assessed catheter failure (removal of a dysfunctional PD catheter). 108 Factors associated with technique failure are summarized on Table 5; more detailed information can be found in Appendix C, Table 6. Increased BMI or categorization as obese was associated with higher rates of technique failure in 3 of 4 studies evaluating that factor.109,112,113 Increased systolic blood pressure (2 studies reporting)103,114 and catheter problems (2 of 3 studies reporting)108,112 were also associated with higher rates of technique failure. African-American race was associated with increased technique failure in 3 of 5 studies that reported results by race103,109,115; a sixth study observed increased technique failure in the indigenous population of Australia/New Zealand. 112 Findings were mixed for presence of diabetes, age, gender, PD type, and geographical distance to the clinic/nephrologist. A small US study found no difference in technique failure based on distance109 while a large Canadian study found lower technique failure with increased geographical distance from the nephrologist.111 The authors noted a slightly higher mortality risk among remote-living PD patients. One study reported higher technique failure in patients with cardiovascular disease and in patients with lower eGFR. 107 Patients from larger dialysis centers had lower rates of technique failure in 2 US studies.78,110 Need for assisted PD was associated with decreased technique failure in a large study from France104 but not in a smaller study from Ireland. 105 49 Home-based versus In-center Dialysis Evidence-based Synthesis Program The temporal study found a lower adjusted risk of technique failure among patients initiating PD between 2001 and 2005 compared to the 1995 to 2000 group (HR 0.89 [95% CI 0.82, 0.98]).116 There was no significant difference between the 2006 to 2009 group and the 1995 to 2000 group (HR 0.95 [95% CI 0.85, 1.06]). Among patients older than 65 years, there was a lower risk of technique failure for both of the more contemporary groups compared to the 1995 to 2000 group. Home Hemodialysis Selection Not all patient homes are suitable for HHD. An observational study from the UK reported on findings after visits to the homes of 249 patients who were medically suitable for HHD.117 Onethird of the homes did not meet the Decent Home Standards. Hazards to health/well-being included overcrowding (57%), dampness/mold growth (33%), inadequate facilities for sanitation and drainage (17%), risk of structural collapse (10%), inadequate domestic hygiene, pests, and refuse (8%), inadequate facilities for storing and preparing food (8%), and inadequate supply of uncontaminated water (3%). Due to spatial, health, and safety concerns, 30% of the homes were not suitable for either HD or PD. A Canadian study of 236 patients initiating HHD or PD looked at differences between HHD and PD patients.118 HHD patients tended to be male (70% vs 50%, P = .05), were younger (46 vs 62 years, P < .001), were less likely to have diabetes (24% vs 45%, P = .003), and had a longer delay between first renal replacement therapy and the start of HHD (4.8 years vs 0.34 years, P = .002). Another Canadian study compared HHD patients (n=15) to PD (n=79) and in-center HD (n=59) patients.119 HHD patients were younger, had a lower BMI, and were more likely working than either PD or in-center HD patients (all P < .05). HHD patients were more likely Englishspeaking than HD patients. There were no differences in eGFR or comorbidity index values at the start of dialysis. Patients reported not choosing HHD because of disinterest (25%), lack of social support (25%), inadequate space (5%), communication issues (5%), and inability to perform own dialysis (3%). A third study from Canada surveyed 66 nocturnal HHD patients and 199 HD patients with no contraindications or other factors limiting ability for HHD.120 The surveys were completed by 85% of the HHD patients and 77% of the HD patients. The nocturnal HHD patients were significantly younger, less likely to have diabetes, and had a higher physical quality of life (SF12). There was no difference in gender, the mental component of the SF-12, perceived ability for self-care, perceived social support, or anxiety. HHD patients were more likely to be comfortable with self-cannulation, believe they will receive as good care as with HD, believe they can properly perform nocturnal HHD, and be less fearful of a catastrophic event. 50 Home-based versus In-center Dialysis Evidence-based Synthesis Program Table 5. Overview of Factors Evaluated for Technique Failure (Switch from PD to In-center HD) Study (N) Country African American/ Race ↑ BMI or Obesity Diabetes ↑Systolic BP CVD ↓GFR ↑ Age Gender Catheter problems CAPD (vs APD) Other Shen 2013103 (1587) USA ↑ ↑ ↓ female ↑Disabled ↑On Medicare ↔Others Lobbedez 2012104 (9882) France ↔ ↓ ↑ ↓Assisted PD (vs selfcare) ↑HD before PD Smyth 2012105 (148) Ireland ↔ ↔ ↔Etiology of ESRD ↔Catheter method ↔Comorbidities ↔Assisted PD Taveras 2012106 (235) USA ↔ ↑ ↔ Kolesnyk 2010107 (709) Netherlands ↑ ↑ ↑ ↑ ↔ Singh 2010108 (315) USA ↔ ↔ ↔ ↔ ↔ ↑ Jaar 2009109 (262) USA ↑ ↑ ↔ ↔ ↔Geographical distance to clinic ↔Others Plantinga 2009110 (236) USA ↓Clinic with > 50 PD patients Tonelli 2007111 (26,775) Canada ↓Geographical distance to