patients were invited to visit both HD and PD unit, received information booklets and films, and were encouraged to discuss dialysis with current patients. The intervention included nephrologist placement of PD catheters, identification and training of family members or nursing home staff, increased social support, early ESRD education, and provision of in-center intermittent PD for selected patients. Individual elements of the program were evaluated. There were significant increases in the number of PD patients following training of nursing home staff, training of family members and providing support, early ESRD education, improving home conditions, and nephrologist catheter placement. The percent of patients choosing PD increased from 19% to 76% (P = .001) and the percent of dialysis patients at the facility who were on PD increased from 16% to 40%. In the study from Spain (reported above), patients who received PD by their choice had lower mortality than those forced to accept PD for medical reasons (3.5% vs 20.4%, P < .001).76 The peritonitis rate was also lower (0.46 vs 0.82 per year at risk, P < .05). Findings from an interview with 188 HD and PD patients who began dialysis at least 3 months prior found no significant difference in “depressed mood” (Beck Depression Index score > 9) but higher quality of life (General Health Perceptions score ≥ 70) in HD patients compared to PD patients. 100 There were no differences across modalities in patients reporting negative effects of their current dialysis modality for aspects of daily life (ability to perform daily tasks, ability to control your life, relationships, getting needed sleep, anxiety, or interest in sex). There was a difference in feelings about how you look with a higher percentage of CCPD patients reporting a negative effect compared to either CAPD patients or HD patients. Using a time trade-off format, approximately 38% of HD patients would switch to CAPD if it increased survival time by 20%; approximately 66% would switch for a 100% increase. Similar values were reported for CAPD and CCPD patients in regard to switching to HD. Choice of PD or HD was reported for patients from the NECOSAD cohort.84 Of 1,346, 864 (64%) made their own choice (52% HD, 48% PD). The choice of HD was significantly more likely for age groups 55 to 65 years, 65 to 70 years, and 70 years and older compared to 18 to 40 years. There was no significant difference for patients 40 to 55 compared to 18 to 40 years. Females and patients living alone were significantly more likely to choose HD while patients with greater serum albumin and who received pre-dialysis care were more likely to choose PD. Technique survival at 12 months for patients who chose their dialysis modality was 93% for HD patients and 74% for PD patients. At 24 months, the corresponding values were 91% and 62%. Two studies from the US looked at factors associated with choice of PD as initial dialysis modality. 101,102 In a study of 2,344 incident HD and 670 incident PD patients, black or “other” 48 Home-based versus In-center Dialysis Evidence-based Synthesis Program race (vs white), lower socioeconomic status, and older age (65 to 74 years vs 45 to 54 years) were associated with decreased likelihood of selecting PD. Gender, renal diagnosis, and timing of referral were not significant predictors. 101 In an earlier study of over 10,000 patients, African American race (vs white), age 20 or older (versus under 20 years), moderately or severely impaired functional status (vs normal), 12 or fewer years of education, and not being a home owner were associated with decreased choice of PD.102 Employment or student status and living with family members were associated with increased use of PD although in a multivariable analysis, the association was not statistically significant. Gender was also not a significant predictor of choice. Technique Survival Fourteen studies presented data on factors associated with technique survival for PD.78,103-115 One additional study reported change in technique survival over time comparing data from patients initiating PD between 1995 and 2000 with data from patients initiating PD from 2006 and 2009.116 Additional study information is presented in Appendix C, Table 6. Eight studies were from the US, 78,103,106,108-110,113,115 2 from the Netherlands,107,114 2 from Canada, 111,116 and one each from Australia/New Zealand, 112 France, 104 and Ireland. 105 Sample sizes ranged from 118114 to 41,197.113 There were 7 registry studies, 78,103,104,111-113,116 4 reports from prospective clinical cohort studies,107,109,110,114 and 4 retrospective studies, each from a single center.105,106,108,115 Follow-up times ranged from 1 to 9 years. Across the studies, the patient populations were similar with the exception of one study that enrolled only patients 75 years of age or older. In the remaining studies, mean ages ranged from 50 to 68 years and 49% to 65% were male. Technique failure was defined in most studies as a switch from PD to HD. Four studies identified switches of 30 days or more103,109,110,112 while others included switches of 60104,113 or 90 days or more. 111,116 Five studies did not specify a duration of HD.78,105,106,114,115 One study defined failure as a permanent switch to HD or death on PD.107 Another study