Predialysis Health, Dialysis Timing, and Outcomes among Older United States Adults

CLINICAL EPIDEMIOLOGY www.jasn.org

Predialysis Health, Dialysis Timing, and Outcomesamong Older United States Adults

Deidra C. Crews,*† Julia J. Scialla,‡ Jiannong Liu,§ Haifeng Guo,§ Karen Bandeen-Roche,|

Patti L. Ephraim,†¶ Bernard G. Jaar,*†** Stephen M. Sozio,*† Dana C. Miskulin,††

Navdeep Tangri,‡‡ Tariq Shafi,*† Klemens B. Meyer,†† Albert W. Wu,¶§§||¶¶*** Neil R. Powe,†††

and L. Ebony Boulware,†¶|| for the Developing Evidence to Inform Decisions aboutEffectiveness (DEcIDE) Patient Outcomes in End Stage Renal Disease Study Investigators

*Division of Nephrology, Department of Medicine, ||Division of General Internal Medicine, Department of Medicine,and ***Department of Surgery, Johns Hopkins University School of Medicine, Baltimore, Maryland; †Welch Center forPrevention, Epidemiology and Clinical Research, Johns Hopkins Medical Institutions, Baltimore, Maryland; ‡Divisionof Nephrology and Hypertension, Department of Medicine, University of Miami Miller School of Medicine, Miami,Florida; §Chronic Disease Research Group, Minneapolis Medical Research Foundation, Minneapolis, Minnesota;Departments of |Biostatistics, ¶Epidemiology, §§Health Policy and Management, and ¶¶International Health, JohnsHopkins Bloomberg School of Public Health, Baltimore, Maryland; **Nephrology Center of Maryland, Baltimore,Maryland; ††Division of Nephrology, Tufts University School of Medicine, Boston, Massachusetts; ‡‡Division of Nephrology,Department of Medicine, Seven Oaks General Hospital, University of Manitoba, Winnipeg, Manitoba, Canada; and†††Department of Medicine, San Francisco General Hospital and University of California, San Francisco, California

ABSTRACTStudies of dialysis initiation timing have not accounted for predialysis clinical factors that could impactpostdialysis outcomes.Weexamined theassociationofpredialysis healthwith timingofdialysis initiation inolderadult patients in the United States and contrastedmorbidity andmortality outcomes among patients with early[estimated GFR (eGFR)$10 ml/min per 1.73 m2] versus later (eGFR,10 ml/min per 1.73 m2) initiation. Weincluded all patients from the US Renal Data System who initiated dialysis between 2006 and 2008, were$67yearsold, andhad$2yearsofpriorMedicarecoverage (n=84,654).Wecalculatedpatients’propensity to initiatedialysis early and matched patients by propensity scores. Cox models were used to compare risks of mortalityand hospitalization among initiation groups. Themajority (58%) of patients initiated dialysis early. Early initiatorswere more likely to have had AKI, multiple congestive heart failure admissions, and other hospitalizations pre-ceding initiation. Among propensity-matched patients (n=61,930), early initiation associated with greater all-cause (hazard ratio [HR], 1.11; 95% confidence interval [95%CI], 1.08 to 1.14), cardiovascular (CV; HR, 1.13; 95%CI, 1.09 to 1.17), and infectious (HR, 1.13; 95%CI, 1.06 to 1.22)mortality and greater all-cause (HR, 1.03; 95%CI,1.01 to 1.05) and infectious (HR, 1.10; 95% CI, 1.07 to 1.13) hospitalizations. There was no difference in CVhospitalizations.Among theseolder adults, early dialysis initiation associateswithgreatermortality andhospital-izations, evenafter accounting forpredialysis clinical factors. Thesefindingsdonot support thecommonpracticeof early dialysis initiation in the United States.

J Am Soc Nephrol 25: 370–379, 2014. doi: 10.1681/ASN.2013050567

There has been a global trend toward initiation ofdialysis at higher levels of kidney function, partic-ularly among older adults.1 This trend may havebeen driven, in part, by clinicians’ beliefs that ear-lier initiation could better preserve patients’ nutri-tional status, thereby decreasing their risks ofmorbidity and mortality.2–4 However, recent stud-ies have questioned the ability of dialysis to prevent

Received May 31, 2013. Accepted July 21, 2013.

Published online ahead of print. Publication date available atwww.jasn.org.

Correspondence: Dr. Deidra C. Crews, Division of Nephrology,Johns Hopkins Bayview Medical Center, 301 Mason F. LordDrive, Suite 2500, Baltimore, MD 21224. Email: [email protected]

Copyright © 2014 by the American Society of Nephrology

370 ISSN : 1046-6673/2502-370 J Am Soc Nephrol 25: 370–379, 2014

http://www.jasn.org

mailto:[email protected]

mailto:[email protected]

functional decline in institutionalized older adults5 and sug-gest greater mortality with early dialysis initiation in severalpatient groups.6–8 Definitive clinical trials have been chal-lenged by the influence of patient and clinician preferencesand the unpredictable course that often accompanies renalfunction decline. The Initiating Dialysis Early and Late(IDEAL) trial, which found no difference in survival, didnot achieve the desired difference in kidney function betweenearly and late initiators, because the majority of patients ran-domized to the late-start group initiated dialysis before reach-ing the target estimated GFR (eGFR). At initiation of dialysisin the IDEAL trial, the mean eGFR was 12.0 ml/min in theearly-start group compared with 9.8 ml/min in the late-startgroup.9 Furthermore, the IDEAL trial may not be generaliz-able to older adults in the United States, because the studyincluded a relatively healthy ESRD population (e.g., 5.5% ofstudy patients had congestive heart failure [CHF]9 comparedwith 32.5% of incident dialysis patients in the United States in2007 had CHF1). Thus, rigorous observational studies stillhave an important role in elucidating optimal dialysis initia-tion strategies among older United States adults.

To our knowledge, which was corroborated by a recentmeta-analysis,7 prior studies of outcomes related to timing ofdialysis initiation have not carefully examined the predialysishealth course (e.g., incidence of AKI or CHF hospitalization)or specifically focused on older adults, the fastest growingUS ESRD population.10 Furthermore, studies have not exam-ined specific causes of morbidity andmortality experienced bypatients after their earlier versus later di-alysis initiation.7

We examined the association of olderUnited States adult patients’ predialysishealth with timing of dialysis initiationand compared morbidity and mortalityoutcomes after early versus later dialysisinitiation using data from predialysisMedicare claims linked to the US RenalData System (USRDS).

RESULTS

Patient Characteristics by Timing ofDialysis InitiationA total of 84,654 patients was included inour study. Exclusions are summarized inFigure 1. Mean age of the study populationwas 76.7 (SD=6.3) years. Median eGFR atdialysis initiation was 11.0 (interquartilerange=8.1–14.3) ml/min per 1.73 m2.The majority (57.7%) initiated dialysisearly (eGFR.10 ml/min per 1.73 m2).Early and later initiators differed substan-tially (Table 1). Examination of predialysisclinical factors revealed AKI to be more

common among early initiators, whereas there were no differ-ences in uremia or hyperkalemia between the groups. Predial-ysis CHF hospitalizations and total hospital days were morecommon among early initiators.

Manycharacteristicswere independentlyassociatedwithearlyinitiation after adjustment for all other variables (Table 1).Among predialysis factors, AKI, uremia, and hyperkalemiawere negatively associated with early initiation. In contrast,CHF admissions and total hospital days were positively associ-ated with early dialysis initiation.

Propensity Matching of CohortA total of 61,930 (71.5%) patients was propensity matched(Figure 1). Median eGFR was 13.3 ml/min per 1.73 m2 forearly initiators and 7.8 ml/min per 1.73 m2 for later initiators.Patient characteristics were well balanced after propensitymatching (Table 1). Most (78.9%) of unmatched patientswere early initiators, with a median eGFR of 13.0 ml/minper 1.73 m2, and larger proportions had experienced AKI(67.5%), greater than two CHF hospitalizations (17.6%),and/or greater than 15 total hospital days (26.4%) in the6 months before dialysis initiation compared with the matchedpatients. Forty-six percent of unmatched patients (n=10,396)died during follow-up.

All-Cause and Cause-Specific MortalityA total of 25,151 (40.6%) patients in our matched cohort diedduring follow-up. Estimatedmedian survival was 23.4 months

Figure 1. Overview of cohort formation.

J Am Soc Nephrol 25: 370–379, 2014 Dialysis Timing among Older Adults 371

www.jasn.org CLINICAL EPIDEMIOLOGY

Table 1. Patient characteristics by timing of dialysis initiation before and after propensity score matching

Patient Characteristics

Before Match, % (n=84,654) Logistic Regression ofPredictors of Early

Initiationa Odds Ratio(95% CI)

After Match, %b (n=61,930)

Later (eGFR,10 ml/min

per 1.73 m2)

Early (eGFR‡10 ml/min

per 1.73 m2)


per 1.73 m2)


per 1.73 m2)

n (%) 35,769 (42.3%) 48,885 (57.7%) — 30,965 (50.0%) 30,965 (50.0%)Mean age, yr (SD) 76.7 (6.3) 77.5 (6.4) — 77.0 (6.3) 77.0 (6.4)Age categories, yr (%)67–69 15.0 12.0 Reference 13.8 13.770–74 26.2 23.4 1.11 (1.05 to 1.16) 25.3 25.375–79 25.6 26.1 1.26 (1.20 to 1.32) 25.9 25.980–84 20.5 23.0 1.40 (1.33 to 1.47) 21.4 21.5$85 12.8 15.4 1.56 (1.47 to 1.65) 13.7 13.6

Men 47.7 58.7 Reference 51.1 50.7Women 52.3 41.3 0.59 (0.57 to 0.60) 49.0 49.3RaceWhite 76.6 78.4 Reference 77.0 76.8Black 18.4 18.0 1.12 (1.07 to 1.16) 18.6 18.8Other 5.0 3.6 0.78 (0.73 to 0.84) 4.4 4.4

Primary cause of ESRD (%)Diabetes mellitus 35.4 42.9 Reference 38.4 38.3Hypertension 37.5 37.4 0.89 (0.85 to 0.93) 38.2 38.2GN 11.7 7.0 0.65 (0.62 to 0.69) 9.3 9.3Other/missing/unknown 15.5 12.7 0.79 (0.75 to 0.83) 14.1 14.2

Year of dialysis initiation2006 37.0 33.8 Reference 35.7 35.82007 32.6 33.1 1.15 (1.11 to 1.19) 32.8 32.82008 30.4 33.1 1.27 (1.22 to 1.31) 31.4 31.4

Dialysis modalityHemodialysis 96.2 95.3 Reference 95.9 95.9Peritoneal dialysis 3.8 4.7 1.26 (1.10 to 1.45) 4.1 4.1

Vascular accessArteriovenous fistula 13.7 13.7 Reference 14.1 14.1Arteriovenous graft 4.0 4.2 1.08 (0.99 to 1.17) 4.1 4.3Catheter 77.4 76.1 0.89 (0.86 to 0.93) 76.5 76.3

Comorbid conditions c

CHF 61.0 76.1 1.32 (1.27 to 1.38) 66.6 67.0Atherosclerotic heart disease 58.3 72.3 1.16 (1.12 to 1.20) 63.5 63.6Cerebrovascular disease 29.1 34.9 1.04 (1.01 to 1.08) 31.0 31.1Diabetes mellitus 58.5 68.2 1.26 (1.21 to 1.31) 62.4 62.5Chronic obstructive pulmonary disease 32.9 42.9 1.19 (1.16 to 1.23) 35.4 35.5Peripheral vascular disease 45.2 54.2 1.08 (1.04 to 1.11) 48.1 48.2Other CV disease 50.2 63.2 1.17 (1.13 to 1.21) 54.3 54.4Dysrhythmia 39.9 54.4 1.23 (1.19 to 1.27) 43.8 43.8Cancer 24.8 23.4 0.91 (0.88 to 0.94) 24.0 24.0Gastrointestinal bleeding 15.9 20.0 1.04 (1.00 to 1.08) 17.0 17.0Liver disease 2.8 4.1 1.37 (1.26 to 1.49) 3.0 3.1

Nursing home resident 8.7 11.6 1.17 (1.11 to 1.23) 9.4 9.5Functional limitation 16.7 20.0 1.04 (1.00 to 1.08) 17.6 17.7Mean serum albumin, g/dl (SD) 3.2 (0.7) 3.2 (0.7) — 3.2 (0.7) 3.2 (0.7),3.0 33.5 32.1 0.96 (0.92 to 1.01) 32.9 33.13.0–3.4 17.5 17.9 Reference 17.6 17.6.3.4 27.3 27.2 1.04 (0.99 to 1.08) 27.5 27.4

Mean hemoglobin, g/dl (SD) 10.0 (1.6) 10.4 (1.5) — 10.1 (1.6) 10.3 (1.5),11 66.9 61.6 0.74 (0.71 to 0.77) 65.2 65.111.0–12.0 15.6 18.7 Reference 16.6 16.8.12.0 9.3 11.4 1.03 (0.98 to 1.09) 10.0 9.9

372 Journal of the American Society of Nephrology J Am Soc Nephrol 25: 370–379, 2014


(later initiators=24.9 months; early initiators=22.1 months;P,0.001) (Figure 2). Overall, death from a cardiovascular(CV) cause was more common than infectious causes of death(175.7 versus 47.5 per 1000 person-years). Early initiation wasassociated with greater risk of all-cause, CV, and infectiousmortality compared with later initiation (Table 2). Subgroupanalyses of all-cause mortality generally revealed similar find-ings across subgroups (P interaction.0.05 for all) (Figure 3).However, among patients with an arteriovenous fistula (AVF)or arteriovenous graft (AVG) as their dialysis access, there wasno statistically significant difference in survival comparingearly with later initiation (hazard ratio [HR], 1.05; 95% con-fidence interval [95% CI], 0.98 to 1.13).

All-Cause and Cause-Specific HospitalizationsOver two thirds of patients (n=42,912) were hospitalized atleast one time in the follow-up period. Risk of all-cause hos-pitalization was greater for early initiators (HR, 1.03; 95% CI,1.01 to 1.05); risk of infectious hospitalization was also greater(HR, 1.10; 95%CI, 1.07 to 1.13) (Table 2). However, there wasno difference in risk of CV hospitalization between groups(HR, 0.99; 95% CI, 0.96 to 1.01). We analyzed the risk ofCHF hospitalization specifically, which also revealed no dif-ference between groups (HR, 0.98; 95% CI, 0.93 to 1.03).Results for all-cause hospitalization in the subgroups tested

showed no difference or greater risk of hospitalization amongearly initiators (Supplemental Appendix A). The Andersen–Gill approach with hospitalization-dependent hazard yieldedthe same results.

Sensitivity AnalysesWe conductedmultiple sensitivity analyses to test our findings.In an analysis in which we did not include variables reflectingpatients’ predialysis health status (i.e., AKI, uremia, hyperka-lemia, CHF admissions, and all-cause hospital days), findingswere very similar to our primary analysis, although they yieldeda slightly greater association between early initiation and mor-tality. For example, our sensitivity analysis, which did not in-clude the predialysis variables, yielded a relative HR of all-causemortality among those patients initiating dialysis early versuslater of 1.12 (95% CI, 1.09 to 1.15), which was slightly greaterin magnitude than our primary findings (HR, 1.11; 95% CI,1.08 to 1.14). We observed similar findings for CV disease mor-tality (HR, 1.15; 95%CI, 1.11 to 1.20) for analyses not includingthe predialysis variables compared with our primary analyses,which yielded anHRof 1.13 (95%CI, 1.09 to 1.17). In a secondanalysis restricted to hemodialysis patients, findings were sim-ilar to our primary analysis (HR for all-cause mortality, 1.11;95% CI, 1.08 to 1.14; HR for all-cause hospitalization, 1.03;95% CI, 1.01 to 1.05). A third analysis, in which we included

Table 1. Continued

Patient Characteristics

Before Match, % (n=84,654) Logistic Regression ofPredictors of Early

Initiationa Odds Ratio(95% CI)

After Match, %b (n=61,930)


per 1.73 m2)


per 1.73 m2)


per 1.73 m2)


per 1.73 m2)

BMI, kg/m2 (SD) 27.1 (6.5) 27.5 (6.8) — 27.2 (6.6) 27.4 (6.9),18.5 4.6 4.2 1.10 (1.03 to 1.19) 4.4 4.418.5–25 37.6 36.1 Reference 36.8 36.625–30 30.9 30.0 0.95 (0.92 to 0.99) 30.6 30.6.30 25.8 28.4 1.12 (1.07 to 1.16) 27.0 27.2

Predialysis clinical factorsd c

AKI 61.0 63.9 0.77 (0.74 to 0.80) 60.9 61.0Uremia 36.1 36.5 0.94 (0.91 to 0.96) 36.0 36.0Hyperkalemia 32.0 32.0 0.87 (0.84 to 0.90) 31.7 31.8CHF hospital admissionsNone 65.2 50.4 Reference 61.3 61.21 22.2 25.1 1.07 (1.03 to 1.12) 24.3 24.21–2 7.8 12.8 1.33 (1.25 to 1.41) 8.9 9.0.2 4.8 11.7 1.68 (1.57 to 1.81) 5.6 5.6

Total hospital days0 55.4 43.2 Reference 52.5 52.3,6 18.1 18.1 1.12 (1.07 to 1.17) 18.5 18.46–15 14.9 18.4 1.25 (1.19 to 1.31) 16.0 16.1.15 11.6 20.4 1.52 (1.44 to 1.61) 13.0 13.2

Not all categorical variables sum to 100% because of missing data.aAll other variables are included in each logistic regression model.bP values comparing later and early groups after matching were all .0.05.cReference group is patients without the condition.dPredialysis clinical factors were each measured in 6 months preceding dialysis initiation.



http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2013050567/-/DCSupplemental

patients who recovered kidney function, yielded findings sim-ilar to our primary analysis (HR for all-cause mortality, 1.10;95% CI, 1.08 to 1.13; and HR for all-cause mortality and all-cause hospitalization, 1.03; 95% CI, 1.01 to 1.05, comparingearly to later initiators). A fourth analysis among patients pre-sumed to be well nourished (i.e., serum albumin$3.5 mg/dl orbody mass index [BMI]$25 kg/m2) and thus, less likely thanthose patients with poorer nutritional status to have their GFRfalsely estimated because of lower creatinine production11

also yielded similar findings to our primary analysis. Amongpatients with serum albumin$3.5 mg/dl (n=17,016), HRsfor all-cause mortality and all-cause hospitalization were1.10 (95% CI, 1.04 to 1.16) and 1.03 (95% CI, 0.99 to1.07) for early versus later initiators, respectively. Amongpatients with a BMI$25 kg/m2 (n=35,737), the HRs forall-cause mortality and all-cause hospitalization were 1.11(95% CI, 1.07 to 1.15) and 1.04 (95% CI, 1.01 to 1.06) forearly versus later initiators, respectively.

Because creatinine-based estimates of GFR could be par-ticularly unstable in the setting of AKI,12 we performed a fifth

analysis to explore whether our findingswere consistent among patients withoutAKI in the days leading up to dialysis initi-ation. In this analysis, we excluded 32,797patients who had documented AKI (denovo or AKI on CKD) in the 30 days upto the date of dialysis initiation (53% ofour matched cohort) and patients withacute tubular necrosis (ATN) as their causeof ESRD (n=1859). Among the remaining28,782 patients, we found similar findingsto our primary analysis (HR, 1.17; 95% CI,1.12 to 1.22 for all-cause mortality; HR,1.02; 95%CI, 0.99 to 1.05 for all-cause hos-pitalizations) comparing early with laterinitiators.

Finally, to confirmthat ourfindingswereconsistent among those patients with ade-quate predialysis nephrology care for ad-vancedCKDandwithout AKI as their causeof ESRD, we performed a sixth analysis andincluded patients who had (1) at least 12

months of outpatient nephrology care before initiation, (2) anESRD cause other than ATN, and (3) an International Classi-fication of Diseases, Ninth Revision, Clinical Modification(ICD-9-CM) code of 585.4 (CKD stage 4) during the 2 yearsbefore initiation. Of 84,654 patients, 27,098 patients met ourcriteria for adequate predialysis care, and 19,554 (72.2%) pa-tients were successfully matched. Similar to our primary anal-ysis, risks for all-cause mortality and all-cause hospitalizationwere greater for early compared with later initiators (HR, 1.12;95% CI, 1.07 to 1.18; HR, 1.04; 95% CI, 1.00 to 1.08, respec-tively).

DISCUSSION

In this national study,we found that early dialysis initiationwascommon among older United States adults and associated withtheir greatermortality and hospitalizations, even after account-ing for their predialysis health status and health care use. Ourfindings were consistent across multiple subgroups, including

Figure 2. Proportion of patients surviving from time of dialysis initiation, among thoseinitiating at eGFR ,10 and $10 ml/min per 1.73 m2.

Table 2. Clinical outcomes by timing of dialysis initiation

OutcomeN for Analysisa Number of Events Incidence Rate per 1000 person-yrs Relative Risk of Early Versus Later

(Referent) Timing of DialysisInitiation HR (95% CI)Early Later Early Later Early Later

All-cause mortality 30,965 30,965 12,999 12,152 402.2 360.6 1.11 (1.08 to 1.14)CV mortality 30,965 30,965 6049 5551 187.2 164.7 1.13 (1.09 to 1.17)Infectious mortality 30,965 30,965 1637 1496 50.7 44.4 1.13 (1.06 to 1.22)

All-cause hospitalization 30,731 30,771 62,967 63,857 2079.8 2017.8 1.03 (1.01 to 1.05)CV hospitalization 30,731 30,771 19,512 20,552 623.7 629.9 0.99 (0.96 to 1.01)Infectious hospitalization 30,731 30,771 16,829 15,911 538.5 487.5 1.10 (1.07 to 1.13)

aN for analysis for hospitalizations is unbalanced between early and later initiators because of patients being followed from day 8 after dialysis initiation for hos-pitalization analyses.



those patients presumed to have had adequate predialysiscare and those patients without AKI in the month precedingdialysis initiation.

To our knowledge, no previous study has examined out-comes associatedwith early versus later dialysis initiation amongolder adults.7 Previous observational studies in other popula-tions have shown conflicting findings, with the majority ofstudies showing greater mortality risk with early dialysis initi-ation8,13–25 and some studies showing no difference26 or im-proved survival for earlier initiators.27–29 These studies haveincorporated limited or no information on critical predialysisfactors, which could heavily influence the timing of dialysisinitiation and morbidity and mortality after dialysis, includingnephrology care, comorbidity, and common reasons for earlyinitiation (e.g., AKI, frequent hospitalization, and CHF exac-erbations). These factors are especially important for olderadults, because they suffer a greater burden of comorbid illnesscompared with their younger counterparts,30 often progress toESRD after episodes of AKI,31 and may be more likely thanyounger persons to initiate dialysis under suboptimal circum-stances (e.g., emergent inpatient settings).32 In our study, weidentified previously unreported independent risk factors forearly dialysis initiation (CHF hospital admissions and all-causehospitalization days in the 6 months preceding initiation) andfound that the consideration of these factors and other

predialysis factors served to attenuate, at least somewhat, therisk of poor outcomes associated with early initiation.

The consistencyof ourfindings across numerous subgroupsshould be considered in conjunction with the recent study byBao et al.8 reporting that frailty was associated with early ini-tiation and attenuated the relation of early dialysis initiationand mortality among a subset of 1576 USRDS registrants (ad-justedHR, 1.08; 95%CI, 0.98 to 1.19 per 5ml/min per 1.73m2).In our study, we found that nursing home residence was in-dependently associatedwith early initiation among older adultsbut that mortality was statistically significantly greater for earlyinitiators regardless of nursing home or functional status, likelyowing to our larger sample size, older adult population, andadministrative ascertainment of our measures of functionallimitation.However, whenwe examined risk of hospitalization,there was no difference between early and later initiators eitherresiding in a nursing home or with functional limitations (Sup-plemental Appendix A). Considered together, our study andthe study by Bao et al.8 suggest that early dialysis initiation doesnot benefit patients with functional limitations. Our findingsamong vascular access subgroups also deserve comment. Wefound no statistical difference in survival or hospitalizationsamong patients with an AVF or AVG (less than 25% of pa-tients). This finding could be attributed to several potentialreasons, including these patients’ greater likelihood to have

Figure 3. Hazard of mortality was statistically significantly greater for early as compared to later dialysis initiators across all patientsubgroups except those patients with an AV fistula or graft. AV, arteriovenous.





received predialysis care, their potential better health, which isreflected by vasculature able to support access creation, or smallnumbers of patients in this group, yielding too few patients toobserve statistical differences in clinical outcomes. Patients withworse or unstable health might be less likely to obtain early vas-cular access and start dialysis earlier because of factors that wemay not have been able to completely account for in our analysis.

Our examination of the association of timing of dialysisinitiation with cause-specific morbidity and mortality under-scoresmechanisms throughwhich early dialysis initiationmaybe harmful. Patients with preserved urine output at 1 year, forexample, have greater survival and less inflammation com-pared with patients with anuria.33,34 Because over 50% of en-dogenous renal function may be lost in the first few months ofhemodialysis treatment,35 patients initiated early mayexperience a more rapid change in their inflammatory staterelative to those patients initiated later who have lost residualfunction more gradually, which could increase early initiators’risk of morbidity and mortality. Additionally, this rapid loss ofresidual renal function could make early initiators more sus-ceptible to deleterious electrolyte shifts during dialysis, whichmight explain our finding that risk of CVmortality (e.g., suddencardiac death36) was especially high among this population.

A minority of patients remained unmatched after ourpropensity analysis. They likely represent older US adults whodo not have a choice regarding timing of dialysis initiationbecause of comorbid disease severity. Our study findings are,therefore, least applicable to this population. In contrast, ourfinding of deleterious outcomes associated with early dialysisinitiation when we analyzed patients more likely to have had achoice (e.g., those patients with predialysis nephrology care andseparately, those patients without AKI) may have importantclinical implications. Our findings are most applicable to thesepatients who may benefit from improved efforts to engage inshared decisionmaking about dialysis initiation,37 becausemanyolder adults with kidney failure may choose to forgo dialysisaltogether.38

Our study has limitations. First, our approach did notaddress lead time bias (i.e., the erroneous observation that ear-lier dialysis improves survival, because patients beginning di-alysis at a higher eGFR entered our analysis earlier than thosepatients beginning later). Survivor bias is also possible (i.e.,early initiators may not have been able to survive to becomelater initiators). Future studies that anchor patient recruitmentand analyses at a common eGFR well above the level that pa-tients are typically initiated on dialysis are urgently needed.Second, we used eGFR to define timing of dialysis initiation,which is not the only measure used by nephrologists.39–41 Also,creatinine-based measures of eGFR may misclassify patientswith poor nutritional status and/or sarcopenia,11,42 which isoften seen among older dialysis patients.43 Furthermore, cre-atinine-based measures of eGFR have not been well studiedamong Asian Americans,44,45 which could have influencedour results. In an attempt to account for these limitations, wecarefully matched patients on numerous predialysis factors,

markers of nutritional status, and comorbidities. We alsoconducted a sensitivity analysis among patients presumed tobewell nourished and found results consistent with our primaryanalysis, similar to a prior study.21 Third, although our analysisincluded propensity matching on 28 variables, residual con-founding remains plausible in our study, such as in other obser-vational studies. Fourth, despite our sensitivity analysis excludingpatients with AKI, we lacked a precise measure of the circum-stances surrounding dialysis initiation (i.e., elective versus urgentinitiation46). Fifth, we did not assess patient-reported out-comes such as quality of life, which is an area worthy of futurestudy.47 The limitations of our study are balanced by our in-clusion of all older adult incident dialysis patients in the UnitedStates, ascertainment of predialysis health status using claimsdata, and cause-specificmortality assessment through a linkageto the National Death Index (NDI; the gold standard of USmortality databases48).

In conclusion, early dialysis initiation among US olderadults is common and associated with greater mortality andhospitalizations comparedwith later initiation.This associationwas present even after rigorous adjustment for patients’ pre-dialysis health status and health care use, and it was consistentacross multiple subgroups. These findings do not support thecommon practice of early dialysis initiation among older adultsin the United States.

CONCISE METHODS

OverviewAspart of the Agency forHealthcareResearch andQualityDeveloping

Evidence to Inform Decisions about Effectiveness Network Patient

Outcomes in ESRD Study,49 we performed a national observational

study to compare outcomes related to early versus later dialysis ini-

tiation among older adults in the United States.

Data SourcesWe used the USRDS ESRD database, a national registry of patients

receiving renal replacement therapy. It includes information from

the ESRD Medical Evidence Form of the Centers for Medicare and

Medicaid Services (CMS; form CMS-2728), which includes de-

mographics, serum creatinine, cause of ESRD, and comorbidities. It

also includes the ESRD Death Notification Form (CMS-2746), indi-

cating date and cause of death, as well asMedicare Part A institutional

claims andMedicare Part B physician/supplier claims.10 For patients

with Medicare as their primary insurer at dialysis initiation, USRDS

also maintains their Parts A and B claims from up to 3 years before

dialysis initiation.10 In this study, we also linked USRDS data to NDI

data from the Centers for Disease Control and Prevention.48

Study PopulationWe included 84,654 US dialysis patients ages 67 years or older who

initiated peritoneal dialysis or hemodialysis (in-center and home)

between January 1, 2006 andDecember 31, 2008 andhadbothMedicare

Parts A and B coverage as their primary insurer at least 2 years before



dialysis initiation. We excluded patients without a serum creatinine

recorded on form CMS-2728 or with renal function that recovered

within90days afterdialysis initiation. Follow-upwas throughDecember

31, 2008.

Assessment of Predialysis Health Status and HealthCare UseWe used predialysis Medicare claims for up to 2 years before dialysis

initiation and form CMS-2728 to determine factors related to dialysis

timing. Factors considered included (1) diagnosis claims for AKI,

uremia, or hyperkalemia in 6 months before dialysis initiation (Sup-

plemental Appendix B), (2) number of hospitalizations for CHF and

total all-cause hospital days in 6 months before dialysis initiation

(Supplemental Appendix C), and (3) history of outpatient nephrol-

ogy visits and codes for CKD stage 4 (ICD-9-CM code 585.4; as an

indicator of longitudinal nephrology care) in 2 years preceding initia-

tion. Patient comorbidities (CHF, atherosclerotic heart disease, cerebro-

vascular accident/transient ischemic attack, peripheral vascular disease,

dysrhythmia, other cardiac disease, chronic obstructive pulmonary dis-

ease, diabetes, cancer, gastrointestinal bleeding, and liver disease) were

defined based on formCMS-2728 and 2 years of predialysis claims using

previously described methods (Supplemental Appendix B).50 Nursing

home residency status and presence of functional limitations (which

include the inability to ambulate or transfer and/or a need for assistance

with daily activities) were also ascertained from form CMS-2728.

Timing of Dialysis InitiationWe calculated each patient’s eGFR at dialysis initiation using serum

creatinine from form CMS-2728 and the four-variable Modification

of Diet in Renal Disease Study equation.51 We defined patients’ ini-

tiation as early if eGFRwas$10ml/min per 1.73m2 and later if eGFR

was ,10 ml/min per 1.73 m2.

OutcomesOur primary outcomes were all-cause mortality and all-cause hos-

pitalization. SecondaryoutcomeswereCVdeath, infectiousdeath,CV

hospitalizations, and infectious hospitalizations. We derived

patients’ dates and causes of death from NDI data, which were avail-

able on 99.3% of deaths (death data from form CMS-2746 was used

for the remainder). We used ICD-9-CM diagnosis codes from Medi-

care Part A claims to define cause-specific hospitalizations and Inter-

national Classification of Diseases, Tenth Revision diagnosis codes to

define NDI causes of death (Supplemental Appendix C).

Formortality analyses, patientswere followed fromdialysis initiation

todeath or censored at the time of kidney transplantation, loss to follow-

up, renal function recovery, or end of study follow-up (December 31,

2008). To exclude hospitalizations directly related to inpatient dialysis

initiation, patients were followed from day 8 after dialysis initiation for

analyses of hospitalizations and additionally censored fordeath or endof

Medicare as the primary insurance payer.

Statistical AnalysesPropensity Matching.We anticipated that patient characteristics would vary between early

and later initiation groups. Therefore, we matched the groups based

on their propensity for early initiation calculated from a logistic re-

gression model including 28 predialysis variables (demographics, renal

complication history, comorbidities, predialysis AKI, hospitalizations,

etc.) as predictors. We then used a greedy match algorithm52 to match

an early initiation patient to a later initiation patient with a 1:1 ratio

and an initial requirement of a propensity difference less than 0.00001.

If no match was found, we then released the requirement of 0.00001–

0.0001 and so on until the patient was matched or could not be

matched with a requirement of 0.1. We compared patient character-

istics before and after the match using chi-squared tests.

Outcome Analyses.After establishing the propensity-matched cohort, we compared

outcomes in the early versus later initiation groups using the Kaplan–

Meier method and Cox proportional hazards regression models. Be-

cause patients’ characteristics were well balanced through propensity

matching, we did not adjust these models.53 Proportionality was

checked using Schoenfeld residual plots.54 For hospitalization ana-

lyses, which included repeated events, a Cox proportional hazards

regression model with the Andersen–Gill approach55 was used. In

case the previous hospitalization changed the probability of the next

hospitalization, as a sensitivity analysis, different baseline hazards

were allowed for different hospitalizations (i.e., the 1st, the 2nd,

etc.) up to the 10th hospitalization. For all analyses, robust standard

error calculations were performed taking into account potential clus-

tering by dialysis center.

We performed subgroup analyses with tests for interaction for the

primary end points of all-cause mortality and hospitalization to

examine thepossibilityof effectmodification. Subgroupsweredefined

by diabetes status, race (because of reports of survival differences by

race56,57), dialysis modality, hemodialysis access (catheter or AVF/

AVG), nursing home residence, and functional limitation status. Be-

cause patient characteristics among the early and later groups were

no longer balanced within some of the subgroups, the analyses were

adjusted for all other characteristics.

Sensitivity AnalysesWe performed sensitivity analyses to test our findings. First, we

conducted an analysis without the inclusion of predialysis health

status variables. Second, we restricted our analysis to hemodialysis

patients and reperformed our propensity match. Third, we repeated

our analysis including patientswho recovered renal functionwithin90

days after dialysis initiation. Fourth, we performed two analyses, where

werestrictedourpropensity-matchedcohort to thosepatientspresumed

to bewell nourished (either serum albumin$3.5mg/dl or BMI$25 kg/

m2). Fifth, we conducted an analysis restricted to propensity-matched

patients without AKI in the 30 days leading up to dialysis initiation and

without ATNas their cause of ESRD. Sixth, to confirm that our findings

were consistent among those patients with adequate predialysis ne-

phrology care for advanced CKD, we performed an analysis including

patients who had (1) at least 12 months of outpatient nephrology care

before initiation, (2) an ESRD cause other thanATN, and (3) an ICD-9-

CMcode of 585.4 (CKD stage 4) during the 2 years before initiation.We

then conducted the propensity-matching procedure to establish a

matched cohort in this subset and repeated our primary models.








ACKNOWLEDGMENTS

D.C.C. was supported by the Amos Medical Faculty Development

Program of the Robert Wood Johnson Foundation. J.J.S. was supported

by theNational InstitutesofHealth(NIH)National Institute forDiabetes,

Digestive and Kidney Diseases (Grant K23 DK095949). T.S. was sup-

ported by the NIHNational Institute for Diabetes, Digestive and Kidney

Diseases (Grant K23-DK-083514). The Developing Evidence to Inform

Decisions about Effectiveness (DEcIDE) Network Patient Outcomes in

ESRD Study was supported by Agency for Healthcare Research and

Quality (AHRQ) Contract HHSA290200500341I (Task Order #6).

Portions of this manuscript were presented as abstracts at the 2011

American Society of Nephrology Kidney Week in Philadelphia, PA,

November 8–13, and at the 2012 American Society of Nephrology

Kidney Week in San Diego, CA, October 31–November 4.

Funding sources for this manuscript and its authors had no role in the

design and conduct of the study; collection, management, analysis, or

interpretation of the data; preparation, review, or approval of the man-

uscript; or the decision to submit it for publication. Identifiable infor-

mation on which this report, presentation, or other form of disclosure is

based is confidential and protected by federal law, Section 903(c) of the

PublicHealthServiceAct, 42USC299a-1(c).Any identifiable information

that is knowingly disclosed is disclosed solely for the purpose for which

it has been supplied. No identifiable information about any individual

supplying the information or described in it will be knowingly disclosed

except with the prior consent of that individual. The data reported here

have been supplied by the US Renal Data System. The interpretation and

reporting of these data are the responsibility of the authors and in no way

shouldbeseenasanofficialpolicyor interpretationof theUSGovernment.

The DEcIDE Network Patient Outcomes in End-Stage Renal

Disease Study Team consists of members from the Johns Hopkins

University (D.C.C.,K.B.-R., P.L.E., B.G.J., S.M.S., T.S.,A.W.W., L.E.B.,

Courtney Cook, Josef Coresh, Jeonyong Kim, Yang Liu, Jason Luly,

Aidan McDermott, Wieneke Michels, Paul Scheel, and Jing Zhou),

University of California at San Francisco (N.R.P.), Chronic Disease

Research Group (J.L., H.G., Allan Collins, Robert Foley, David Gil-

bertson, Charles Herzog, and Wendy St. Peter), Cleveland Clinic

Foundation (Joseph Nally, Susana Arrigain, Stacey Jolly, Vicky Konig,

XiaoboLiu, SankarNavaneethan, and JesseSchold),UniversityofNew

Mexico (Philip Zager), Tufts University (D.C.M. and K.B.M.), Uni-

versity ofMiami (J.J.S.), University ofManitoba (N.T.), andAcademic

Medical Center, The Netherlands (Wieneke Michels).

DISCLOSURESNone.

REFERENCES

1. O’Hare AM, Choi AI, Boscardin WJ, Clinton WL, Zawadzki I, Hebert PL,Kurella Tamura M, Taylor L, Larson EB: Trends in timing of initiation ofchronic dialysis in theUnited States.Arch InternMed 171: 1663–1669, 2011

2. National Kidney Foundation: NKF-DOQI clinical practice guidelines forperitoneal dialysis adequacy. National Kidney Foundation. Am J Kid-

ney Dis 30[3 Suppl 2]: S67–S136, 1997

3. Tattersall J, Dekker F, Heimburger O, Jager KJ, Lameire N, Lindley E,Van Biesen W, Vanholder R, Zoccali C; ERBP Advisory Board: When tostart dialysis: Updated guidance following publication of the InitiatingDialysis Early and Late (IDEAL) study. Nephrol Dial Transplant 26:2082–2086, 2011

4. Ledebo I, Kessler M, van Biesen W, Wanner C, Wiecek A, Prichard S,Argilés A, Ritz E: Initiationof dialysis-opinions froman international survey:Report on the Dialysis Opinion Symposium at the ERA-EDTA Congress,18 September, 2000, Nice.Nephrol Dial Transplant 16: 1132–1138, 2001

5. Kurella Tamura M, Covinsky KE, Chertow GM, Yaffe K, Landefeld CS,McCulloch CE: Functional status of elderly adults before and after ini-tiation of dialysis. N Engl J Med 361: 1539–1547, 2009

6. Rosansky SJ, Clark WF, Eggers P, Glassock RJ: Initiation of dialysis athigher GFRs: Is the apparent rising tide of early dialysis harmful orhelpful? Kidney Int 76: 257–261, 2009

7. Susantitaphong P, Altamimi S, Ashkar M, Balk EM, Stel VS, Wright S,Jaber BL: GFR at initiation of dialysis and mortality in CKD: A meta-analysis. Am J Kidney Dis 59: 829–840, 2012

8. Bao Y, Dalrymple L, Chertow GM, Kaysen GA, Johansen KL: Frailty,dialysis initiation, and mortality in end-stage renal disease. Arch Intern

Med 172: 1071–1077, 20129. Cooper BA, Branley P, Bulfone L, Collins JF, Craig JC, Fraenkel MB,

Harris A, Johnson DW, Kesselhut J, Li JJ, Luxton G, Pilmore A, Tiller DJ,Harris DC, Pollock CA; IDEAL Study: A randomized, controlled trial ofearly versus late initiation of dialysis.N Engl J Med 363: 609–619, 2010

10. US Renal Data System:USRDS 2011Annual Data Report: Atlas of End-StageRenal Disease in the United States, Bethesda, MD, National Institutes ofHealth,National InstituteofDiabetes andDigestiveandKidneyDisease, 2011

11. Beddhu S, SamoreMH, RobertsMS, StoddardGJ, Pappas LM, CheungAK: Creatinine production, nutrition, and glomerular filtration rate es-timation. J Am Soc Nephrol 14: 1000–1005, 2003

12. Bouchard J, Macedo E, Soroko S, Chertow GM, Himmelfarb J, Ikizler TA,Paganini EP, Mehta RL; Program to Improve Care in Acute Renal Disease:Comparison ofmethods for estimating glomerular filtration rate in critically illpatientswith acute kidney injury.NephrolDial Transplant25: 102–107, 2010

13. Beddhu S, Samore MH, Roberts MS, Stoddard GJ, Ramkumar N,Pappas LM, Cheung AK: Impact of timing of initiation of dialysis onmortality. J Am Soc Nephrol 14: 2305–2312, 2003

14. Kazmi WH, Gilbertson DT, Obrador GT, Guo H, Pereira BJ, Collins AJ,Kausz AT: Effect of comorbidity on the increased mortality associatedwith early initiation of dialysis. Am J Kidney Dis 46: 887–896, 2005

15. Shiao CC, Huang JW, Chien KL, Chuang HF, Chen YM, Wu KD: Early initi-ationof dialysis and late implantationof catheters adversely affect outcomesof patients on chronic peritoneal dialysis. Perit Dial Int 28: 73–81, 2008

16. Sawhney S, Djurdjev O, Simpson K, Macleod A, Levin A: Survival anddialysis initiation: Comparing British Columbia and Scotland registries.Nephrol Dial Transplant 24: 3186–3192, 2009

17. Stel VS, Dekker FW,Ansell D, AugustijnH, CasinoFG,Collart F, Finne P,Ioannidis GA, SalomoneM, Traynor JP, Zurriaga O, Verrina E, Jager KJ:Residual renal function at the start of dialysis and clinical outcomes.Nephrol Dial Transplant 24: 3175–3182, 2009

18. LassalleM, LabeeuwM, Frimat L, Villar E, Joyeux V, CouchoudC, StengelB: Age and comorbidity may explain the paradoxical association of anearly dialysis start with poor survival. Kidney Int 77: 700–707, 2010

19. Hwang SJ, YangWC, Lin MY, Mau LW, Chen HC: Impact of the clinical con-ditions at dialysis initiation on mortality in incident haemodialysis patients: Anational cohort study inTaiwan.NephrolDialTransplant25:2616–2624,2010

20. Wright S, Klausner D, Baird B,WilliamsME, Steinman T, TangH, RagasaR, Goldfarb-Rumyantzev AS: Timing of dialysis initiation and survival inESRD. Clin J Am Soc Nephrol 5: 1828–1835, 2010

21. Rosansky SJ, Eggers P, Jackson K, Glassock R, Clark WF: Early start ofhemodialysis may be harmful. Arch Intern Med 171: 396–403, 2011

22. ClarkWF,NaY,RosanskySJ, SontropJM,MacnabJJ,GlassockRJ, EggersPW, Jackson K, Moist L: Association between estimated glomerular fil-tration rate at initiation of dialysis and mortality. CMAJ 183: 47–53, 2011



23. EvansM, Tettamanti G,NyrénO, Bellocco R, ForedCM, Elinder CG:Nosurvival benefit from early-start dialysis in a population-based, in-ception cohort study of Swedish patients with chronic kidney disease.J Intern Med 269: 289–298, 2011

24. Wilson B, Harwood L, Locking-Cusolito H, Chen SJ, Heidenheim P,Craik D, ClarkWF: Optimal timing of initiation of chronic hemodialysis?Hemodial Int 11: 263–269, 2007

25. Molnar MZ, Streja E, Kovesdy CP, Hoshino J, Hatamizadeh P, GlassockRJ, Ojo AO, Kalantar-Zadeh K: Estimated glomerular filtration rate atreinitiation of dialysis and mortality in failed kidney transplant recipi-ents. Nephrol Dial Transplant 27: 2913–2921, 2012

26. Traynor JP, Simpson K, Geddes CC, Deighan CJ, Fox JG: Early initia-tion of dialysis fails to prolong survival in patients with end-stage renalfailure. J Am Soc Nephrol 13: 2125–2132, 2002

27. Korevaar JC, Jansen MA, Dekker FW, Jager KJ, Boeschoten EW,Krediet RT, Bossuyt PM; Netherlands Cooperative Study on the Ade-quacy of Dialysis Study Group: When to initiate dialysis: Effect of pro-posed US guidelines on survival. Lancet 358: 1046–1050, 2001

28. Tang SC,HoYW, TangAW,ChengYY, Chiu FH, LoWK, Lai KN;HongKongPeritonealDialysisStudyGroup:Delaying initiationofdialysis till symptomaticuraemia—is it too late?Nephrol Dial Transplant 22: 1926–1932, 2007

29. Coronel F, Cigarran S, Herrero JA: Early initiation of peritoneal dialysisin diabetic patients. Scand J Urol Nephrol 43: 148–153, 2009

30. Liu J, Huang Z, Gilbertson DT, Foley RN, Collins AJ: An improved co-morbidity index for outcome analyses among dialysis patients. KidneyInt 77: 141–151, 2010

31. Ishani A, Xue JL, Himmelfarb J, Eggers PW, Kimmel PL, Molitoris BA,Collins AJ: Acute kidney injury increases risk of ESRD among elderly.J Am Soc Nephrol 20: 223–228, 2009

32. MendelssohnDC, Curtis B, Yeates K, Langlois S,MacRae JM, SemeniukLM, Camacho F,McFarlane P; STARRT Study Investigators: Suboptimalinitiation of dialysis with and without early referral to a nephrologist.Nephrol Dial Transplant 26: 2959–2965, 2011

33. Shafi T, Jaar BG, Plantinga LC, Fink NE, Sadler JH, Parekh RS, PoweNR,Coresh J: Association of residual urine output with mortality, quality oflife, and inflammation in incident hemodialysis patients: The Choicesfor Healthy Outcomes in Caring for End-Stage Renal Disease (CHOICE)Study. Am J Kidney Dis 56: 348–358, 2010

34. van der Wal WM, Noordzij M, Dekker FW, Boeschoten EW, Krediet RT,Korevaar JC, Geskus RB; Netherlands Cooperative Study on the Ade-quacy of Dialysis Study Group (NECOSAD): Full loss of residual renalfunction causes higher mortality in dialysis patients; findings from a mar-ginal structural model. Nephrol Dial Transplant 26: 2978–2983, 2011

35. Jansen MA, Hart AA, Korevaar JC, Dekker FW, Boeschoten EW, KredietRT; NECOSAD Study Group: Predictors of the rate of decline of residualrenal function in incident dialysis patients.Kidney Int62: 1046–1053, 2002

36. Alves TP, Wang X, Wright JT Jr., Appel LJ, Greene T, Norris K, Lewis J;AASK Collaborative Research Group: Rate of ESRD exceeds mortalityamong African Americans with hypertensive nephrosclerosis. J Am SocNephrol 21: 1361–1369, 2010

37. Schell JO, Germain MJ, Finkelstein FO, Tulsky JA, Cohen LM: An in-tegrative approach to advanced kidney disease in the elderly. AdvChronic Kidney Dis 17: 368–377, 2010

38. Hemmelgarn BR, JamesMT, Manns BJ, O’Hare AM, Muntner P, RavaniP, Quinn RR, Turin TC, Tan Z, Tonelli M; Alberta Kidney Disease Net-work: Rates of treated and untreated kidney failure in older vs youngeradults. JAMA 307: 2507–2515, 2012

39. Weiner DE, Stevens LA: Timing hemodialysis initiation: A call for clinicaljudgment. Am J Kidney Dis 57: 562–565, 2011

40. Lameire N, Van Biesen W: The initiation of renal-replacement therapy—just-in-time delivery. N Engl J Med 363: 678–680, 2010

41. Kurella Tamura M, O’Hare AM, McCulloch CE, Johansen KL: Signs andsymptoms associated with earlier dialysis initiation in nursing homeresidents. Am J Kidney Dis 56: 1117–1126, 2010

42. Grootendorst DC, Michels WM, Richardson JD, Jager KJ, BoeschotenEW, Dekker FW, Krediet RT; NECOSAD Study Group: The MDRD for-mula does not reflect GFR in ESRD patients. Nephrol Dial Transplant26: 1932–1937, 2011

43. Ohkawa S, Odamaki M, Ikegaya N, Hibi I, Miyaji K, Kumagai H: Asso-ciation of age with muscle mass, fat mass and fat distribution in non-diabetic haemodialysis patients.Nephrol Dial Transplant 20: 945–951,2005

44. Stevens LA, Claybon MA, Schmid CH, Chen J, Horio M, Imai E, NelsonRG, Van Deventer M, Wang HY, Zuo L, Zhang YL, Levey AS: Evaluationof the Chronic Kidney Disease Epidemiology Collaboration equationfor estimating the glomerular filtration rate in multiple ethnicities.Kidney Int 79: 555–562, 2011

45. Gerchman F, Tong J, Utzschneider KM, Hull RL, Zraika S, Udayasankar J,McNeely MJ, Andress DL, Leonetti DL, Boyko EJ, FujimotoWY, Kahn SE:Superiority of theModification of Diet in Renal Disease equation over theCockcroft-Gault equation in screening for impaired kidney function inJapanese Americans. Diabetes Res Clin Pract 77: 320–326, 2007

46. Crews DC, Jaar BG, Plantinga LC, Kassem HS, Fink NE, Powe NR: In-patient hemodialysis initiation: Reasons, risk factors and outcomes.Nephron Clin Pract 114: c19–c28, 2010

47. Johansen KL: Time to rethink the timing of dialysis initiation.Arch InternMed 171: 382–383, 2011

48. Cowper DC, Kubal JD,Maynard C, Hynes DM: A primer and comparativereviewofmajorUSmortalitydatabases.AnnEpidemiol12: 462–468,2002

49. Boulware LE, Tangri N, Ephraim PL, Scialla JJ, Sozio SM, Crews DC,Shafi T, Miskulin DC, Liu J, St Peter W, Jaar BG, Wu AW, Powe NR,Navaneethan SD, Bandeen-Roche K; DEcIDE ESRD Patient Outcomesin Renal Disease Study Investigators: Comparative effectiveness stud-ies to improve clinical outcomes in end stage renal disease: TheDEcIDEpatient outcomes in end stage renal disease study. BMC Nephrol 13:167, 2012

50. Hebert PL, Geiss LS, Tierney EF, EngelgauMM, Yawn BP, McBean AM:Identifying persons with diabetes using Medicare claims data. Am JMed Qual 14: 270–277, 1999

51. Levey AS, Bosch JP, Lewis JB, Greene T, Rogers N, Roth D; Modifica-tion of Diet in Renal Disease Study Group: A more accurate method toestimate glomerular filtration rate from serum creatinine: A new pre-diction equation. Ann Intern Med 130: 461–470, 1999

52. Parsons LS: Reducing Bias in a Propensity Score Matched-Pair SampleUsing Greedy Matching Techniques, Cary, NC, SAS Institute, 2001

53. D’Agostino RB Jr.: Propensity scores in cardiovascular research. Cir-culation 115: 2340–2343, 2007

54. Grambsch PM, Therneau TM: Proportional hazards tests and diag-nostics based on weighted residuals. Biometrika 81: 515–526, 1994

55. Andersen PK, Gill RD: Cox regression model for counting processes—a large sample study. Ann Stat 10: 1100–1120, 1982

56. Kucirka LM, Grams ME, Lessler J, Hall EC, James N, Massie AB,Montgomery RA, Segev DL: Association of race and age with survivalamong patients undergoing dialysis. JAMA 306: 620–626, 2011

57. Crews DC, Sozio SM, Liu Y, Coresh J, Powe NR: Inflammation and theparadox of racial differences in dialysis survival. J Am Soc Nephrol 22:2279–2286, 2011

This article contains supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2013050567/-/DCSupplemental.





Predialysis Health, Dialysis Timing, and Outcomes among Older United States Adults

Documents