Chronic maintenance dialysis is an expensive procedure and Centers for Medicare and Medicaid Services extends coverage to all patients with ESRD who require dialysis or transplantation. When this was implemented in 1972, only about 10,000 patients were receiving dialysis, a number that has increased to over 469,000 patients with a cost of 30.9 billion dollars in 2013 [5]. This accounts for 7.1% of the overall Medicare paid claims cost for less than 1% of the total Medicare population [1].

The exact cost per patient per year depends upon the modality used, with HD being the most expensive at approximately $85,000 per patient per year (PPPY), followed by PD, which costs approximately $70,000 PPPY. Transplant is the most cost effective therapy with an expenditure of approximately $30,000 PPPY [1]. It is significant to point out that the difference in the expenditure of HD and PD in the US is not driven by a lower reimbursement to the dialysis units [6]. The amount paid to the dialysis units is the same for HD and PD but the higher cost for the HD population is mainly attributed to the cost of inpatient care and medication use [7]. Based on these numbers, one can deduce that PD is a financially attractive option for the ever-increasing population of ESRD patients in the US.

Although PD has been used as an RRT modality since 1976, the rapid growth of the ESRD population in the early 2000s was mostly due to patients undergoing ICHD [8]. Financial incentives for ICHD and concerns regarding the outcomes on PD were among the major reasons for this disproportionate increase in ICHD and as of 2008, less than 7% of the prevalent ESRD population was on PD [9]. The bundling of dialysis-related services led to a renewed interest in PD nationally with a 50% increase in the prevalent PD population from 2008 to 2013 (45,000 patients were on PD in 2013 compared to 30,000 in 2008) [1].

Despite this increase, the rate of PD utilization in the US is much lower than other countries like Hong Kong, Australia, New Zealand and Canada [10]. This difference cannot be solely attributed to variance in patient characteristics but is rather a result of obstacles impeding the growth of PD in our health care system. Lack of informed decision-making in ESRD patients is the biggest barrier. A quarter of the patients starting dialysis receive no pre-ESRD care but even more worrying is the fact that two-thirds of the patients are not even offered PD as an option despite the fact that 87% of patients would be eligible for it [1, 11, 12]. It is extremely concerning that these numbers challenge the basic principles of autonomy and patient-centered care.

Another important issue is the lack of familiarization with PD in providers since most nephrology training programs focus on HD [13]. Based on the results of a recent survey, 88% of nephrology training program directors felt that PD fellow training was limited and 60% endorsed personal inadequate PD training [14]. As physicians, our practice is limited to what we are most comfortable with. Therefore, these training limitations translate into lower use of PD by providers during independent practice.

In addition to provider related aspects, the most important factor in determining dialysis modality selection is patient choice. Despite being presented with the option of PD, a substantial number of patients choose to undergo HD. Patients report the fear of something catastrophic happening at home without health care provider supervision, lack of space at home and feeling of social isolation as main barriers to selecting PD [15].

In summary, both provider and patient related issues limit the use of PD in the US, which have to be addressed at a national level. Several initiatives like the Home Dialysis University for fellows are addressing the deficiency in provider training. However, most importantly as a team of health care providers, we should emphasize enhancement in patient education and patient empowerment, allowing them to make a decision that suits their lifestyle.

All ESRD patients should be assessed for PD eligibility. There are very few absolute contraindications to PD, which include lack of residence permitting PD, morbid obesity, large unrepaired abdominal wall hernias, expanding abdominal aortic aneurysm and active diverticulitis [16]. Most other factors like impaired vision, hearing, lack of dexterity to make PD connections, immobility and dementia are barriers, and these can potentially be overcome if a patient has assistance at home [16] (Table 1.1).

After evaluation of PD suitability, patients should then be offered a free choice as a part of modality education with written material, websites, videos, group lectures and one-to-one sessions on an as needed basis. The key here is to let the patients decide, as they are more likely to do better with the modality that they are interested in. Most studies show that half of the patients would choose PD if given the right [17].

Historically, the studies comparing outcomes of PD and HD have focused on mortality and yielded controversial results. An ideal comparison would have been a randomized controlled trial, which has been attempted in the past with very low recruitment rates. Earlier epidemiologic studies based on US renal data system (USRDS) registry showed that PD was associated with a 19% increase in mortality [18]. This became the cornerstone of the argument that PD is somehow an inferior therapy compared to HD. However, there has been a significant improvement in outcomes of PD since then as shown in a study by Mehrotra et al., where the composite outcome of mortality and change in modality over an 8 year period (between 1996 and 2003) showed a 17% improvement in PD outcomes as opposed to HD outcomes where there was no significant improvement [19]. More recent registry data from the USRDS and Denmark shows that there is no significant mortality difference based on the modality for RRT [20, 21]. In the US, the 5-year survival for patients starting RRT between 2002 and 2004 was 33% for PD compared to 35% for HD with no statistical difference.

Residual renal function (RRF) in dialysis patients contributes to small and middle molecular clearance and has effects on mortality with every 0.5 ml/min increase in glomerular filtration rate (GFR) being associated with a 9% lower risk of mortality [22]. HD is associated with a much faster rate of RRF decline (3.7 ml/min compared to 1.4 ml/min for PD at 12 months) which might be related to rapid changes in fluid homeostasis [23]. Moreover, selecting PD as an RRT method prior to transplantation has shown some positive effects on graft function [24].

However, patient outcomes are not only about biomedical outcomes but psychological outcomes are equally important-more so in some cases. PD is associated with more patient satisfaction. Patients receiving PD were much more likely than patients on HD to rate their dialysis care as excellent (86% vs 56% respectively) and including excellent ratings for each specific aspect of clinical care [25]. PD also allows greater flexibility in terms of travel and employment.

To summarize, PD and HD have similar medical outcomes but PD allows more flexibility. Ultimately the choice of RRT modality should be made by patients based on which modality is better suited to their lifestyle as the emphasis for patients is mostly on how they live-rather than how long [26].