ESRD provide
rs
192 Introduction 19
4 Dif
erences
in pro
vider
growth
{dia
lysis
trea
tmen
ts} {u
nit lo
catio
n &
gro
wth
} {af
filia
tion
& p
rofit
stat
us}
196 Patient characteris
tics b
y unit aY
liatio
n {patie
nt dem
ogra
phics &
clin
ical
par
amet
ers }
{pre
vale
nt p
atie
nts,
diag
nose
s, &
mod
alit
y}
198 Provider compliance with K/DOQI g
uidelines
{anem
ia tr
eatm
ent}
{vasc
ular a
cces
s} {s
erum
alb
umin
, cre
atin
ine,
& ch
oles
tero
l} {p
atie
nt &
staf
f vac
cina
tion
s}
200 Reuse practices {trends} {re
use by ESRD network
} {ger
micides
& ESRD net
works}
{clin
ical p
aram
eter
s by r
euse
germ
icid
e }
20
2 Su
mm
ary
CHAPTER
Rossettih
Does the road wind up-hill all the way?Yes, to the very end.Will the day’s journey take the whole long day?From morn to night, my friend....Shall I meet other wayfarers at night?Those who have gone before.Then must I knock, or call when just in sight?They will not keep you standing at that door.
Shall I find comfort, travel-sore and weak?Of labour you shall Ind the sum.Will there be beds for me and all who seek?Yea, beds for all who come.
“Uphill”
Christina
192 ESRD providersh
xpansion of the ESRD program, particularly of the dialysis population,
has required providers to keep pace by increasing the number of units in
which patients receive services. Reimbursement for dialysis services, how-
ever, is capitated on a per treatment basis, with provisions for staff and supplies, overhead,
and ancillary support services. This capitation places continued pressure on providers to
reduce costs, particularly since rates of payment have changed little since their implementa-
tion in 1982. The consolidation of dialysis providers into large chains, and the transition of
these providers from non-profit to for-profit status, are both consistent with the need for
increased capital to build dialysis units; it has become increasingly difficult for non-profit
and independent units to thrive.
In Chapter Twelve we provide detailed information on the costs associated with dialysis
services, while in this chapter we present data on recent trends in the dialysis infrastructure
and on how these changes relate to provider profit status and chain status, growth in the
number of dialysis treatments, and differences in the geographic distribution of providers
across ESRD networks, states, and Health Service Areas.
Starting with 2002, facility profit status is now determined from the CMS Dialysis Facil-
ity Compare data, as this information is no longer collected in the CMS Annual Facility
Survey. Overall, the number of total providers increased 6 percent, within the 4–8 percent
annual increase noted since 1996. The total number of hemodialysis treatments grew 27
percent between 1998 and 2002, while the number of peritoneal dialysis treatments fell 21
percent over the same period.
Dialysis chains—defined by the USRDS as corporations owning 20 or more freestanding
units that are located in more than one state—continue to grow, and own the majority of
units in the southern and southeastern states. Non-chain units are more widely dispersed
across the country, particularly in the Upper Midwest and the northeastern states. By ESRD
network, there are dramatic differences in unit ownership; in Network 2 (New York), for
example, only 20 percent of units are owned by chains, compared to 82 percent in Network
8 (Alabama, Mississippi, and Tennessee). Most of the recent growth in the number of both
units and patients has occurred in the large chains—Fresenius, Gambro, DaVita, Renal Care
Group, Dialysis Clinics Inc., and National Nephrology Associates.
Unit density varies considerably, from a mean of 0.83 units per 100,000 people in Califor-
nia and rural areas to 4.2 units per 100,000 people in the southern and central states.
Introducion
90 91 92 93 94 95 96 97 98 99 00 01 02
stin
ufo
reb
muN
0
500
1,000
1,500
2,000
2,500
3,000
3,500Freestanding for-profit
Hospital facility Transplant & dialysis centerTransplant center
Freestanding non-profit Hospital center
Chapt
er hi
ghlig
hts
193H2004 USRDS Annual Data Report
{11.1} Counts of dialysis & transplant units, by CMS certification typedata are obtained from CMS’s annual End-Stage Renal DiseaseFacility Survey, CMS Independent Renal Facility CostReports, & the CMS “Dialysis Facility Compare”website. The leveling out of the number offreestanding, for-profit units in 2002 isdue to the reclassification of unitsin the survey.
The care delivered by providers has been a focus of the
USRDS for the last three years. Clinical results by provider
vary widely. DaVita, for instance, appears to have the great-
est percent of patients with hemoglobins of 12 g/dl and higher,
and epoetin doses that are among the highest as well. Levels
of dialysis therapy, in contrast, as measured by urea reduc-
tion ratio (URR), show that, 86 percent of hemodialysis pa-
tients across the country meet the K/DOQI URR target of
≥65 percent. The distribution of patients by vascular access
is similar across chains, with patients treated by National
Nephrology Associates most likely to have a synthetic graft.
And patients receiving therapy in units owned by Renal Care
Group are considerably more likely to receive a pneumococ-
cal pneumonia vaccination than those in other units.
The CDC has monitored hemodialyzer reuse for many
years through its annual National Surveillance of Dialysis-
Associated Diseases in the United States. Survey data from
2002 were not available in time for this ADR; we present,
then, data from the 2001 survey, which show a decline in the
percent of units reusing dialyzers. Reuse practices vary widely
nationwide. The most striking geographic differences are re-
lated to the practice of using bleach as a cleaning agent. In
Louisiana, for example, two-thirds of dialysis units use bleach
in their reuse process, while in the Great Plains states bleach
is used in less than 3 percent of all units. Interestingly, units
that do not reuse their dialyzers appear to have the lowest
proportion of patients with hemoglobin levels of 12 g/dl and
above; this may be related to the patient population in these
units, and merits further investigation. Generally, there is
little difference by reuse practice or germicide use when look-
ing at hemoglobin levels, epoetin doses, serum albumin lev-
els, and delivered dialysis therapy.
Provider-level information presented here can be exam-
ined in conjunction with data in Chapter Six on hospitaliza-
tion and mortality ratios (pages 134–137), in Chapter Seven
on transplantation ratios (page 145), and in Chapter Twelve
on PMPM costs for clinical services (page 211). After ad-
dressing basic demographic differences, morbidity and mor-
tality ratios are similar across providers. And there are clear
differences in resource utilization between providers, yet out-
comes on a gross scale are similar.
With attention being paid to the K/DOQI guidelines, many
basic elements of dialysis care have improved in recent years.
To improve patient outcomes even further, new and more
detailed assessments may be required. The impact of
diabetes, insulin resistance, and the heavy inflamma-
tory load carried by both the chronic kidney dis-
ease and dialysis populations, for instance, each
merit investigation. And while effective inter-
ventions are yet to be defined, observa-
tional studies support hypotheses that
inflammation may be an impor-
tant area to consider for clini-
cal trials. These efforts
would include current
trials addressing the
efficacy of daily
dialysis.
{11.2} In most
networks the
number of perito-
neal dialysis treat-
ments has fallen or re-
mained relatively stable,
though growth has occurred in
Networks 9, 10, and 18. {11.5} The
number of patients receiving treat-
ment in chain-affiliated units has in-
creased six-fold since 1993, while the number
treated in non-chain units fell after peaking in the
early 1990s, and has remained stable during the past
four years. {11.12} While the dialysis chains, particularly
Fresenius, have grown significantly, there has been little
change in the number of non-chain and hospital-based units.
{11.19} Renal Care Group appears to have the highest rates of vaccina-
tion for influenza and pneumococcal pneumonia, and Gambro among the
lowest rates. {11.21} Dialyzer reuse grew during the 1990s, but since 1997 has
fallen slightly.
194 ESRD providersh
)sn
oillim
ni(st
nemtaert
retnec-
nI
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.01998
2002
ESRD network
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
)sd
nasu
oht
ni(st
nemtaerT
0
5
10
15
20
25
30
35
40
Hemodialysis
Peritoneal dialysis
89 2 0 89 2 0 89 20 89 20 89 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 2 0 8 9 2 0 8 9 20 89 20
s tin
ufot
ne creP
0
20
40
60
80
100
Chain-affiliated
Non-chain
All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
Units addedUnits unchanged
Units dropped
etween 1998 and 2002, the num-
ber of in-center hemodialysis treat-
ments increased 28 percent (Figure
11.2). Growth ranged from 16 percent in
Networks 5 and 8 to nearly 37 percent in
Networks 16 and 17—the northwestern
states and northern California. Changes in
the number of peritoneal dialysis treatments
have been less consistent nationwide.
Figure 11.3 illustrates growth in the
number of chain-affiliated and non-chain
dialysis units. The Mid-Atlantic region and
East Coast have seen the greatest increase
in chain-affiliated units, while growth in
non-chain units is more widespread, with
the highest concentrations occurring in the
northern half of the country.
In approximately half of the ESRD net-
works the proportion of units that are
chain-affiliated has remained relatively
stable, while Networks 8, 9, 10, 12, and 13
saw an increase in that proportion between
1998 and 2002 (Figure 11.4). The greatest
growth—from 16 to 46 percent—occurred
in Network 16. While the proportion of
chain-affiliated units doubled in Network
2 (New York), this network continues to
have the greatest percentage of indepen-
dently owned units, at almost 80 percent.
Figure 11.5 illustrates the parallel growth
of unit and patient counts, as well as thedramatic changes occurring in the compo-
sition of the ESRD program since 1988.
The number of patients receiving treatment
in chain-affiliated units, for instance, has
increased six-fold since 1993, while the
number treated in non-chain units fell af-
ter peaking in the early 1990s, and has re-
mained stable during the past four years.
From 1998 to 2002 overall growth in unit
and patient counts was quite similar, at 21–
24 percent, but there are considerable varia-
tions across the country (Figure 11.6). In
Networks 3, 4, and 10, for example, the in-
crease in the number of units has been far
higher than that in the number of patients.
In the western states of Networks 16, 17,
and 18, in contrast, growth in the number
of patients has far outpaced that of new units.
Three-quarters of ESRD patients are
treated in freestanding, for-profit dialysis
units (Figure 11.7). Growth in the for-profit
sector generally parallels that seen with
chain affiliation (Figure 11.9). In Networks
6, 7, 13, and 14, nearly 90 percent of units
are for-profit, while in Network 2 almost
{11.2} Dialysis treatments, by ESRD network & modality
Diferences in provider growth
Chain-affiliated & non-chain units
{11.3} Unit growth between 1998 & 2000
{11.4} Unit distribution, by ESRD network
Chain-affiliated units Non-chain units
195H2004 USRDS Annual Data Report
All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
0
700
1,400
2,100
2,800
0
50
100
150
200
0
700
1,400
2,100
2,800
3,500
)sd
nasu
oht
ni(st
neitapf
ore
bmu
N
0
50
100
150
200
250Profit status
Chain status
88 90 92 94 96 98 00 02
stin
ufo
reb
muN
0
1,000
2,000
3,000
4,000
0
50
100
150
200
250
300Freestanding/hospital-based
Chain: units
Chain: patients
Profit: units
Profit: patients
FS: units
FS: patients
HB: units
HB: patients
Non-profit: unitsNon-profit: patients
ESRD network
All 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18
2002-8991,eg
nahc
tne cre P
0
10
20
30
40
50
89 20 89 20 89 20 89 2 0 8 9 2 0 89 2 0 89 20 89 20 89 20 89 20 89 20 89 2 0 89 2 0 89 2 0 89 2 0 89 2 0 89 20 8 9 20 89 20
st in u f
otne creP
0
20
40
60
80
100
89 20 89 20 8 9 2 0 8 9 2 0 89 20 8 9 20 8 9 20 8 9 20 8 9 20 8 9 20 89 2 0 89 2 0 89 2 0 89 2 0 89 2 0 89 20 8 9 2 0 89 20 89 20
s t inu f
otn ec re P
0
20
40
60
80
100
96 97 98 99 00 01 02
stneita
pfot
nec reP
0
20
40
60
80
100
2.60+ (4.21)1.89 to <2.601.47 to <1.89 No units
1.09 to <1.47below 1.09 (0.83)
Transplant & dialysis center
Hospital facility
Hospital center
Freestanding non-profit
Freestanding for-profit
Units
Patients
Non-chain:
patients
units
Non-chain:
Unknown
Non-profit
Profit
Hospital-based
Freestanding
{11.5} Unit & patient counts, by unit type {11.6} Percent change in the number of units & patients, 1998 to 2002, by ESRD network
{11.9} Distribution of for-profit & non-profit units, by ESRD network
{11.10} Distribution of freestanding & hospital-based units, by ESRD network
{11.8} Units per 100,000 population, 2002{11.7} Pt distribution, by CMS certification type
70 percent continue to be operated on a
non-profit basis.The number of units available per
100,000 population differs widely across the
country, with a five-fold difference between
upper and lower quintile averages (Figure
11.8). In the country as a whole, 81 percent
of units are freestanding (Figure 11.10). In
the southeastern states of Networks 6, 7, and
8 this number reaches more than 90 per-
cent, while Network 2 (New York) has the
largest proportion of hospital-based units, at
49 percent.
{Figures 11.2–10} data obtained from CMS’s annualEnd-Stage Renal Disease Facility Survey, CMS Inde-pendent Renal Facility Cost Reports, & the CMS “Di-alysis Facility Compare” website. {Figures 11.8} dataalso obtained from estimates of the United States 2002census, based on the 2000 census. x {Figure 11.2}Transient treatments, which account for less than 1percent of all treatments, are not included. Hemodialy-sis includes outpatient hemodialysis & hemodialysistraining treatments; peritoneal dialysis includes out-patient IPD treatments & IPD, CAPD, & CCPD train-ing treatments. {Figure 11.8} 2002, by HSA, unadjusted.
The CDC did not conduct a survey in 1998, & data for2002 were not available as this book went to press. Fig-ure 2.38, in Chapter Two, contains a map of the ESRDnetworks; a list of network contacts can be found onpage 230 of Appendix A.
196 ESRD providersh
)sraey(e
ganae
M
56
58
60
62
64
elamef
tnecreP
40
42
44
46
48
50Age Gender: Female
stneita
pfo
tnecreP
0
20
40
60
80
100
citebai
dt
necr eP
34
36
38
40
42
44
46
Race
Diabetic status: Diabetics
stneita
pfo
tnecreP
0
20
40
60
80
100
)ld/
g(ni
bol
go
meH
9.6
9.8
10.0
10.2
10.4
10.6
Primary diagnosis
Modality
stneita
pfot
necreP
0
20
40
60
80
100
stneita
pfot
necreP
0
20
40
60
80
100
stneita
pfo
tnecreP
25
30
35
40
All 1 2 3 4 5 6 NC HB U
stneita
pfo
tnecre P
45
50
55
60
65
70
Unit affiliation (see table at right for codes)
All 1 2 3 4 5 6 NC HB U
/m2 )
gk(IM
B
25
26
27
28
29
All 1 2 3 4 5 6 NC HB U
eGFR
(ml/
min
/1.7
3 m
2 )
9.0
9.5
10.0
10.5
11.0
Hispanic ethnicity
Mean hemoglobin at initiation Percent receiving EPO at initiation
Percent with albumin < test’s lower limit Mean BMI at initiation Mean eGFR at initiation
Hispanic-MexicanHispanic-otherNon-HispanicOther/unknown
Hemodialysis
Peritoneal dialysis
White
Black
Native American
Asian
Other/unknown
Diabetes
Hypertension
Glomerulonephritis
Cystic kidney
Other/unknown
{11.11} Characteristics of incident dialysis patients, by unit affiliation, 2002
Patient qaracterisics, by unit aYliation
197H2004 USRDS Annual Data Report
Unit affiliation (see table below for codes)
1 2 3 4 5 NC HB0
200
400
600
800
1,000
1,200 Number of units Number of patients (in thousands)
1 2 3 4 5 NC HB0
10
20
30
40
50
60
70
80
)sraey(e
ganae
M
48
50
52
54
56
58
60
62
selameft
necreP
40
42
44
46
48Age Gender: Female
s tnei ta
pfo
tnecreP
0
20
40
60
80
100
citebai
dtnecreP
24
28
32
36
40
44Race Diabetic status: Diabetics
Unit affiliation (see table below for codes)
All 1 2 3 4 5 6 NC HB U
stneita
pfo
tnecr eP
0
20
40
60
80
100
All 1 2 3 4 5 6 NC HB U
stneita
pfo
tnecreP
0
20
40
60
80
100Primary diagnosis Modality
Hemodialysis
Peritoneal dialysis
White
Black
Native American
Asian
Other/unknown
Diabetes
Hypertension
Glomerulonephritis
Cystic kidney
Other/unknown
1998
2002
he mean age of the incident ESRD
population is 62.3, and with the
exception of patients in facilities
owned by Dialysis Clinics Incorporated,
mean age varies little by unit affiliation (Fig-
ure 11.11). Differences in racial distribu-
tion among units are also unremarkable, as
are differences by diabetic status, although
there are fewer diabetics in hospital-based
units. The majority of patients in all units
have diabetes as their primary cause of renal
failure, and over 90 percent are on hemodi-
alysis. Mean hemoglobins at initiation av-
erage close to 10 g/dl, and on a unit level do
not appear related to whether or not pa-
tients receive EPO prior to starting therapy.
The percentage of patients with serum al-
bumins below the test’s lower limit is high-
est for hospital-based units, and mean BMIs
and eGFRs are similar for all unit types.
The number of chain-affiliated dialysis
units, and their patient populations, have
increased quite dramatically since 1998 (Fig-
ure 11.12). The number of patients treated
by Renal Care Group, for example, has
doubled, while the population in units
owned by Fresenius has grown 42 percent.
In units not affiliated with chains, in con-
trast, the number of patients has increased
only 2 percent (growth in unit counts is
slightly higher, at 7 percent), and it has fallen
more than 12 percent in units that are hos-
pital-based.
The prevalent ESRD population tends to
be younger than the incident population,
with a mean age of 56.9 across all units (Fig-
ure 11.13). The distribution of females by
affiliation is similar. Across units the per-
cent distribution of prevalent diabetics is simi-
lar, and is slightly lower than that found in
the incident population.
{Figures 11.11–12} incident ESRD patients, 2002.Facility data obtained from the CMS annual End-Stage Renal Disease Facility Survey, the CMS Inde-pendent Renal Facility Cost Reports, & the CMS“Dialysis Facility Compare” website. {Figure 11.13}December 31 point prevalent ESRD patients, 2002.Facility data obtained from the CMS annual End-Stage Renal Disease Facility Survey, the CMS Inde-pendent Renal Facility Cost Reports, & the CMS“Dialysis Facility Compare” website. x {Figure11.11} The lower limit of albumins measured by bro-mcresol purple is 3.2 g/dl, & by bromcresol green is 3.5g/dl. {Figure 11.12} Chain 6 did not exist in 1998.
{11.12} Unit & patient counts, by unit affiliation
{11.13} Characteristics of prevalent dialysis patients, by unit affiliation, 2002
All · All unitsChain 1 · FreseniusChain 2 · GambroChain 3 · DaVita
Chain 4 · Renal Care GroupChain 5 · Dialysis Clinics, Inc.Chain 6 · Nat’l Nephrology Assoc.NC · Non-chain units
HB · Hospital-based unitsU · Unknown affiliation
198 ESRD providersh
stinuf
otnecreP 0
20
40
60
80
100Hemoglobin (g/dl) EPO dose (units)
Unit affiliation (see table at right for codes)
All 1 2 3 4 5 6 NC HB U All 1 2 3 4 5 6 NC HB U0
20
40
60
80
100
<9
9-<10
10-<11
11-<12
12+
<7,000
7,000- <11,000
11,000- <16,000
16,000- <23,000
23,000+
<60
60-<65
65-<70
70-<75
75+
<22-<2.6
2.6-<3.03.0-<3.6
3.6+Median URR (%): hemodialysis Mean weekly Kt/V: peritoneal dialysis
Unit affiliation (see table at right)
All 1 2 3 4 5 6 NC HB
stneita
pfo
tnecreP
0
20
40
60
80
100
11.85+ (11.93)11.75 to <11.8511.68 to <11.75
11.59 to <11.68below 11.59 (10.46)
17,484+ (18,546) 16,593 to <17,48415,433 to <16,593
14,146 to <15,433below 14,146 (11,007)
11.85+ (12.00)11.75 to <11.8511.68 to <11.75
11.59 to <11.68below 11.59 (10.74)
17,484+ (18,564)16,593 to <17,48415,433 to <16,593
14,146 to <15,433below 14,146 (11,393)
80.5+ (85.0)76.6 to <80.574.4 to <76.6
69.3 to <74.4below 69.3 (55.2)
80.5+ (87.4)76.6 to <80.574.4 to <76.6
69.3 to <74.4below 69.3 (64.9)Insuff. data
Cuffed catheter Non-cuffed catheter AV fistula AV graft
cross all unit affiliations, mean
hemoglobin levels are at or above
11 g/dl—the K/DOQI target—in
77–89 percent of prevalent dialysis patients
(Figure 11.14). In units owned by DaVita
and Renal Care Group, 44–51 percent of
patients have a mean hemoglobin of 12 g/dl
or greater.
The percent of dialysis patients receiv-
ing mean weekly EPO doses of 16,000 units
and above ranges from 34 in units owned
by Dialysis Clinics, Inc. to 43 in those owned
by DaVita.
K/DOQI also set a target urea reduction
ratio (URR) guideline of ≥ 65 percent, which
in the CPM dataset is met by 86 percent of
hemodialysis patients—from 84 percent of
patients in non-chain and hospital-based
units to 91 percent of those treated in units
owned by Gambro. Sixty percent of perito-
neal dialysis patients in the CPM dataset
have a mean weekly Kt/V of 2.6 or above.
Chain-affiliated units with the highest
mean hemoglobin levels are more highly
concentrated west of the Mississippi River;
non-chain units with the lowest hemoglo-
bins are located primarily in the southeast
and southwestern states (Figure 11.15).
Variations in mean EPO dose track accord-
ing to hemoglobin levels, demonstrating a
general tendency within providers to main-tain established hemoglobin targets.
The percent of patients with internal
accesses is slightly higher in chain-affiliated
units across the country (Figure 11.16).
K/DOQI guidelines recommend in-
creased use of arteriovenous fistulas. In
2001, 31 percent of hemodialysis patients
had this type of vascular access, and use var-
ied little by unit affiliation (Figure 11.17).
Overall, less than a third of incident di-
alysis patients initiate therapy with a serum
albumin above the test’s lower limit (Figure
{11.14} Anemia treatment & dialysis adequacy, by unit affiliation, 2002
Provider compliance with K/DOQI guidelines
{11.15} Geographic variations in anemia treatment, by unit affiliation, 2002
{11.16} Geographic variations in the % of pts with an internal access, by unit affiliation, 2002 CPM data
{11.17} Vascular access use, by unit aff., 2001
Mean hemoglobin: chain Non-chain
Mean EPO dose: chain Non-chain
Chain Non-chain
199H2004 USRDS Annual Data Report
stneita
pfo
tnecreP
0
20
40
60
80
100
Unit affiliation (see table below for codes)
All 1 2 3 4 5 6 NC HB U All 1 2 3 4 5 6 NC HB UAll 1 2 3 4 5 6 NC HB U
Albumin > test’s lower limit Serum creatinine <10 mg/dl Hemoglobin 11 g/dl
All 1 2 3 4 5 6 NC HB
etarn
oitaniccav
hcaeg
ni tro
pe rsti
nu f
o%
0
20
40
60
80
100
Unit affiliation (see table below for codes)
All 1 2 3 4 5 6 NC HB
Influenza Pneumococcal pneumonia
Unit affiliation (see table below)
All 1 2 3 4 5 6 NC HB
detaniccav
tnecreP
80
82
84
86
88
90
92
94
<25%
25-49%
50-74%
75-100%
Unknown
Not offered
>_
{11.18} Nutritional & hemopoietic parameters in incident dialysis patients, by unit affiliation, 2002
{11.19} Dialysis patient vaccinations, by unit affiliation, 2001 {11.20} Hepatitis B: staff vaccination rates, 2001
All · All unitsChain 1 · FreseniusChain 2 · GambroChain 3 · DaVitaChain 4 · Renal Care GroupChain 5 · Dialysis Clinics, Inc.Chain 6 · Nat’l Nephrology Assoc.NC · Non-chain unitsHB · Hospital-based unitsU · Unknown affiliation
11.18), and over 80 percent be-
gin treatment with a serum
creatinine of less than 10 mg/
dl. Slightly more than one-
quarter of new patients begin
dialysis with a hemoglobin at
or above the recommended
K/DOQI guideline of 11 g/dl.
In 2001, nearly half of the
dialysis units across the country reported giving influenza vaccina-
tions to 75–100 percent of their patients (Figure 11.19). Sixty-two
percent of units owned by Renal Treatment Centers achieved this
rate, compared to fewer than one-third of those owned by Gambro.
Vaccinations for pneumococcal pneumonia are not offered in
41 percent of units nationwide. Units owned by Renal Treatment
Centers are again most likely to vaccinate their patients, with al-
most one-third of these units reporting a pneumococcal pneumo-
nia vaccination rate of 75–100 percent in 2001 (Figure 11.19).
Hepatitis B vaccination rates for dialysis unit staff range from
85 to 92 percent, with an overall rate of 87 percent nationwide
(Figure 11.20).
{Figure 11.14} period prevalent dialysis patients, 2002. Hemoglobin graph in-cludes only patients treated with EPO, & the mean hemoglobin represents the av-erage hemoglobin value for the year across all patients. EPO dose adjusted forinpatient days. URR & Kt/V data obtained from 2002 CPM data, & include onlypatients who are in both the USRDS & CPM databases. {Figure 11.15} periodprevalent dialysis patients, 2002. Hemoglobin maps include only patients treatedwith EPO, & the mean hemoglobin represents the average hemoglobin value forthe year across all patients. EPO dose adjusted for inpatient days. {Figure 11.16}prevalent hemodialysis patients from the 2002 CPM data who are also in theUSRDS database; current access determined from CPM data. {Figure 11.17}prevalent hemodialysis patients. {Figure 11.18} incident dialysis patients, 2002.{Figures 11.17 & 11.19–20} data obtained from the CDC’s National Surveillanceof Dialysis-Associated Diseases in the United States. x {Figure 11.18} Thelower limit of albumins measured by bromcresol purple is 3.2 g/dl, & by bromcresolgreen is 3.5 g/dl.
200 ESRD providersh
90 91 92 93 94 95 96 97 99 00 01
stin
ufot
necreP
0
20
40
60
80
100
90 91 92 93 94 95 96 97 99 00 01
stin
ufo
tnecre P
0
20
40
60
80
100
ESRD network
17 18 9 15 13 7 11 6 14 All 16 10 8 5 3 12 1 4 2
esuer
taht
stin
ufo
tne cre P
0
20
40
60
80
100
Reuse No reuse
Reuse, chain
No reuse, chain
Reuse, non-chain
No reuse, non-chain
Units addedUnits unchanged
Units dropped
fter peaking at 82 percent in 1997, the percent of units
that reuse their dialyzers has dropped slightly, to 76 per-
cent in 2001 (Figure 11.21). With Fresenius’ recent deci-
sion to discontinue reuse in its dialysis units, this decrease is likely
to become more significant.
Changes in reuse practices are occurring primarily in chain-
affiliated units (Figure 11.22). Since 1990, the proportion of non-
chain units reusing dialyzers has fluctuated only between 65–70
percent, while in chain-owned units the proportion fell from 96–
98 percent in the
early 1990s to 82
percent in 2001.
Only 54 percent
of Network 2 units
reuse dialyzers (Fig-
ure 11.23); this net-
work also contains
a large proportion of
non-profit and hos-
pital-based units. In
Networks 17, 18,
and 9, in contrast,
88–91 percent of
{11.21} Reuse practices, overall
Reuse pracices
{11.22} Reuse practices, by chain status {11.23} Reuse practices, by ESRD network, 2001
{11.24} Changes in the location of reuse & non-reuse units between 1997 & 2001
units practice reuse.
The map of units that reuse hemodialyzers parallels that of
chain-affiliated and non-chain units. There is a high concentra-
tion of reuse units in the eastern half of the country, with densities
highest in the Mid-Atlantic and East Coast states (Figure 11.24).
The map of non-reuse units shows no discernable pattern other
than higher concentrations in the eastern United States, an area
that contains the majority of dialysis units.
Due to its lack of carcinogenic properties and its efficacy as a
disinfectant, the use of peracetic acid in the reuse process has grown
steadily, from 49 percent in 1990 to 62 percent in 2002 (Figure
11.25). Rates of peracetic acid use do, however, vary quite widely
across the country (Figure 11.26). In Networks 12, 16, and 11, the
germicide is used in 78–80 percent of units; in Networks 3, 13, and
14, however, 41–48 percent of units continue to use formalde-
hyde. The wide use of peracetic acid is reflected geographically in
Figure 11.27. Formaldehyde use is concentrated in units located in
the eastern third of the country, and should decrease even more
with Fresenius’ decision to discontinue reuse.
The percentage of units using bleach is highest in the southern
and eastern states; this is most likely a direct correlation with the
location of units using formaldehyde as a disinfectant (Figure 11.28).
Differences in hemoglobin levels, EPO use, serum albumin
Reuse units Non-reuse units
201H2004 USRDS Annual Data Report
0
20
40
60
80
100 Hemoglobin (g/dl) EPO dose (units)
Germicide (see table at right for codes)
PA F G B B+G NR
stneita
pfot
n ecre P
0
20
40
60
80
100
PA F G B B+G NR
Serum albumin (g/dl) Urea reduction ratio (%)
ESRD network
12 16 11 18 15 5 6 9 17 2 All 4 1 7 3 13 10 8 14
stneita
pfo
tn ecreP
0
20
40
60
80
100
90 91 92 93 94 95 96 97 99 00 01
stin
ufo
tnecreP
0
20
40
60
80
100
38.8+ (61.1)29.6 to <38.818.8 to <29.6
2.9 to <18.8below 2.9 (0.0)
Formaldehyde
Glutaraldehyde
Peracetic acid
Other (includes heat & Amuchina)
Peracetic acid Formaldehyde Glutaraldehyde Other
<99-<10
10-<1111-<1212+
<33-<44-<55+
<60
60-<65
65-<70 70-<75 75+
<7,0007,000-<11,000
11,000-<16,000
16,000-<23,000
23,000+
Other
Peracetic acidGlutaraldehyde
Formaldehyde
{11.25} Germicide use, overall {11.26} Germicide use, by ESRD network, 2001
{11.27} Location of units, by germicide type, 2001
{11.29} Clinical parameters, by germicide type, 2001
{11.28} Percent of units using bleach, 2001
PA · Peracetic acidF · FormaldehydeG · GlutaraldehydeB · BleachB + G · Bleach + germicideNR · No reuse
levels, and dialysis adequacy are unremark-
able when based on germicide type (Figure
11.29).
{Figures 11.21–28} data obtained from the CDC’sNational Surveillance of Dialysis-Associated Diseasesin the United States, 2001. Units in Figures 11.24 &11.27 are mapped by zip code. {Figure 11.29} preva-lent hemodialysis patients treated in a known pro-vider, 2001. Data on hemoglobin & EPO dose includeonly EPO-treated patients. Albumin & URR data ob-tained from CPM data, include only patients who arein both the CPM & USRDS database, & reflect themedian value for each patient. EPO doses adjustedfor IP days. x {Figure 11.25} “Other” includes heatand Amuchina.
The CDC did not conduct a survey in 1998, & data arenot available for 2002. Figure 2.38, in Chapter Two,contains a map of the ESRD networks; a list of networkcontacts can be found on page 230 of the Appendix.
Chap
terJo
summary
Chapter summary202 ESRD providersh
Provider compliance with K/DOQI guidelines{Figure 11.14 In the prevalent population, average hemoglobin levels vary by chain, with units owned by DaVita and Renal Care Grouphaving the greatest percentage of patients with hemoglobins of 12 g/dl or higher for a year. EPO doses are highest in units owned by DaVita.The percent of dialysis patients receiving mean weekly EPO doses of 16,000 units and above ranges from 34 in units owned by DialysisClinics, Inc. to 43 in units owned by DaVita.} {Figure 11.19 Patient vaccination rates differ across the dialysis providers. Renal Care Groupappears to have the highest rates of vaccination for influenza and pneumococcal pneumonia, and Gambro among the lowest rates.}
Reuse practices{Figure 11.21 Dialyzer reuse grew during the 1990s, but since 1997 has fallen slightly.} {Figure11.22 Since 1990, the proportion of non-chain units reusing dialyzers has fluctuated onlybetween 65–70 percent, while in chain-owned units the proportion fell from 96–98 percent inthe early 1990s to 82 percent in 2001.} {Figure 11.25 The use of peracetic acid in the reuseprocess has grown steadily since 1990, from 49 to 62 percent.} {Figure 11.29 Hemoglobinlevels and erythopoietin doses vary slightly in relation to the type of germicide used by a dialysisunit, while serum albumin levels and urea reduction ratios are relatively consistent. Since reusepractices and germicide use are highly dependent on providers, these results need to be factoredinto analyses of provider-specific outcomes.}
Maps: National means & patient populationsFigure number 11.15 11.15 11.15 11.15 11.16 11.16
Hgb ch Hgb nch EPO ch EPO nch chain n-chainOverall value for all pts 11.7 11.7 16,894 15,770 75.7 74.0Total patients 130,357 68,183 130,357 68,183 4,677 2,657Overall value for pts mapped 11.7 11.7 16,865 15,762 75.8 74.1Missing HSA/state: pts dropped 2,798 1,387 2,798 1,387 102 1,040
Patient characteristics by unit afiliation{Figure 11.11 The distribution of incident patients by age is generally comparable across providers. There isconsiderable variation by provider in the pre-dialysis treatment of anemia, with the percent of patientsreceiving EPO prior to initiation ranging from 29 to 38. The percentage of patients with serum albuminsbelow the test’s lower limit is highest in hospital-based units, and mean BMIs and eGFRs are similar for allunit types.} {Figure 11.12 While the dialysis chains, particularly Fresenius, have grown significantly, there hasbeen little change in the number of non-chain and hospital-based units.}
Diferences in provider growth{Figure 11.2 Since 1998 the greatest growth in the number of in-center hemodialysis treatments has occurred in Network 17,and the smallest growth in Network 5. In most networks the number of peritoneal dialysis treatments has fallen or remainedrelatively stable, though significant growth has occurred in Networks 9, 10, and 18.} {Figure 11.5 The number of patientsreceiving treatment in chain-affiliated units has increased six-fold since 1993, while the number treated in non-chain unitsfell after peaking in the early 1990s, and has remained stable during the past four years.}
Introduction{Figure 11.1 The number of freestanding units continues to grow, and has more than tripledover the last twelve years.}