinical encounters with PD patients may be more meaningful and time consuming. The PD training fee may account for some of this differential, although this is not always claimed and is a one-time distribution. Despite gaining Medicare coverage in the month of PD initiation, insufficient insurance prior to dialysis initiation is associated with lower PD uptake, likely reflecting the default to HD in the absence of sufficient patient education on home dialysis options (19). Expanding kidney failure education to these individuals, particularly if they will qualify for Medicare at the time of PD initiation, is an attractive policy albeit one with legislative barriers.

The major disincentive for nephrologists occurs prior to dialysis initiation, where the time required to provide meaningful education to patients and their families regarding PD while concurrently managing their comorbid conditions may result in unreimbursed effort. Currently, Medicare provides a Kidney Disease Education benefit only for patients with CKD stage 4, but this benefit is underutilized (1), with reimbursement lagging that of a typical evaluation and management (E/M) clinic visit. The proposed ETC model expands the Kidney Disease Education benefit to include CKD stage 5 and 5D for six months following dialysis initiation, with additional clinical staff, including dieticians and social workers, able to provide education.

For dialysis providers, PD is a mixed proposition. Typically, per-patient provider costs to manage a PD patient are less than those for in-center hemodialysis, although increases in PD solution and equipment costs could reduce this difference. The major conflict involves the opportunity cost of PD, which is a potential unfilled chair at a hemodialysis facility. Filling an existing empty hemodialysis chair will result in greater financial benefits than treating a patient with PD, providing an incentive to initiate and maintain patients in hemodialysis facilities. Additionally, facilities with PD programs are at higher risk of quality incentive program (QIP) penalties (20). The dialysis adequacy metric currently incorporates both PD and hemodialysis adequacy into a single measure. The HD adequacy metric is topped out, representing the ease of achieving solute clearance, with fewer than 10% of facilities having hemodialysis adequacy results below the achievement threshold; in contrast, a third of facilities that offer PD have PD adequacy results below the achievement threshold (20). Additionally, there is no patient-experience measure in PD, magnifying the financial importance of other measures for facilities that provide PD.

Take home: Multiple financial incentives favor hemodialysis over peritoneal dialysis, despite potentially lower costs associated with PD. Increasing PD requires realigning incentives, including: 1) Reimbursing more for PD; 2) Correcting QIP disincentives for PD; 3) Controlling costs of PD supplies to maintain financial incentives for PD; 4) Establishing PD centers of excellence beyond dialysis facilities including hospitals, skilled nursing facilities, and rehabilitation centers that are able to provide high quality care when a PD patient is not at home; and 5) Providing sufficient reimbursement for assisted PD.

Conclusions

Although in-center hemodialysis is the current default kidney replacement therapy in the US, the Advancing American Kidney Health Initiative represents an immediate opportunity to increase PD utilization in the US and improve kidney care. While there is no consensus on the number of US dialysis patients who should utilize PD, the current 10% PD prevalence is far too low. Increasing the utilization of PD requires broad changes in knowledge, infrastructure and incentives. Patients, family members, nephrologists and non-nephrology providers must be aware of PD, recognize advantages associated with PD, view potential barriers as surmountable, and support the uptake and, critically, maintenance of PD. Financial incentives must be appropriately aligned to promote PD uptake and maintenance, with support for assisted PD and PD in settings outside the home.