US
RD
S
USRDS ASN 2002
United States Renal Data System: 2002 ASN Symposium
Chronic Kidney Disease in the Medicare Population:
Progression to ESRD vs Death,
Morbidity, Quality of Care, and Cost
US
RD
S
USRDS ASN 2002
Structure of the USRDS
Larry Agodoa, MD
Co-Project Officer
United States Renal Data System
NIH, NIDDK, DKUHD
US
RD
S
USRDS ASN 2002
Administrative Oversight of the USRDS
Department of Health &
Human Services (DHHS)
Centers for Medicare and
Medicaid Services (CMS)
National Institutes of
Health (NIH)
Office of Clinical
Standards & Quality
(OCSQ)
Office of Strategic
Planning (OSP)
National Institute of Diabetes
and Digestive and Kidney
Diseases (NIDDK)
Division of Kidney,
Urologic, and Hematologic
Diseases (DKUHD)
ESRD
Networks
CMS USRDS
Project
Coordinator
USRDS
CC
NIH Project
Officers
Cardiovascular
SSC
Quality of Life
SSC
Nutrition
SSC
Economic
SSC
US
RD
S
USRDS ASN 2002
USRDS Investigative Centers
Coordinating Center-Director Allan Collins MD, Deputy Director
Bertram Kasiske MD
Cardiovascular SSC-Director Charles Herzog MD, Deputy
Director Blanche Chavers MD
Economic SSC- Director Lawrence Hunsicker MD, Deputy
Director John Brooks PhD
Rehabilitation/Quality of Life SSC-Director Nancy Kutner PhD,
Deputy Director Donna Brogan PhD
Nutrition/Malnutrition SSC-Director Glen Chertow MD, Deputy
Director George Kaysen MD MPH
2002 Annual Data Report: online at WWW.USRDS.ORG
ADR with enclosed CD shipped October 31, 2002
United States Renal Data System
US
RD
S
USRDS ASN 2002
Chronic Kidney Disease in the Medicare Population
USRDS data: Paul Eggers, PhD
Trends in Diabetes in the General Population, CKD, likelihood of death vs ESRD and progression of comorbid conditions
Allan Collins MD FACP: Coordinating Center
Morbidity and preventive care in the general Medicare, CKD and dialysis populations
Lawrence Hunsicker MD: Economic SSC
Comparing Cost of Medicare non-CKD, CKD and dialysis populations
John Brooks PhD: Economic SSC
US
RD
S
USRDS ASN 2002
Getting Research Data Sets from the USRDS
Paul Eggers, PhD
Co-Project Officer
United States Renal Data System
NIH, NIDDK, DKUHD
US
RD
S
USRDS ASN 2002
Approved USRDS requests for standard analytic files: November 2001 to October 2002
Category N
Transplantation/Immunosuppression 10
Ethnicity/Race/Disparities 6
Cardiovascular/PVD 6
Costs 3
Other (modality, nutrition, HIV, GI, vascular access, center effects, VA)
13
Total 38
US
RD
S
USRDS ASN 2002
Approved USRDS manuscripts using SAF files: November 2001 to October 2002
Category N
Transplantation/Immunosuppression 11
Ethnicity/Race/Disparities 1
Cardiovascular/PVD 3
Costs 2
Other (Nutrition, Timing/Initiation, center effects, hip fracture, quality of life)
12
Total 29
US
RD
S
USRDS ASN 2002
Outline for USRDS study proposals:
USRDS ADR
US
RD
S
USRDS ASN 2002
Medical: A. Collins MD, L. Hunsicker MD, M.
McBean MD MPH, A. Murray MD MPH
Biostatistics and epidemiology: D. Gilbertson PhD, SL. Li MS, C. Solid
MS, J. Ma PhD, J. Xue DVM PhD, T. Roberts MS, T. Louis PhD
Economics: W. Manning PhD, S. Li MS, J. Liu PhD,
J. Ebben BS, J. Brooks PhD
Data systems: S. Chen MS, C Arko MS,
F Dalleska, MS, E. Frazer BS
Editorial & Graphics: S. Everson PhD,
D. Berrini BS, E. Constantini MA
Chronic Kidney Disease in the Medicare Population: Acknowledgement
USRDS team contributing to this presentation
US
RD
S
USRDS ASN 2002
Trends in Diabetes in the General Population, CKD, likelihood of
death vs ESRD and progression of comorbid conditions
Allan J. Collins, MD FACP
Director, USRDS Coordinating Center
US
RD
S
USRDS ASN 2002
Chronic Kidney Disease is a Major Comorbid Condition Within the Medicare Population
The Medicare program in the US currently consumes 2.3% of the gross domestic product and it is projected to more than double to 5.4% by 2030 (CBO* 9-18-2002).
Chronic diseases in the Medicare population are a major public health concern and cost to the US.
Approximately 20-26 million Americans have some degree of kidney damage (NKF K/DOQI Guidelines Feb 2002), which contributes to the ever increasing number of individuals advancing to ESRD.
*Congressional Budget Office
US
RD
S
USRDS ASN 2002
Perspectives on chronic disease: Medicare Beneficiaries and their Cost of Care* 5% Medicare sample, 1997 cohort (CBO 9-18-02)
50
2
20
42
47
9
25
5
0
20
40
60
80
100P
erc
ent
of
patients
and c
osts
(%
)
Fee-for-Service
Beneficiaries Cost of Care
*CBO Testimony: Dan L. Crippen, Director
% Patients % Medicare
spending
US
RD
S
USRDS ASN 2002
Percent of Medicare Beneficiaries and spending by the number of Chronic Conditions*: 5% Medicare sample, 1997 (CBO 9-18-2002)
47%
22%
15%
16%
*CBO Testimony: Dan L. Crippen, Director
88%
7%
4%
1%
3+ conditions 2 conditions 1 condition none
Medicare beneficiaries Medicare spending
US
RD
S
USRDS ASN 2002
Prevalence of DM in the General Population:
CDC Annual Survey*; all age groups
1992 1994 1996 1998 2000
Rate
per m
illion
pop
ula
tion
0
50
100
150
200
250
300
350
Diabetes
Hypertension
Glomerulonephritis
Cystic kidney
All
Incident rates in the ESRD population Prevalence of diabetes in the general population
1990 1992 1994 1996 1998 2000
Rate
per 1,0
00 p
op
ula
tion
4.5
5.0
5.5
6.0
6.5
7.0
7.57.5
*Ali H. Mokdad, PhD, et al Diabetes Trends in the U.S.: 1990-1998
Ali H. Mokdad, PhD, et al The continuing Epidemics of Obesity and Diabetes in the United States
Data taken from CDC’s Behavioral Risk Factor Surveillance System (BRFSS) in 2000
51% increase
US
RD
S
USRDS ASN 2002
Costs of the ESRD & Medicare programs figure 12.2, dollars in 2000 are inflated by 2% to account for costs incurred not reported
0
1
2
3
4
5
6 ES
RD
's p
erc
en
tag
e o
f Me
dic
are
pro
gra
m
20001999199819971996199519941993199219910
5
10
15To
tal E
SR
D d
olla
rs(b
illion
s)
100
150
200
250
To
tal M
ed
ica
re d
olla
rs(b
illion
s)
USRDS 2002 ADR
ESRD Program $
Total Medicare Program $
ESRD as a % of Medicare
4.5%
5.8%
To
tal
do
llar
s ($
Bil
lio
ns)
E
SR
D as a %
of M
edicare
US
RD
S
USRDS ASN 2002
ESRD as a Chronic Disease: Trends in Patient Counts 1978-2000
0
50,000
100,000
150,000
200,000
250,000
300,000
350,000
400,000
1978
1980
1982
1984
1986
1988
1990
1992
1994
1996
1998
2000
Pa
tie
nt
co
un
ts
Incident ESRD
Prevalent dialysis
Prevalent ESRD
USRDS 2002 ADR
96,192
275,053
378,862
US
RD
S
USRDS ASN 2002
Chronic Kidney Disease as a Major Comorbid Condition Within the Medicare Population
How does the burden of Chronic Kidney Disease compare to the other major diseases such as
Diabetes Mellitus or Congestive Heart Failure?
US
RD
S
USRDS ASN 2002
Distribution of DM, CHF and CKD in the Senior US Population: NHANES vs Medicare Dx Codes
Identify the DM and CHF in NHANES III from patient history questions (Bench Mark)
Identify DM, CHF and CKD from diagnosis codes within the Medicare Part A & B claims (clinically apparent disease) DM identified by established methods used by CMS,
NCQA and health plans*
CHF and CKD identified by the same method used for DM*
Compare NHANES and Medicare claims methods to validate the approach and then identify the trends in CVD complications associated with CKD
* Hebert et al, American Journal of Medical Quality 1999, 14:270-277
Participated Medicare Current Beneficiary Survey (MCBS) Access to Care 1992-1993
US
RD
S
USRDS ASN 2002
13.6
16.317.0
17.8
10.0
11.0
12.0
13.0
14.0
15.0
16.0
17.0
18.0
19.0
1992-1993* 1996-1997 1997-1998 1998-1999
Cohort year
Pe
rce
nt
pa
tie
nts
Defining DM in the Medicare population: time trends & growth
* Hebert et al, American Journal of Medical Quality 1999, 14:270-277
Participated Medicare Current Beneficiary Survey (MCBS) Access to Care 1992-1993
Validation cohort
31%
US
RD
S
USRDS ASN 2002
Identified diseases, by data source figure p.24, patients age 65 & older, comparison regressed to 1990
Diabetes Congestive
heart failure
Chronic
kidney dis.
Perc
ent
of
pa
tients
0
2
4
6
8
10
12
14
NHANES III
General Medicare
*Stage IV & V (1.1%), eGFR <30 ml/min, NHANES III
*
Imputed eGFR, 34 ml/min: General Medicare population with clinically apparent disease from Dx codes (1.8%), 1990
: 1996-1999 data regressed to 1990
: 1988-1994
USRDS 2002 ADR
Imputed eGFR for the
most current Medicare
1997-1998 cohort
40.1 ml/min (3.5% of
Medicare as CKD)
US
RD
S
USRDS ASN 2002
Distribution of the Senior US Population by eGFR (NHANES III data 1988-1994) figure p.25, patients age 65 & older
eGFR groups (NHANES III)
0-<15 30-<45 60-<7515-<30 45-<60 75-<90
90+
Nu
mb
er
of
pa
tie
nts
(m
illio
ns)
0
2
4
6
8
10Patient counts Hemoglobin
All 0-<1515-<30
30-<4545-<60
60-<7575-<90
90+
Hem
oglo
bin
(g
/dl)
10
11
12
13
14
15
eGFR-(Coresh et al method)
: Age 65+ 5.9 million seniors have
stage III-V CKD
29%
20%
50%1%
0%
eGFR 90+ eGFR 60<90
eGFR 30<60 0.1%
eGFR:
15<30
eGFR:
<15
CKD
Stage III CKD
Stage
IV CKD
Stage
V
eGFR in the
normal range
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Incidence of ESRD during one year follow-up by DM and CKD in general Medicare Patients 5% Medicare sample (100% estimate), 1997-1998 cohort
80.31
19.23
15.27
12.86
41.73
26.18
2.72
0
20
40
60
80
100
Perc
ent
of
patients
(%
)
DM,CKD
DM, noCKD
No DM,CKD
No DM,no CKD
Entry period
General Medicare
ESRD during one
year follow-up
1.7 19.7%
80.8%
N=1,265,831(25,316,620) N=1,711(34,220)
19,940,320
688,680
4,007,980
406,800
% P
atients
General Medicare
est. by group
US
RD
S
USRDS ASN 2002
Percent of patients advancing to ESRD or died during two year follow-up by DM, CKD vs dialysis 5% Medicare sample , 1996-1997 cohort (RR: Death vs ESRD)
9.4 14.7
29.040.0
85.073.2
65.1 60.0
24.6
90.5
0
20
40
60
80
100
No DM, no
CKD
DM, no
CKD
No DM,
CKD
DM, CKD dialysis
Perc
ent
of
patie
nts
(%
)
Event free
ESRD
Death
Status in the entry period
0.07 0.31 2.25
5.85
RR: 5.0 RR: 11 RR: 47 RR: 134
Follow-up: 12-31-1999
(Prev. 1997)
US
RD
S
USRDS ASN 2002
Prevalence and Incidence of Cardiovascular Disease in the Medicare non-CKD and CKD populations: 5% Medicare data 1996-1997 entry, followed thru 1999
1996-1997 Medicare Age 67+
(all pts survive the entire period) 1-1-96 12-31-97 12-31-98 12-31-99
Define CKD and CVD from
Medicare Dx codes*
in the entry
Follow-up patients w/o CVD
(remove ESRD, HMO pts)
Determine
Probability
of a new
Dx of CVD for
all outcome groups
Track patients
To ESRD,
new Dx of CKD,
death or thru 12-99
Period of
Advancing
comorbidity
Final outcome
period
Study design
* Hebert et al, American Journal of Medical Quality 1999, 14:270-277
Participated Medicare Current Beneficiary Survey (MCBS) Access to Care 1992-1993
All patients alive
US
RD
S
USRDS ASN 2002
Prevalence of CVD in general Medicare during 1996-97: age 67+; CKD during entry
80
52
40
27
36
55
44
23
11 10 12
25
45
24
11 11 12
26
0
20
40
60
80
100
All CVD ASHD CHF CVA/TIA PVD Other CVD
Perc
ent
of
patients
(%
)
CKD (N=27,974)
Non-CKD (N=983,281)
All patients (N=1,011,255)
All patients: % with CVD Dx
US
RD
S
USRDS ASN 2002
28
1311
912
1917
75 5 5
11
0
10
20
30
40
All CVD ASHD CHF CVA/TIA PVD Other
CVD
Ra
te p
er
10
0 p
t-y
rs CKD Non-CKD
Unadjusted incidence by type of CVD per 100 patient-years at risk during 1998
All patients: new CVD Dx
US
RD
S
USRDS ASN 2002
0.00
0.10
0.20
0.30
0.40
0 1 2 3 4 5 6 7 8 9 101112
CKD/ESRD (N=8,167)
Died (N=17,017)
All (N=554,024)
No events (N=528,840)
0.00
0.10
0.20
0.30
0.40
0 1 2 3 4 5 6 7 8 9 101112
ESRD (N=124)
Died (N=422)
All (N=5,712)
No events (N=5,166)
Life table estimates for probability of incident CVD during 1998 by CKD and patient groups Non-CVD patients 1996-’97 and did not die or develop ESRD/CKD in 1998
CKD patients*
Follow-up time (month)
Pro
babili
ty o
f C
VD
Non-CKD patients*
0.00
0.10
0.20
0.30
0.40
0 1 2 3 4 5 6 7 8 9 101112
CKD/ESRD (N=8,167)
Died (N=17,017)
All (N=554,024)
No events (N=528,840)
*All Log-rank tests for differences
among patient groups: p<.0001
RR: 1.61 for new CVD in CKD
0.241
0.152
US
RD
S
USRDS ASN 2002
Life table estimates for probability of incident CHF during 1998 by CKD status and patient groups Non-CHF patients 1996-’97, alive thru 1998, by outcome in 1999
CKD patients
Follow-up time (month)
Pro
babili
ty o
f C
HF
Non-CKD patients
0.00
0.10
0.20
0.30
0 1 2 3 4 5 6 7 8 9 101112
ESRD (N=318)
Died (N=1,747)
All (N=16,671)
No events (N=14,606)
0.00
0.10
0.20
0.30
0 1 2 3 4 5 6 7 8 9 101112
CKD/ESRD (N=17,203)
Died (N=36,141)
All (N=879,100)
No event (N=825,756)
0.024
0.060
RR: 2.25 for new CHF in CKD
*All Log-rank tests for differences
among patient groups: p<.0001
0.045
0.108
US
RD
S
USRDS ASN 2002
Conclusions
CKD is common is the Medicare population
Medicare Claims Dx of CKD represents only those patients with more advanced disease (late stage III+ disease: mean eGFR 34 ml/min)
The Medicare population is 5 to 134 times more likely to die than to ever reach ESRD
CVD is twice as common and advances at 1.7-2.5 times the rate in the CKD vs the non-CKD population
CVD advances at a similar rate in those individuals that are destined to CKD, ESRD or death
Premature Death is the major issue
for the CKD population!
US
RD
S
USRDS ASN 2002
Morbidity and preventive care in the general Medicare,
CKD and dialysis populations
Lawrence Hunsicker, MD
Director, Economic Special Study Center
US
RD
S
USRDS ASN 2002
Issues to be Addressed
Mortality and hospitalization rates among the Medicare CKD patients, though lower than among the ESRD population, are much higher than among non-CKD patients.
Preventive care is lagging in the Medicare CKD population, as it is among patients with ESRD.
Preventive care (in particular, influenza vaccination) has the potential to reduce both deaths and illness in both the CKD and ESRD populations.
US
RD
S
USRDS ASN 2002
All-cause mortality in the general Medicare & dialysis populations, patients age 65+, all patients figure 9.9, period prevalent patients, 1999, unadjusted
Cardiovascular
disease
No cardiovascular
disease
50
150
250
350
450WhiteAll
De
ath
s p
er
1,0
00
pa
tie
nt
ye
ars
at
risk
50
Non-CKD CKD Dialysis Non-CKD CKD Dialysis
50
150
250
350
450Native American Asian
De
ath
s p
er
1,0
00
pa
tie
nt
ye
ars
at
risk
Death
s p
er
1,0
00 p
atient years
at risk
USRDS 2002 ADR
Dialysis
US
RD
S
USRDS ASN 2002
All-cause mortality in the general Medicare & dialysis populations, patients age 65+, by diabetic status figure 9.6, period prevalent patients, 1999, unadjusted
Cardiovascular
disease
No cardiovascular
disease
Diabetic Non-diabetic
Non-CKD CKD Dialysis Non-CKD CKD Dialysis
De
ath
s p
er
1,0
00
pa
tie
nt
ye
ars
at
risk
0
50
100
150
200
250
300
350
400
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
All-cause mortality in the general Medicare & dialysis populations, patients age 65+, by gender figure 9.7, period prevalent patients, 1999, unadjusted
Cardiovascular
disease
No cardiovascular
disease
Non-CKD CKD Dialysis Non-CKD CKD Dialysis0
50
100
150
200
250
300
350
400Male Female
De
ath
s p
er
1,0
00
pa
tie
nt
ye
ars
at
risk
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
All-cause mortality in the general Medicare & dialysis populations, by patient age figure 9.8, period prevalent patients, 1999, unadjusted
Cardiovascular
disease
No cardiovascular
disease
Non-CKD CKD Dialysis Non-CKD CKD Dialysis0
100
200
300
400
5000-44 45-64
De
ath
s p
er
1,0
00
pa
tie
nt
ye
ars
at
risk
USRDS 2002 ADR
Cardiovascular
disease
No cardiovascular
disease
Dialysis Non-CKD CKD Dialysis Non-CKD CKD Dialysis0
100
200
300
400
50065-74 75+
US
RD
S
USRDS ASN 2002
Expected remaining lifetime in patients with increasing morbidity, by age figure 9.25, chronic kidney disease & diabetes, prevalent dialysis patients, 2000
65-74 75-84 85+ Exp. remaining lifetime (yrs) 0
5
10
15
20
Dialysis
G e ne ral M e di c a re: C K D, DM
G e ne ral M e di c a re: C K D, NDM
G e ne ral M e di c a re: No C K D, DM
G e ne ral M e di c a re: No C K D, NDM
65-74 75-84 85+Exp. re
main
ing lifetim
e (
yrs
)
0
5
10
15
20
Dialysis
General Medicare: CKD, DM
General Medicare: CKD, NDM
General Medicare: No CKD, DM
General Medicare: No CKD, NDM
CKD, DM
CKD, NDM
Non-CKD, DM
Non-CKD, NDM
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Hospital admissions, by diagnosis, patient population, & diabetic status figure 6.12, prevalent patients age 65 & older, 1999
All-cause CV Infectious OtherAll-cause CV Infectious Other
Ad
mis
sio
ns p
er
1,0
00
pa
tie
nt
ye
ars
at
risk
0
500
1,000
1,500
2,000
2,500Diabetic Non-diabetic
General Medicare:
non-CKD
General Medicare:
CKD
Dialysis
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Cardiovascular admissions, by diagnosis, age, & patient population figure 6.10, prevalent patients age 65 & older, 1999
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Issues to be Addressed
Mortality and hospitalization rates among the Medicare CKD patients, though lower than among the ESRD population, are much higher than among non-CKD patients.
Preventive care is lagging in the Medicare CKD population, as it is among patients with ESRD.
Preventive care (in particular, influenza vaccination) has the potential to reduce both deaths and illness in both the CKD and ESRD populations.
US
RD
S
USRDS ASN 2002
Pre- & post-ESRD patients diabetic care: receiving lipid testing figure 5.8, patients 67+, by race/ethnicity
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Pre- & post-ESRD patients diabetic care: receiving glycosylated hemoglobin testing figure 5.9, patients 67+, by race/ethnicity
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Lipid monitoring in diabetic patients: general Medicare and ESRD patients figure 5.12, patients age 65–75, by HSA, unadjusted
Percent of patients
65.1+ (72.0)
60.4 to <65.1
56.4 to <60.4
42.9 to <56.4
below 42.9 (35.9)
ESRD patients General Medicare patients
Percent of patients
65.1+ (68.9)
60.4 to <65.1
56.4 to <60.4
42.9 to <56.4
below 42.9 (NA) USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Glycosylated hemoglobin (HbA1c) testing in diabetic patients: Medicare vs ESRD pts figure 5.14, patients age 65–75, by HSA, unadjusted
Percent of patients
78.4+ (81.1)
75.1 to <78.4
72.2 to <75.1
61.5 to <72.2
below 61.5 (51.5) USRDS 2002 ADR
ESRD patients General Medicare patients
Percent of patients
78.4+ (80.5)
75.1 to <78.4
72.2 to <75.1
61.5 to <72.2
below 61.5 (60.7)
US
RD
S
USRDS ASN 2002
Issues to be Addressed
Mortality and hospitalization rates among the Medicare CKD patients, though lower than among the ESRD population, are much higher than among non-CKD patients.
Preventive care is lagging in the Medicare CKD population, as it is among patients with ESRD.
Preventive care (in particular, influenza vaccination) has the potential to reduce both deaths and illness in both the CKD and ESRD populations.
US
RD
S
USRDS ASN 2002
Influenza Vaccine Delivery and Effectiveness in End-Stage Renal Disease*
Studied two influenza epidemic years: 1997, 1998.
Prevalent dialysis patients.
Billing data for influenza immunization sought in period from 1 September – 31 December.
Data for hospitalization and mortality sought in subsequent period from 1 January – 28 February, the CDC defined influenza seasons for ‘98 and ‘99.
All analyses corrected for baseline comorbidities existing prior to immunization period.
* Gilbertson DT et al, Kidney International, 2002, (In Press)
US
RD
S
USRDS ASN 2002
Influenza vaccination rate*: 97-98 & 98-99, By modality compared to HP 2000/2010
49%
40%
60%
90%
40%
49%
0.0%
20.0%
40.0%
60.0%
80.0%
100.0%
Hemodialysis Peritoneal
Dialysis
Healthy People
2000 Objective
Healthy People
2010 Objective
Perc
en
t
1997-1998
1998-1999
*Gilbertson, et al, Kidney International (in press), 2002
US
RD
S
USRDS ASN 2002
Influenza vaccination rate*: 98-99, By modality, age and race
0.0
20.0
40.0
60.0
80.0
100.0
< 18 18 - 39 40 - 64 65 +
0.0
20.0
40.0
60.0
80.0
100.0
White Black Native American Asian Other
HD
PD
HD
PD
Pe
rce
nt va
ccin
ate
d
*Gilbertson, et al, Kidney International (in press), 2002
Age
Race
US
RD
S
USRDS ASN 2002
Any-cause Hosp.
Influenza/ Pneumonia
Hosp.
Bacteremia/ Viremia/
Septicemia Hosp.
Respiratory Infection
Hosp.
Any-cause Death
Cardiac Death
Infectious Death
Other Death
Hosp. or Death
0.4
0
.6
0.8
1
.0
1.2
1
.4
1.6
Odds Ratios for Hospitalization and Mortality: Vaccinated vs. Not Vaccinated^
Odd
s R
atio
*
*Upper limit for Bacteremia/Viremia/Septicemia, PD = 2.01
Hemodialysis
Peritoneal Dialysis
1.6
1.4
1.2
1.0
0.8
0.6
0.4
Odds ratio for death Odds ratio for Hospitalization
Od
ds r
atio
^Gilbertson, et al, KI (in press), 2002
US
RD
S
USRDS ASN 2002
Pre- & post-ESRD patients (age 67+) care for influenza vaccinations figure 5.4, by age, & race/ethnicity, fall of pre-ESRD to fall of post-ESRD
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Percent of patients receiving influenza vaccinations figure 5.1, hemodialysis patients, by HSA, 1999 unadjusted
Percent of patients
66.3+ (71.3)
60.2 to <66.3
55.2 to <60.2
47.2 to <55.2
below 47.2 (37.6)
US
RD
S
USRDS ASN 2002
Issues to be Addressed
Mortality and hospitalization rates among the Medicare CKD patients, though lower than among the ESRD population, are much higher than among non-CKD patients.
Preventive care is lagging in the Medicare CKD population, as it is among patients with ESRD.
Preventive care (in particular, influenza vaccination) has the potential to reduce both deaths and illness in both the CKD and ESRD populations.
US
RD
S
USRDS ASN 2002
Comparing Medicare Costs for non-
CKD, CKD and dialysis Patients
John M. Brooks, PhD
Deputy Director, Economic Special Study Center
US
RD
S
USRDS ASN 2002
*Hebert et al,“Identifying Persons with Diabetes Using Medicare Claims Data”,
American Journal of Medical Quality 1999, 14:270-277
Non-Dialysis Sample Definitions
From Medicare 5% sample, excluding patients:
→ in HMO or with ESRD during 1997-1998
→ not continuously enrolled in Medicare Part A
and Part B during 1997-1998
→ less than 67 in 1999
• CKD: At least 1 inpatient or 2 outpatient CKD claims during 1997-1998 (*methodology)
non-CKD sample size: 1,066,608
CKD sample size: 38,781
US
RD
S
USRDS ASN 2002
Dialysis Sample Definition
From USRDS database, excluding patients
→ in HMO during 7/1/1998-12/31/1998
→ not continuously on dialysis from 7/1/1998
through 12/31/1998 after 90 days on initiation
of ESRD
→ less then 67 in 1999
Dialysis Sample: 61,679
US
RD
S
USRDS ASN 2002
Medicare Costing Methodology
Inpatient, outpatient, Part B, SNF, Hospice
CKD and non-CKD patients:
→ All costs from 1/1/99 to either death, end of
entitlement, ESRD initiation or 12/31/99.
Dialysis patients:
→ All costs from 1/1/99 to either death,
transplant, lost-to-follow-up, or 12/31/99.
US
RD
S
USRDS ASN 2002
Medicare Costing Methodology con’t
Estimated average per member per month (PMPM) total cost.
Divided PMPM non-dialysis cost into:
→ Reactive Costs: costs in response to a
medical condition
→ Proactive Costs: not reactive
US
RD
S
USRDS ASN 2002
Patient Characteristics by Patient Group
Non-CKD CKD Dialysis
Patients Patients Patients
Sample size 1,066,607 38,781 61,697
Age group (years) %
67-74 45.1 34.2 51.7
75-84 40.5 44.8 41.2
85 + 14.4 21.1 7.1
Gender %
Male 38.8 47.5 47.5
Female 61.2 52.5 52.5
Race %
White 89.3 83.5 61.8
African American 6.9 11.9 32.2
Other 3.8 4.6 6.0
US
RD
S
USRDS ASN 2002
97%
0.28%3.5%
Distribution of Medicare Expenditures & Population by Diagnosis Group: Age 67+
88%
9%
3%non-CKD
CKD
Dialysis
Medicare Expenditures: 1999 Medicare population: 97-98
US
RD
S
USRDS ASN 2002
505
1,373
5,223
0
1,000
2,000
3,000
4,000
5,000
Non-CKD Patients CKD Patients Dialysis Patients
Unadjusted Average Total PMPM Costs in
1999 by Patient Group
US
RD
S
USRDS ASN 2002
Non-CKD CKD Dialysis
Patients Patients Patients
Sample size (5%) 1,066,608 38,781 61,697(100%)
Age group (years)
67-74 405 1,328 5,270
75-84 553 1,402 5,178
85 + 694 1,385 5,126
Gender
Male 533 1,388 5,052
Female 487 1,359 5,377
Race
White 501 1,317 5,092
African American 559 1,689 5,492
Other 491 1,576 5,064
PMPM Total Cost by Group and
Demographics in 1999
US
RD
S
USRDS ASN 2002
% Effect on
Risk Factor PMPM Coefficient SE P-value
Intercept 4.193 0.005 <.0001
Age 67-74: Reference
Age 75-84 +80.3% 0.589 0.006 <.0001
Age 85 + +160% 0.957 0.008 <.0001
Female: Reference
Male -19.3% -0.214 0.005 <.0001
White: Reference
Black -38.1% -0.479 0.010 <.0001
Other race -35.9% -0.445 0.014 <.0001
Non-CKD: Reference
CKD +467% 1.74 0.014 <.0001
Multiple Regression Estimates on the Log of
PMPM Total Cost in 1999
US
RD
S
USRDS ASN 2002
Adjusted Comparison of CKD and non-CKD costs
3.63.2
2.2
4.6
3.2
2.2
4.8
3.22.8
5.9
3.8
2.5
0.0
1.0
2.0
3.0
4.0
5.0
6.0
7.0
67-74 75-84 85+ 67-74 75-84 85+
Re
lati
ve
co
st
(CK
D/n
on
-CK
D)
White
Black
Male Female
US
RD
S
USRDS ASN 2002
0
500
1000
1500
2000
2500
3000
Non-CKDPatients
CKD Patients Dialysis Patients
Av
era
ge
PM
PM
Co
sts
Proactive
Reactive
Non-Dialysis Reactive and Proactive
Medicare PMPM Costs by Group in 1999
71%
($359)
29%
($145) ↓
19%
($253) ↓
81%
($1,080)
85%
($2,352)
15%
($429)
$504
$1,333
$2,781
US
RD
S
USRDS ASN 2002
Effect of pro-active treatment on costs: Influenza vaccination in dialysis pts, age 67+
1,530
1,766
1,400
1,450
1,500
1,550
1,600
1,650
1,700
1,750
1,800
Vaccinated not vaccinated
$236 PMPM
Average monthly Hospitalization costs
in the first 4 months of 1999
Vaccinated in the fall 1998
(n=33,164) (n=28,533)
Me
an
IP
PM
PM
1-1
to
4-3
0-9
9
US
RD
S
USRDS ASN 2002
Assuming $50 vaccination, savings per patient:
(236 x 4) – 50 = $894
Total potential cost savings during influenza season:
28,533 x $894 = $25,508,502
Flu Vaccination Medicare Cost
Implications
US
RD
S
USRDS ASN 2002
Summary of the economic impact of CKD
ESRD patients age 67+ represent 0.3% of the Medicare population and consume 10 times their proportion of expenditures.
CKD patients age 67+ represent 3.5% of the Medicare population and consume 9% of the Medicare budget.
CKD patients are 4.5-6 times more expensive than the non-CKD general Medicare population.
Influenza vaccination appears to be a cost effective preventive health care measure and should be recommended for ALL ESRD patients!
US
RD
S
USRDS ASN 2002
Major issues in CKD
Given the burden of disease and high rates of heart failure and death, how frequently do nephrologists see this vulnerable population to guide care?
US
RD
S
USRDS ASN 2002
Life table estimates for probability of first outpatient nephrologist visit during 1998 CKD patients 1996-’97, alive thru 1998, by outcome in 1999
Follow-up time (month)
Pro
ba
bili
ty o
f vis
it
0.00
0.10
0.20
0.30
0.40
0.50
0 1 2 3 4 5 6 7 8 9 10 11 12
ESRD (N=617)
All (N=27,974)
Alive (N=23,006)
Died (N=4,351)
US
RD
S
USRDS ASN 2002
Overall Summary
There are 12.6 times as many patients with CKD as patients with ESRD in the Medicare system.
CKD patients are less sick than the patients with ESRD, but they are much sicker than the non-CKD patients, and their care is disproportionately expensive.
CKD patients are about 5 – 10 times more likely to die than to reach ESRD. Nephrologists see them late if at all.
They receive less preventive care than is recommended.
US
RD
S
USRDS ASN 2002
Overall Conclusions
• There appears to be a self-fulfilling pessimism about the care of patients with CKD (as for patients with ESRD), who receive less than recommended proactive preventive care.
We need to determine which aspects of traditional preventive care are effective in this population and do more to apply the aspects that are effective.
The nephrology community needs to educate and to work with the larger medical community to address the unmet needs of the very large population with CKD short of ESRD.
US
RD
S
USRDS ASN 2002
US
RD
S
USRDS ASN 2002
US
RD
S
USRDS ASN 2002
Adjusted relative risks for incident CVD and CHF during 1998 between CKD and non-CKD patients Non-CVD or non-CHF patients 1996-’97, alive thru 1998
Rela
tive
ris
k
2.25
1.61
0.0
0.5
1.0
1.5
2.0
2.5
CVD CHF
N=559,736 N=895,771 Study population:
Reference: Non-CKD patients
Adjusted for age, gender, and race
US
RD
S
USRDS ASN 2002
Effect of Proactive Treatment on Costs:
Flu Vaccinations for Dialysis Patients
Inpatient PMPM Costs in first 4 months of 1999
Dialysis patients with/without flu vaccination in fall of 1998
Number PMPM
With Vaccination 33,164 $1,530
Without Vaccination 28,533 $1,766
Vaccinated patients had $236 lower Medicare inpatient costs per month.
US
RD
S
USRDS ASN 2002
All-cause mortality in the general Medicare & dialysis populations, patients age 65+, 1999, unadjusted figure 9.4, period prevalent general Medicare & period prevalent dialysis pts
No chronic kidney disease Chronic kidney disease Dialysis
Death
s per
1,0
00patientyears
at risk
0
100
200
300
400
Cardiovascular disease
No cardiovascular disease
Patients by cardiovascular disease
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Expected remaining lifetime in patients with increasing morbidity, by age figure 9.26, diabetes, chronic kidney disease, & chronic heart
Exp. remaining lifetime (yrs) 65-74 75-84 85+ 0
4
8
12
16
Dialysis
General Medicare: DM, CKD, CHF
General Medicare: NDM, CKD, CHF
General Medicare: DM, CKD
General Medicare: DM only
Exp. re
main
ing lifetim
e (
yrs
)
65-74 75-84 85+0
4
8
12
16Dialysis
General Medicare: DM, CKD, CHF
General Medicare: NDM, CKD, CHF
General Medicare: DM, CKD
General Medicare: DM only
USRDS 2002 ADR
US
RD
S
USRDS ASN 2002
Number of patients with a pre-ESRD outpatient nephrologist visit
No visit
50.4%
Nephrology
visit
49.6%N = 55,087 N = 54,234
1995-1998
OP Nephrology visits within two years of ESRD
US
RD
S
USRDS ASN 2002
Cumulative percent of patients with a 1st outpatient nephrologist visit pre-ESRD: age 67+
0
20
40
60
80
100
-24 -22 -20 -18 -16 -14 -12 -10 -8 -6 -4 -2
Months from the first ESRD service date
Cu
mu
lati
ve p
erc
en
t o
f p
ati
en
ts
Pts with OP nephrology visits= 54,234
First service month
US
RD
S
USRDS ASN 2002
Life table estimates for probability of first outpatient nephrologist visit during 1998 CKD patients 1996-’97, alive thru 1998, by outcome in 1999
Follow-up time (month)
Pro
ba
bili
ty o
f vis
it
0.00
0.10
0.20
0.30
0.40
0.50
0 1 2 3 4 5 6 7 8 9 10 11 12
ESRD (N=617)
All (N=27,974)
Alive (N=23,006)
Died (N=4,351)
US
RD
S
USRDS ASN 2002
Cumulative Comorbidity Dx in Medicare Patients Advancing to ESRD: Pts age 67+, two years before ESRD incidence
0
10
20
30
40
50
60
70
80
-24 -21 -18 -15 -12 -9 -6 -3 0 3
ASHD
CHF
PVD
CVATIA
CARDO
ESRD 1st service month
Time in Months
Pre-ESRD period ESRD period
US
RD
S
USRDS ASN 2002
Pneumonia vaccination rates in the two years pre-ESRD figure 5.6, ESRD patients 67+, by HSA, unadjusted
Map: Pneumoncoccal vaccination
US
RD
S
USRDS ASN 2002
Life table estimates for probability of incident CVD during 1998 by CKD and patient groups Non-CVD patients 1996-’97 and did not die or develop ESRD/CKD in 1998
CKD patients*
Follow-up time (month)
Pro
babili
ty o
f C
VD
Non-CKD patients*
0.00
0.10
0.20
0.30
0.40
0 1 2 3 4 5 6 7 8 9 101112
ESRD (N=124)
Died (N=422)
All (N=5,712)
No events (N=5,166)
0.00
0.10
0.20
0.30
0.40
0 1 2 3 4 5 6 7 8 9 101112
CKD/ESRD (N=8,167)
Died (N=17,017)
All (N=554,024)
No events (N=528,840)
*All Log-ranks test for differences
among patient groups: p<.0001
0.083
0.14
RR: 1.61 for new CVD in CKD
US
RD
S
USRDS ASN 2002
Adjusted relative risks for incident CVD and CHF during 1998 between CKD and non-CKD patients Non-CVD or non-CHF patients 1996-’97, alive thru 1998
Rela
tive
ris
k
2.25
1.61
0.0
0.5
1.0
1.5
2.0
2.5
CVD CHF
N=559,736 N=895,771 Study population:
Reference: Non-CKD patients
Adjusted for age, gender, and race
US
RD
S
USRDS ASN 2002
ESRD patients, by diagnosis & origins in the general Medicare population figure p.18, 1997-1998 general Medicare patients
All patients
Percent of patients
0 102 03 0 405 06 07 08 09 0 100
ESRD ptsin followup
period
Gen. Med.pts during
entry period
DM/CKD/CHF
DM/CKD
CHF/CKD
CKD only
DM/CHF
CHF only
DM only
None
4.4%
67%
USRDS 2002 ADR