Clinical practices and outcomes in elderly hemodialysis patients: results from the Dialysis Outcomes and Practice Patterns Study (DOPPS)

Abstract

Background and objectives: Demand for hemodialysis among elderly patients is increasing worldwide. Although clinical care of this high-risk group is complex and challenging, no guidelines exist to inform hemodialysis practices. The Dialysis Outcomes and Practice Patterns Study (DOPPS) provides a unique opportunity to assess dialysis practices and associated outcomes among elderly versus younger patients on chronic in-center hemodialysis in 12 countries.

Design, setting, participants, & measurements: Clinical characteristics, dialysis practices, and outcomes of elderly versus younger patients were compared among participants in four DOPPS regions in 2005 through 2007.

Results: Although participant mean age increased over time in all DOPPS countries, the percentage of elderly varied widely. Overall, comorbidities and malnutrition were more common in the elderly. Fistulae were used less frequently among elderly versus younger patients in Europe and North America but not in Australia, New Zealand, and Japan. No difference in treatment time was observed between elderly and younger patients after normalizing for body weight. In all regions, ultrafiltration rates were lower among elderly patients. Elderly patients reported poorer quality of life with respect to the physical but not mental component scores. Mortality risk was three- to sixfold higher in the elderly group, whereas causes of death overall were similar for elderly and younger patients.

Conclusions: Elderly patients represent a different proportion of DOPPS participants across countries, possibly reflecting differences in policies and clinical practices. In general, hemodialysis practices in the elderly reflected each region's clinical patterns, with some variation by age group depending upon the practice.

Figure 1.

Mean age of Dialysis Outcomes and Practice Patterns Study (DOPPS) participants at study entry across DOPPS phases. On the basis of cross sections of participants in DOPPS I (1998 through 1999, with the exception of the United States (US), where most DOPPS I participants were enrolled in 1997; n = 8542), DOPPS II (2002 through 2003; n = 9117), and DOPPS III (2005 through 2006; n = 8161). Panel A depicts countries that participated in DOPPS I, II, and III; panel B depicts countries that participated only in DOPPS II and DOPPS III. P value for comparison between DOPPS I and DOPPS II: <0.05 in all countries except the United States; P value for comparison between DOPPS II and DOPPS III: <0.05 in Belgium (BE), Japan (JP), Sweden (SW), and Australia and New Zealand (ANZ). FR, France; IT, Italy; SP, Spain; GE, Germany; UK, United Kingdom; CA, Canada.

Figure 2.

Distribution of patients in age categories across DOPPS countries. Cross section of participants in DOPPS III (2005 through 2007; n = 8161).

Figure 3.

Kidney Disease Quality of Life (KDQoL) physical component summary (PCS) and mental component summary (MCS) scores by age categories across DOPPS regions versus U.S. population norm. Cross section of participants in DOPPS III (2005 through 2007; n = 8161). Europe includes United Kingdom, France, Germany, Italy, Spain, Belgium, and Sweden; North America includes United States and Canada; ANZ represents Australia and New Zealand.

Figure 4.

Survival by age at start of dialysis across DOPPS regions. Probability of survival from start of dialysis by region and age group at dialysis start (calculated using unadjusted Cox regression within each region with delayed entry at time of study enrollment, stratified by age group, and accounting for facility clustering effects). Total number of patients = 8161, and the number of deaths = 1337. P value for 75 years versus 45 years: 0.05 in each region. Europe includes United Kingdom, France, Germany, Italy, Spain, Belgium, and Sweden; North America includes United States and Canada; ANZ represents Australia and New Zealand.

Figure 5.

Mortality risk by age at study entry across DOPPS regions. Cross section of participants in DOPPS III (2005 through 2007; n = 8161); number of deaths n = 1337. Reference (Ref) group: patients <45 years of age. Model adjusted for sex, race, body mass index, duration of ESRD, 14 summary comorbid conditions listed in Table 1 (dementia was included in the neurologic diseases), stratified by country, and accounted for facility clustering effect. Mean age in category: <45 years = 36, 45 to 75 years = 62, ≥75 years = 80. Europe includes United Kingdom, France, Germany, Italy, Spain, Belgium, and Sweden; North America includes United States and Canada; ANZ represents Australia and New Zealand; HR, Hazard Ration; CI, confidence interval.

Figure 6.

Distribution of cause of death by age at study entry across DOPPS regions. Analysis was limited to patients for whom data on cause of death was available (patient n = 8010; number of deaths n = 1188). Number of deaths shown in parentheses for n <5.