program (TOPs) to standard information in a nonrandomized trial.93 One analysis included 30,217 incident patients, 20.057 of whom attended TOPs. A second analysis included 2,800 matched pairs (TOPs or standard education matched on age, gender, race, diabetes, and geographic area). Of the 20,057 TOPs attendees, 27% chose incenter HD, 24% chose home-based HD, 13% chose transplant, 0.2% chose no therapy, and 35% did not make a choice. Follow-up data were available for 5,565. Twenty-five percent started a home-based dialysis therapy (predominantly PD). Among patients who did not attend TOPS, 3% started a home-based dialysis therapy. It was noted that TOPs attendees were younger, more 46 Home-based versus In-center Dialysis Evidence-based Synthesis Program likely white, and had fewer comorbid conditions. Of the 2,800 matched pairs, 24% of TOPs attendees and 4% of non-attendees chose PD (OR 7.73 [95% CI 3.26, 18.32]). Participants in the CHOICE study were asked to complete a survey about satisfaction with dialysis care. 94 The analysis focused on patients from centers that offered both HD and PD. PD patients were more likely to rate as “excellent” the amount of information they received on choosing HD or PD (relative probability 2.65 [95% CI 2.21, 3.02]) and the amount of dialysis information (relative probability 2.07 [95% CI 1.78, 2.32]). A recent retrospective cohort study from Canada identified reasons for not choosing PD after expressing an intention to initiate PD. 95 PD was actually initiated by 59% of those who expressed an intention to initiate PD. Patient reasons included preference for hospital-based treatment (37%) and lack of space in home (1.6%). Medical reasons included an acute start to dialysis (37%), abdominal surgeries (8%), hernia (3%), and obesity (2%). A prospective cohort study from France reported outcomes from patients who expressed a preference for PD or HD. 96 HHD was not an option in the region of France where the study took place. Of 177 patients who received information on dialysis modalities prior to starting dialysis, 82 (46%) preferred PD. Forty-five of these patients went to RRT with 21 (47%) receiving PD. Of 49 patients preferring HD, 33 went to RRT with 32 (97%) receiving HD. Of 34 patients who were undecided, 11 went to RRT with 9 (82%) receiving HD. Twelve patients were reluctant to undergo dialysis. Three went to RRT with all receiving HD. A separate group of 51 patients in this study had been on HD for less than one month at the time of the information sessions having received no formal information prior to starting on HD.96 Fourteen of these patients (27%) preferred PD and, of 12 patients alive at 3 months, 4 (33%) had switched to PD. Twenty-six preferred to stay with HD and 25 were alive at 3 months. Eleven were undecided but all stayed with HD and were alive at 3 months. Reasons for preferring PD included ability to receive treatment at home, autonomy, comfort to travel, and employment compatibility. Reasons for preferring HD included treatment in a medical facility, autonomy, socioeconomic criteria, socializing/security, and reluctance to have an intra-abdominal catheter. Mismatches between preference and treatment were noted only for 29 patients who expressed a preference for PD. The mismatches were due to medical causes (predominantly abdominal contraindications) in 48% and other causes (including medical center transfer, adverse opinion of family or employer, and change of opinion) in 52%. A survey of patients in the UK who had already made a modality choice following an education program reported differences between HD (n=82) and PD (n=24) patients.97 The PD patients were younger than the HD patients, had lower comorbidity scores, and were more likely married and employed or in school, and less likely living alone. Patients who chose PD identified the following factors as significantly more important than did the patients who chose HD: receiving written information on the modality, the modality fitting with lifestyle, and family/home/work circumstances. Patients who chose HD scored past medical history significantly more important than did the patients choosing PD. A study from Italy looked at time of referral relative to start of dialysis (≤ 3 month or > 3 months) and, for patients referred more than 3 months before dialysis, the effects of a unstructured pre-dialysis clinic versus a formal multidisciplinary pre-dialysis care program. 98 47 Home-based versus In-center Dialysis Evidence-based Synthesis Program Patients at the study centers were encouraged to consider PD if they had no major clinical or psychological contraindications or personal unwillingness. Participation in modality selection was less common for patients referred 3 months or less before dialysis (63% vs 78%, P = .015) as was choice of PD (30% vs 48%, P = .006). There was no difference in participation in selection or choice of PD between patients receiving standard pre-dialysis care or multidisciplinary care. More patients receiving multidisciplinary pre-dialysis care had a planned dialysis start compared to those receiving standard care (91% vs 39%, P < .001) and choice of PD was higher in those with a planned start (56% vs 24%, P < .001). A before and after study from the US evaluated the effect of a comprehensive infrastructure change in dialysis care. 99 All