Division of Geriatrics, Department of Medicine, Queen Elizabeth Hospital, Hong Kong, China. 2

Transcription

1 Journals of Gerontology: MEDICAL SCIENCES The Author Published by Oxford University Press on behalf of The Gerontological Society of America. Cite journal as: J Gerontol A Biol Sci Med Sci 2014 March;69(3): All rights reserved. For permissions, please journals.permissions@oup.com. doi: /gerona/glt098 Advance Access publication August 2, 2013 Outcomes in Older Adults With Stage 5 Chronic Kidney Disease: Comparison of Peritoneal Dialysis and Conservative Management Chun Keung Shum, 1,2 Kui Fu Tam, 1,2 Wai Leung Chak, 3 Tuen Ching Chan, 4 Ying Fai Mak, 1 and Ka Foon Chau 3 1 Division of Geriatrics, Department of Medicine, Queen Elizabeth Hospital, Hong Kong, China. 2 Medical Unit, Hong Kong Buddhist Hospital, Hong Kong, China. 3 Division of Nephrology, Department of Medicine, Queen Elizabeth Hospital, Hong Kong, China. 4 Division of Geriatrics, Department of Medicine, The University of Hong Kong, Queen Mary Hospital, Hong Kong, China. Address correspondence to Chun Keung Shum, MBBS, Division of Geriatrics, Department of Medicine, Queen Elizabeth Hospital, 30 Gascoigne Road, Kowloon, Hong Kong, China. shum_ck@yahoo.com.hk Background. Data on the outcomes of older adults receiving peritoneal dialysis (PD), especially those who are dependent and have multiple comorbidities, are scarce. Methods. In a retrospective cohort study, we compared older adults ( 65 years) with stage 5 chronic kidney disease receiving PD (PD group) with those receiving conservative management (conservative group). Baseline characteristics (demographics and clinical, functional, socioeconomic, and laboratory parameters) were collected, and study outcomes (patient survival, emergency hospitalization, institutionalization, and palliative and end-of-life care) were compared between groups. Results. We included 199 eligible participants aged years (mean ± standard deviation 73.8 ± 5.4 years; 157 in the PD group and 42 in the conservative group). The PD group had a longer survival (median [interquartile range]: 3.75 [ ] vs 2.35 [ ] years, p <.001), lower emergency hospitalization rates (1.63 [ ] vs 3.51 [ ] per person-year, p <.01) and hospital days (16.17 [ ] vs [ ] days per person-year, p =.03), and no increased risk of institutionalization compared with the conservative group. Age (hazard ratio [HR] for 1-year increase 1.06, 95% confidence interval [CI] ), modified Charlson s Comorbidity Index (HR 1.36, 95% CI ), impairment in basic activities of daily living (HR 2.11, 95% CI ), and emergency dialysis (HR 1.67, 95% CI ) were independent predictors of mortality in the PD group. Conclusion. PD is a viable treatment option in older adults with stage 5 chronic kidney disease. Age alone should not preclude dialysis. Comprehensive geriatric assessment can prognosticate and facilitate shared decision making to commence dialysis in older adults. Key Words: Chronic kidney disease Elderly Dialysis Conservative management Outcome. Received December 23, 2012; Accepted May 26, 2013 Decision Editor: Stephen Kritchevsky, PhD Age is no longer seen as a contraindication to dialysis. Owing to aging population and improved management of comorbid diseases, there are increasing numbers of older adults with stage 5 chronic kidney disease (CKD) (1,2). Technological advances in renal replacement therapy (RRT), increased acceptance, and more liberal access to dialysis have resulted in rapid expansion in geriatric dialysis population (3 5). The fastest growth has been seen particularly in the age group older than 75 years, many of whom are dependent, frail, and have multiple comorbidities (6,7). In the United States, individuals older than 75 years showed a 67% increase in the incidence of stage 5 CKD compared with 24% for those between 5 and 74 years during , and the highest rates of dialysis initiation were found in those between 75 and 84 years (2). In Hong Kong, there were 3321 peritoneal dialysis (PD) patients in Among them, 1465 (44.1%) were aged 65 years or above (8,9). The benefits of dialysis in older adults, especially those who are very old, are dependent, and have multiple comorbidities, are questionable compared with those of conservative management (4). Hong Kong has a PD-first policy and follows the international guidelines on the standard of dialysis care (9). People with stage 5 CKD will receive PD as the first modality of long-term RRT unless contraindicated. In 2007, 80% of patients with stage 5 CKD (median age 62.3 years) were on PD in Hong Kong (10,11). Previous studies were limited and mainly focused on 308

2 PERITONEAL DIALYSIS IN OLDER ADULTS 309 survival outcomes in older adults receiving hemodialysis. There are little data regarding the efficacy of PD in older adults. We performed a study to compare the survival and nonsurvival outcomes of PD versus conservative management in Chinese older adults. Methods Study Design and Recruitment A retrospective cohort study was conducted in Queen Elizabeth Hospital in Hong Kong serving a cluster of 0.5 million population (Kowloon Central Cluster). In this cluster, all patients before starting long-term RRT would have dialysis assessment. All Chinese older adults ( 65 years) with stage 5 CKD (defined as estimated glomerular filtration rate [egfr] < 15 ml/min/1.73 m 2 using the Modification of Diet in Renal Disease equation) (12) who received dialysis assessment were retrospectively identified over a 7-year period (from July 1, 2003 to June 30, 2010). The date of study entry was the date of dialysis assessment, and the date of study end was on December 31, Participants were followed up till death or for a minimum of 1.5 years if still alive on the date of study end. They were categorized into conservative group (defined as those who received no dialysis or conservative management after dialysis assessment ) and PD group (defined as those who received PD as the first modality of longterm RRT or had made a decision to commence on PD and had begun preparation for the temporary dialysis, PD catheter insertion but had died before dialysis initiation). It was a complex shared decision making between patients/ families and doctors to commence dialysis or conservative management, and the final decision was largely voluntary to the patients. Participants in the conservative group would receive the same level of medical care as those in the PD group except no PD. Regular follow-ups were provided in medical clinics and/or nurse-led clinics to optimize medical management of renal disease (eg, anemia treatment by erythropoietin, fluid and electrolyte balance, management of uremic symptoms), and telephone hotline was provided for symptom control. Data Collection Data were collected by retrospective review of computerized records and case notes. Baseline characteristics of participants were recorded during dialysis assessment : (a) demographics age and gender; (b) clinical data etiology of CKD, modified Charlson s Comorbidity Index (mcci) (13), late referral (defined as dialysis assessment required shortly within 1 month after seen by nephrologists), and emergency dialysis (defined as temporary dialysis required during emergency admission before starting long-term RRT); (c) functional aspects basic activities of daily living (BADL; independent or impaired) and mobility (independent walker, assisted walker, chairbound/bedridden); (d) socioeconomic factors marital status, educational level, living status, receiving Comprehensive Social Security Allowance (CSSA), any helper for PD; and (e) laboratory parameters egfr, sodium, potassium, calcium, phosphate, bicarbonate, albumin, and hemoglobin levels. Other data on participants reasons for no dialysis in the conservative group and need for assistance for PD in the PD group (self-pd [PD performed by participants] and assisted PD [PD performed by their caregivers]), were collected. Outcomes Primary outcome was survival (defined as patient survival from the date of entry into stage 5 CKD to death or study end). The date of entry into stage 5 CKD was the date of first recorded value of egfr less than 15 ml/ min/1.73 m 2 and participants need to fulfill two criteria: (a) at least one value of egfr in the range of ml/ min/1.73 m 2 and (b) all subsequent recorded values of egfr less than 15 ml/min/1.73 m 2. Secondary outcomes included emergency hospitalization rate (number of emergency hospitalizations per year of follow-up), emergency hospital days (number of hospital days due to emergency hospitalizations per year of follow-up), institutionalization (new admission to a nursing home), and palliative and endof-life care (renal palliative care, bothersome interventions during end-of-life period [defined as one of the following invasive interventions within 1 month before death: operative or endoscopic intervention, intubation or mechanical ventilation, cardiopulmonary resuscitation, tube feeding for artificial nutrition]). The study was approved by the Research Ethics Committee at the Kowloon Central Cluster, Hospital Authority, Hong Kong. Statistical Analyses Groups were compared using chi-square test or Fisher exact test for categorical variables and using t test or Mann Whitney U test for continuous variables as appropriate. Kaplan Meier analyses were used to compare survival and institutionalization outcomes with group differences analyzed by log-rank test. Cox proportional regression analysis was used to determine the predictors of mortality. Linear regression was used to determine the predictors of emergency hospitalization after log transformation. Exploratory analyses using each of the variables (namely, demographics, clinical data, functional aspects, socioeconomic factors, laboratory parameters, and treatment modality) were used to assess the independent impact of predictors. Results were expressed as percentage (%) or number (n) for categorical variables, and means ± standard deviation (SD) or median (interquartile range) for continuous variables. In Cox proportional hazards model, hazard ratio (HR) and 95% confidence interval (95% CI) were estimated for

3 310 SHUM ET AL. each independent predictor. In linear regression model, standardized coefficient beta (β) was estimated for each independent predictor after log transformation. All variables had less than 5% missing data. Participants with missing data from dialysis assessment were excluded for the analysis of that variable. Statistical significance was set at p <.05. Statistical analyses were performed using SPSS, Version 19. Results A total of 199 older adults with stage 5 CKD aged years (mean ± SD: 73.8 ± 5.4 years) were included in this study: 42 (21.1%) in the conservative group and 157 (78.9%) in the PD group (Table 1). Participants had a median follow-up of 1.96 years (interquartile range years). In the PD group, 102 (71.8%) required assisted PD. Table 1. Baseline Characteristics Between Conservative and PD Groups at the Time of Dialysis Assessment Baseline Characteristics Conservative Group (n = 42)* PD Group (n = 157)* p Demographics Age (y) 75.3 ± ± Male (%) Etiology of CKD Diabetic nephropathy (%) Hypertensive nephrosclerosis (%) Glomerulonephritis (%) Polycystic kidney (%) Obstructive uropathy (%) Others/unknown (%) Comorbidity mcci 4.6 ± ± Diabetes mellitus (%) Coronary artery disease (%) Congestive heart failure (%) Stroke (%) Peripheral vascular disease (%) Chronic pulmonary disease (%) Liver disease (%) Cancer (%) Dementia (%) Clinical data Late referral (%) Functional Aspects BADL Independent BADL (%) Impaired BADL (%) Mobility Independent walker (%) Assisted walker (%) Chairbound/bedridden (%) Socioeconomic factors Single/divorced/separated (%) Illiterate (%) Living alone/in nursing home (%) Living in public housing (%) Receiving CSSA (%) No helper for PD (%) <.001 Laboratory parameters egfr (ml/min/1.73m 2 ) 6.8 ± ± Sodium (mmol/l) ± ± Potassium (mmol/l) 4.5 ± ± Calcium (mmol/l) 2.2 ± ± Phosphate (mmol/l) 1.9 ± ± Bicarbonate (mmol/l) 20.9 ± ± Albumin (g/l) 36.9 ± ± Hemoglobin (g/dl) 8.7 ± ± Notes: BADL = basic activities of daily living; CKD = chronic kidney disease; CSSA = Comprehensive Social Security Allowance; egfr = estimated glomerular filtration rate; mcci = modified Charlson s Comorbidity Index; PD = peritoneal dialysis. *Results expressed as % or mean ± standard deviation whenever appropriate.

4 PERITONEAL DIALYSIS IN OLDER ADULTS 311 Baseline Characteristics The conservative group was older (75.3 ± 5.7 vs 73.4 ± 5.3 years, p =.04), was more likely to have stroke (28.6% vs 14.0%, p =.03) and receive CSSA (24.4% vs 9.7%, p =.01), and had no helper for PD (58.8% vs 17.6%, p <.001) compared with the PD group. There was no significant difference in other characteristics at baseline (Table 1). Participants Reasons for No Dialysis In the conservative group, reasons for no dialysis were documented in 40 participants, and 35% (n = 14) had more than one reason. The most common reason was inability to perform PD but no helper available or refused institutionalization (40.0%, n = 16), followed by patient s choice (37.5%, n = 15), afraid of treatment and complication burden (22.5%, n = 9), advanced age (20.0%, n = 8), and comorbidities (17.5%, n = 7). Survival By the end of the study, 92.9% (n = 39) died in the conservative group, and 61.1% (n = 96) died in the PD group. The PD group had a longer median survival than the conservative group (median [interquartile range]: 3.75 [ ] vs 2.35 [ ] years, p <.001). Kaplan Meier curves showed survival advantage in the PD group (p <.001; Figure 1). In Cox proportional hazards model, age, mcci, BADL impairment, and emergency dialysis were independent predictors of mortality in the PD group (Table 2). After adjustment for age, mcci, and BADL impairment, survival advantage persisted in the PD group (p <.001, HR 0.46, 95% CI ) compared with the conservative group. In subgroup analyses, the survival advantage was preserved in those receiving PD with low comorbidity or independent BADL but was lost in those with high comorbidity or BADL impairment (Figures 2 and 3). Hospitalization The PD group had significantly lower emergency hospitalization rates (1.63 [ ] vs 3.51 [ ] per person-year, p <.01) and hospital days (16.17 [ ] vs [ ] days per person-year, p =.03) than the conservative group. In multivariate linear regression, independent predictors of emergency hospitalization rate and hospital days (after log transformation) included PD (β = 0.18, p =.01; β = 0.14, p =.04), age (β = 0.21, p <.01; β = 0.17, p =.02), mcci (β = 0.22, p <.01; β = 0.17, p =.01), and BADL impairment (β = 0.20, p <.01; β = 0.22, p <.01). The differences in emergency hospitalization between groups remained significant even after adjustment for age, mcci, and BADL impairment. Figure 1. Kaplan Meier survival curves for participants in conservative and peritoneal dialysis groups. CKD = chronic kidney disease; PD = peritoneal dialysis.

5 312 SHUM ET AL. Table 2. Cox Proportional Hazards Model for Independent Predictors of Mortality in the PD Group Independent Predictors of Mortality in the PD Group HR (95% CI) p Age (y) 1.06 ( ).01 mcci 1.36 ( ) <.001 BADL impairment 2.11 ( ) <.01 Emergency dialysis 1.67 ( ).02 Notes: BADL = basic activities of daily living; CI = confidence interval; HR = hazard ratio; mcci = modified Charlson s Comorbidity Index; PD = peritoneal dialysis. Institutionalization The PD group had no significant increased risk of institutionalization compared with the conservative group (p =.20). Palliative and End-of-Life Care The conservative group was more likely to receive renal palliative care (15.4% vs 0%, p <.001) and less likely to receive bothersome interventions during end-of-life period (47.2% vs 85.9%, p <.001) than the PD group. Figure 2. Comparison of Kaplan Meier survival curves for participants in conservative and peritoneal dialysis groups with low and high comorbidity. mcci = modified Charlson s Comorbidity Index; PD = peritoneal dialysis. Figure 3. Comparison of Kaplan Meier survival curves for participants in conservative and peritoneal dialysis groups with independent and impaired BADL. BADL = basic activities of daily living; PD = peritoneal dialysis.

6 PERITONEAL DIALYSIS IN OLDER ADULTS 313 Discussion PD can be a viable treatment option in older adults with stage 5 CKD. In this study, PD could improve survival and reduce the risk of emergency hospitalization without an increased risk of institutionalization compared with conservative management. To our knowledge, this is the first study to compare the survival and nonsurvival outcomes between PD and conservative management in Chinese older adults with stage 5 CKD. Age alone should not preclude dialysis. In this study, survival benefit was present in older adults receiving PD compared with conservative management. Previous studies mainly focused on comparing survival outcomes between groups of adults predominantly receiving hemodialysis versus conservative management. The studies by Chandna and coworkers, Carson and coworkers, and Murtagh and coworkers showed that people on dialysis lived significantly longer than people managed conservatively, but conservatively managed people were older or more likely to have high comorbidity (14 17). The study by Joly and coworkers (18) also revealed a large survival benefit in octogenarians proposed dialysis, but the group not proposed dialysis was more likely to have diabetes, have a lower functional score, be referred late, and be socially isolated. However, the study by Smith and coworkers (19) showed no significant survival difference between conservatively managed people and people recommended for conservative management but opting for and treated by dialysis. As a whole, survival benefit of dialysis is probably present in selected groups of older adults. In this study, older adults receiving PD had significantly lower emergency hospitalization. A previous study has shown that people on dialysis (predominantly hemodialysis) had higher hospitalization rates and similar hospitalfree survivals compared with conservatively managed people (16). The discrepancy may be explained by homebased PD having advantages over in-center hemodialysis in older adults. PD can avoid continuous and exhaustive trips to and from the hospital, have less intrusion into lifestyle, and better preserve residual renal function with less hemodynamic stress during treatment, and there is no need for vascular access (10,20,21). In this study, the PD group had no increased risk of institutionalization. Institutionalization was found to be an indirect manifestation of functional decline and increased caregiver burden (22,23). We proposed that proper selection of older adults receiving dialysis should not result in unnecessary institutionalization and sacrifice of function. Geriatric population is a heterogeneous population with a variety of functions and comorbidities (10). Chronological age is not equivalent to physical and mental age (11). In this study, mcci and BADL impairment were independent predictors of survival and emergency hospitalization. Past studies also showed that the survival benefit of dialysis was lost in older adults (>75 years) with high comorbidity or ischemic heart disease compared with conservative management (15,17), and a prediction model for 6-month mortality was validated in older adults receiving dialysis including comorbidities and functional status (total dependency for transfers) (24 26). In another study, the use of surprise question ( Would I be surprised if this patient died in the next year? ) was effective in identifying hemodialysis patients who had a higher risk for early mortality. Not surprised group was significantly older, had a higher comorbidity score, and had a lower performance status (27,28). Hence, comprehensive geriatric assessment including comorbidity and functional status should be considered in addition to age in older adults to commence dialysis (24,29 31). In older adults with stage 5 CKD, especially in those who are very old with multiple comorbidities and functional impairment, conservative management is a reasonable alternative. The survival advantage of dialysis may probably be lost in these people, and they may suffer from repeated hospitalizations. Moreover, the conservative group was more likely to receive renal palliative care and less likely to receive bothersome interventions during end-of-life period. Previous studies also found that in conservatively managed people, age older than 75 years and female gender were independent predictors of improved survival (15), and they were more likely to die at home or in a hospice (4,16). Conservative management, also known as maximum conservative management, is not simply defined by no dialysis (32). It shifts the focus from efforts to prolong life to those that focus on symptom control, quality of life, and care support by a multidisciplinary team approach. It embodies active disease management (eg, anemia treatment by erythropoietin, fluid and electrolyte balance, management of uremic symptoms), end-of-life care, and ongoing psychosocial and spiritual support to patients and family/ caregivers (4,5,7). Nevertheless, prognostication and decision making to commence dialysis remain difficult and should be individualized (1,20,32 34). Collaboration between geriatric and nephrology teams can be essential to provide multidisciplinary and multifactorial assessment and intervention in older adults with stage 5 CKD (24,35). There were a number of limitations in the study. First, the study was retrospective in nature, but it is unethical to randomize people to dialysis or conservative management. Second, the sample size was small. However, the benefit can be clearly shown by this sample size. Third, selection bias might be present in this study. A group of older adults with good functioning and low comorbidity might be selected for dialysis assessment (Table 1), and there was no data on the percentage of older adults with stage 5 CKD who were not referred for dialysis assessment. Last, symptom burden, quality of life, and functional outcomes were not assessed. Future large prospective cohort studies investigating symptom burden, quality of life, and functional outcomes in

7 314 SHUM ET AL. addition to survival of older adults receiving dialysis versus conservative management are warranted (15,17). Conclusion PD is a viable treatment option in older adults with stage 5 CKD. Age alone is not a contraindication to dialysis. Comprehensive geriatric assessment can be important to prognosticate and facilitate shared decision making to commence dialysis in older adults. In those who are very old with multiple comorbidities and functional impairment, conservative management is a reasonable alternative. References 1. Brunori G, Viola BF, Maiorca P, Cancarini G. How to manage elderly patients with chronic renal failure: conservative management versus dialysis. Blood Purif. 2008;26: Misra M. Dialysis in the elderly. Blood Purif. 2008;26: Kooman JP, Cornelis T, van der Sande FM, Leunissen KM. Renal replacement therapy in geriatric end-stage renal disease patients: a clinical approach. Blood Purif. 2012;33: Burns A, Carson R. Maximum conservative management: a worthwhile treatment for elderly patients with renal failure who choose not to undergo dialysis. J Palliat Med. 2007;10: Burns A, Davenport A. Maximum conservative management for patients with chronic kidney disease stage 5. Hemodial Int. 2010; 14(suppl 1):S32 S Selvarajah V, Isles C. End-stage renal disease in the very elderly. J R Coll Physicians Edinb. 2007;37: Dasgupta I, Rayner HC. In good conscience safely withholding dialysis in the elderly. Semin Dial. 2009;22: Ho YW, Leung CB, Choy BY, et al. Renal registry and peritoneal dialysis management: the Hong Kong perspective. Perit Dial Int. 2008;28(suppl 3):S12 S Li PK, Law MC, Chow KM, et al. Good patient and technique survival in elderly patients on continuous ambulatory peritoneal dialysis. Perit Dial Int. 2007;27(suppl 2):S196 S Dimkovic N, Oreopoulos D. Management of elderly patients with end-stage kidney disease. Semin Nephrol. 2009;29: Brown EA, Johansson L. Epidemiology and management of end-stage renal disease in the elderly. Nat Rev Nephrol. 2011;7: Ma YC, Zuo L, Chen JH, et al. Modified glomerular filtration rate estimating equation for Chinese patients with chronic kidney disease. J Am Soc Nephrol. 2006;17: Hemmelgarn BR, Manns BJ, Quan H, Ghali WA. Adapting the Charlson Comorbidity Index for use in patients with ESRD. Am J Kidney Dis. 2003;42: O Connor NR, Kumar P. Conservative management of end-stage renal disease without dialysis: a systematic review. J Palliat Med. 2012;15: Chandna SM, Da Silva-Gane M, Marshall C, Warwicker P, Greenwood RN, Farrington K. Survival of elderly patients with stage 5 CKD: comparison of conservative management and renal replacement therapy. Nephrol Dial Transplant. 2011;26: Carson RC, Juszczak M, Davenport A, Burns A. Is maximum conservative management an equivalent treatment option to dialysis for elderly patients with significant comorbid disease? Clin J Am Soc Nephrol. 2009;4: Murtagh FE, Marsh JE, Donohoe P, Ekbal NJ, Sheerin NS, Harris FE. Dialysis or not? A comparative survival study of patients over 75 years with chronic kidney disease stage 5. Nephrol Dial Transplant. 2007;22: Joly D, Anglicheau D, Alberti C, et al. Octogenarians reaching endstage renal disease: cohort study of decision-making and clinical outcomes. J Am Soc Nephrol. 2003;14: Smith C, Da Silva-Gane M, Chandna S, Warwicker P, Greenwood R, Farrington K. Choosing not to dialyse: evaluation of planned non-dialytic management in a cohort of patients with end-stage renal failure. Nephron Clin Pract. 2003;95:c40 c Buemi M, Lacquaniti A, Bolignano D, et al. Dialysis and the elderly: an underestimated problem. Kidney Blood Press Res. 2008;31: Li PK, Chow KM. Peritoneal dialysis patient selection: characteristics for success. Adv Chronic Kidney Dis. 2009;16: Gaugler JE, Duval S, Anderson KA, Kane RL. Predicting nursing home admission in the U.S: a meta-analysis. BMC Geriatr. 2007;7: Tsuji I, Whalen S, Finucane TE. Predictors of nursing home placement in community-based long-term care. J Am Geriatr Soc. 1995;43: Swidler MA. Geriatric renal palliative care. J Gerontol A Biol Sci Med Sci. 2012;67: Couchoud C, Labeeuw M, Moranne O, et al.; French Renal Epidemiology and Information Network (REIN) registry. A clinical score to predict 6-month prognosis in elderly patients starting dialysis for end-stage renal disease. Nephrol Dial Transplant. 2009;24: Hudson M, Weisbord S, Arnold RM. Prognostication in patients receiving dialysis #191. J Palliat Med. 2007;10: Cohen LM, Ruthazer R, Moss AH, Germain MJ. Predicting six-month mortality for patients who are on maintenance hemodialysis. Clin J Am Soc Nephrol. 2010;5: Moss AH, Ganjoo J, Sharma S, et al. Utility of the surprise question to identify dialysis patients with high mortality. Clin J Am Soc Nephrol. 2008;3: Bowling CB, Muntner P. Epidemiology of chronic kidney disease among older adults: a focus on the oldest old. J Gerontol A Biol Sci Med Sci. 2012;67: Bowling CB, Sawyer P, Campbell RC, Ahmed A, Allman RM. Impact of chronic kidney disease on activities of daily living in communitydwelling older adults. J Gerontol A Biol Sci Med Sci. 2011;66: Gopinath B, Harris DC, Burlutsky G, Mitchell P. Use of community support services and activity limitations among older adults with chronic kidney disease. J Gerontol A Biol Sci Med Sci. 2013;68: Fassett RG, Robertson IK, Mace R, Youl L, Challenor S, Bull R. Palliative care in end-stage kidney disease. Nephrology (Carlton). 2011;16: Jassal SV, Kelman EE, Watson D. Non-dialysis care: an important component of care for elderly individuals with advanced stages of chronic kidney disease. Nephron Clin Pract. 2011;119(suppl 1):c5 c Swidler M. Geriatric renal palliative care is coming of age. Int Urol Nephrol. 2010;42: Moranne O, Couchoud C, Vigneau C; PSPA Study Investigators. Characteristics and treatment course of patients older than 75 years, reaching end-stage renal failure in France. The PSPA study. J Gerontol A Biol Sci Med Sci. 2012;67: