Hospitals, blood collection centers, cord blood banks

ATTACHMENT C

AABB conducted the 2005 Nationwide Blood Collection and
Utilization Survey under United States Department of Health
and Human Services Contract HHSP22320042202TE.
Project Directors:
Barbee I. Whitaker, PhD
Project Director for AABB
8101 Glenbrook Road
Bethesda, MD 20814
301.215.6574
[email protected]
LCDR Richard Henry, USPHS
Project Director for HHS
US Department of Health and Human Services
1101 Wootton Parkway
Tower Building, Suite 250
Rockville, MD 20852
[email protected]
Report Authors:
Barbee I. Whitaker, PhD, and Marian Sullivan, MS, MPH

Table of Contents
List of Tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
1. Executive Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Important Trends in the US Blood Supply . . . . . . . . . . . . . . 1
Blood Collection . . . . . . . . . . . . . . . . . . . . . . . . . . 1
Blood Transfusion . . . . . . . . . . . . . . . . . . . . . . . . . . 2
2. Key Findings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
3. Introduction and Methods. . . . . . . . . . . . . . . . . . . . . . . . . . 5
Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Data Collection . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Sampling Frame . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
Sample Selection . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Response Rates . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
Edit and Imputation Procedures . . . . . . . . . . . . . . . . . . . 6
Sampling Weights . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Survey Respondents . . . . . . . . . . . . . . . . . . . . . . . . . 9
Characterization of Respondents . . . . . . . . . . . . . . . . . . 10
Blood Center Nonrespondents . . . . . . . . . . . . . . . . . . . 12
Limitations of the Survey . . . . . . . . . . . . . . . . . . . . . . 12
4. Blood Collected and Processed in the United States . . . . . . . . . . . . 13
Total Collections . . . . . . . . . . . . . . . . . . .
Whole Blood Collections. . . . . . . . . . . . . . .
Red Blood Cells Collected by Apheresis . . . . . . .
Whole Blood and Red Blood Cell Units Released for
Distribution . . . . . . . . . . . . . . . . . . . .
Non-Red-Blood-Cell Components Processed . . . . .

. . . . . . . 13
. . . . . . . 13
. . . . . . . 13
. . . . . . . 15
. . . . . . . 15

5. Blood Transfused in the United States . . . . . . . . . . . . . . . . . . . 19
Whole Blood and Red Blood Cells Transfused
Whole Blood and Red Blood Cell Recipients .
Non-Red-Blood-Cell Components Transfused
Platelet Dosage. . . . . . . . . . . . . . . .
Outdated Units. . . . . . . . . . . . . . . .
Intraoperative Autologous Blood Recovery . .

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

19
19
20
21
22
24

iii

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

6. Component Modification . . . . . . . . . . . . . . . . . . . . . . . . . 25
Irradiation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
Leukocyte Reduction . . . . . . . . . . . . . . . . . . . . . . . . 25
Transfusion of Irradiated and Leukocyte-Reduced Components. . . 28
7. Current Issues in Blood Collection and Screening . . . . . . . . . . . . . 29
Screening Test Losses . . . . . . . . . . . . . . . . . . . . . . . 29
Therapeutic Phlebotomy . . . . . . . . . . . . . . . . . . . . . . 29
Disaster Planning. . . . . . . . . . . . . . . . . . . . . . . . . . 29
8. Current Issues in Blood Transfusion . . . . . . . . . . . . . . . . . . . . 31
Blood Inventory Shortages
Crossmatch Procedures. .
RBC Unit Age . . . . . .
Platelet Product Age . . .
Adverse Reactions . . . .

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

.
.
.
.
.

31
32
32
32
34

9. Component Costs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
Red Blood Cells . . . . . . . .
Fresh Frozen Plasma . . . . . .
Whole-Blood-Derived Platelets.
Apheresis Platelets . . . . . . .
Reimbursement . . . . . . . .
Summary . . . . . . . . . . . .

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

.
.
.
.
.
.

35
35
39
39
39
40

10. Cellular Therapy Products . . . . . . . . . . . . . . . . . . . . . . . . 41
Collections . . . . . . . . . . . . . . .
Processing . . . . . . . . . . . . . . .
Infusion . . . . . . . . . . . . . . . .
Characterization of Reporting Facilities.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

.
.
.
.

41
44
46
46

11. Historical Perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Time Trends . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51
Blood Supply Adequacy . . . . . . . . . . . . . . . . . . . . . . 51
US Population Trends . . . . . . . . . . . . . . . . . . . . . . . 54
12. Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . 55
13. References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

iv

List of Tables

Table 3-1.

Sampling Frame Counts and Sampling Rates . . . . . . . . . . . 7

Table 3-2.

Response Rates by Type of Facility and Surgical Volume . . . . . 7

Table 3-3.

Raw Sampling Weights. . . . . . . . . . . . . . . . . . . . . . 8

Table 3-4.

Weighting Class Adjustments . . . . . . . . . . . . . . . . . . 9

Table 3-5.

Final Sampling Weights. . . . . . . . . . . . . . . . . . . . . 10

Table 3-6.

United States Public Health Service Regions . . . . . . . . . . 11

Table 4-1.

Estimated 2004 Activities of US Blood Centers and
Hospitals for Whole Blood (WB) and Red Blood Cells
(RBCs) (expressed in thousands of units) . . . . . . . . . . . . 14

Table 4-2.

WB/RBC Units Released for Distribution by Blood Group
and Type by Blood Centers and Collecting Hospitals . . . . . . 15

Table 4-3.

Estimated 2004 Activities of US Blood Centers and
Hospitals for Non-Red-Blood-Cell Components
(expressed in thousands of units) . . . . . . . . . . . . . . . . 16

Table 5-1.

Outdated Components as a Percentage of the Total
Volume of Each Type Processed for Transfusion in 2004 . . . . 23

Table 6-1.

Blood Components Modified by Irradiation in All Facilities
in 2004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25

Table 6-2.

Blood Components Modified by Prestorage and Poststorage
Leukocyte Reduction in All Facilities in 2004 . . . . . . . . . . 26

Table 6-3.

Number of Irradiated and Leukocyte-Reduced Blood
Component Units Processed in All Facilities in 2004 and
2001 (expressed in thousands of units) . . . . . . . . . . . . . 26

Table 6-4.

Estimated Number of Blood Component Units Modified by
Irradiation or Leukocyte Reduction and Transfused by All
Facilities in 2004 . . . . . . . . . . . . . . . . . . . . . . . . 27

v

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 6-5.

Number of Irradiated and Leukocyte-Reduced Component
Units Transfused in 2004 and 2001 (expressed in thousands
of units). . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28

Table 8-1.

Cancellation of Elective Surgeries by US Hospitals, 19972004 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31

Table 9-1.

Mean Dollars Paid per Selected Component Unit by
Hospitals in 2004 vs 2001 . . . . . . . . . . . . . . . . . . . . . . 36

Table 9-2.

Average Component Cost by USPHS Region . . . . . . . . . . 37

Table 9-3.

Average Component Cost by Surgical Volume . . . . . . . . . 38

Table 9-4.

CMS Hospital Outpatient Prospective Payment
System Rates for Selected Blood Components. . . . . . . . . . 39

Table 10-1.

Autologous Cellular Therapy Product Collections Performed . . 42

Table 10-2.

Allogeneic Cellular Therapy Product Collections Performed . . 43

Table 10-3.

Cellular Therapy Products Processed . . . . . . . . . . . . . . 45

Table 10-4.

Cellular Therapy Products Issued and/or Infused . . . . . . . . 47

Table 10-5.

Cellular Therapy Product Collections by Hospitals, by
Surgical Volume . . . . . . . . . . . . . . . . . . . . . . . . 49

vi

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

1. Executive Summary

The Department of Health
and Human Services (DHHS)
and AABB conducted the
fourth nationwide survey of
blood services and related activities in July 2005. (Earlier
surveys were conducted in
2002, 2000, and 1998.) The
purpose of the survey (a
12-page questionnaire) was to
assess the amount of blood
collected and transfused in
the United States (US) in 2004
and to provide data to assist
the DHHS in validating its
Blood Availability and Safety
Information System. The facilities surveyed included all
non-hospital-based blood collection centers (blood centers), a sample of hospitals
from the American Hospital
Association database, and, for
the first time, a sample of cord
blood banks.

sentative and to develop
weighted national estimates.

Blood collection per thousand
US population of donor age
(18-65 years) was 85.6 in
2004 compared to 88.0 in
Important Trends in the
2001. This was a decrease of
US Blood Supply
2.7% from the 2001 rate, but
still a larger percentage of the
The supply of screened
population than was reported
allogeneic whole blood (WB) in 1999 (80.8/1,000). This
and red blood cell (RBC)
suggests that donor centers
units that passed all tests
have been able to retain many
(14,560,000) exceeded trans- of the donors who volunfusions of allogeneic
teered after the September 11,
WB/RBCs (13,912,000) by a
2001 terrorist attacks on New
margin of 648,000 units. The York City and Washington,
blood supply was sufficient to DC. The US WB/RBC transfumeet transfusion needs. These sion rate in 2004 was 49.6
data, combined with the low- units per thousand population,
est rate of units outdated in re- statistically unchanged from
cent years, suggest that hospi- 2001. This was a welcome
tals and blood centers have
flattening of the seemingly
become more efficient at de- ever-increasing transfusion
livering the appropriate prod- rate curve for the aging US
uct when needed. Although
population (Figure 11-2).
most hospitals reported few
delays or shortages, when
shortages or delays occurred, Blood Collection
Response rates to the 2005
Nationwide Blood Collection the episodes lasted longer.
Among hospitals reporting un- The NBCUS indicates that the
and Utilization Survey
total supply of WB/RBC units
met blood needs, the mean
(NBCUS) were as follows:
in 2004 was 15,288,000 units
number of days of unmet
blood centers 92.3%
before testing. This is only
nonsurgical
blood
need
in(131/142), hospitals 56.8%
(1,604/2,825), and cord blood creased significantly from 2.1 marginally fewer units (0.2%)
banks 33.3% (3/9). Statistical days in 2001 to 19.27 days in than were reported in 2001,
when increased collections
procedures were used to ver- 2004 (p<0.001).
were attributed to altruistic
ify that the sample was repre-

Executive Summary

1

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

The composition of the source
of the blood supply in 2004
was different from that of previous years. In 2001, traditional allogeneic collections
were responsible for the high
number of units collected. In
2004, these numbers remained high, but because of
the increased use of red cell
apheresis technology in the
face of reduced (and, in the
case of autologous and directed donations, significantly
reduced) traditional blood
collections. Red cell apheresis
collections increased 202%
over 2001, yielding approximately 824,000 RBC units.
Figure 1-1 illustrates the
growth in use of red cell
apheresis.

15000

12000
Thousands of Units

donation after the September
11, 2001 terrorist attacks.
Blood centers were responsible for the collection of
14,305,000 units or 93.6% of
the supply; hospitals collected
983,000 units or 6.4% of the
total.

9000

6000

3000

0
1999

2001

Traditional Allogeneic Collections

2004

RBC Apheresis Collections

Figure 1-1. Growth in use of RBC apheresis technology.

Blood Transfusion

not significantly different from
the 2001 total. However, the
The number of WB/RBCs
proportion of transfused platetransfused in 2004 totaled
lets that were collected by
14,182,000 units—a small,
apheresis increased 10% from
but not statistically significant, 7,582,000 to 8,338,000
increase of 2% over 2001 to- platelet concentrate equivatals. Allogeneic units translent units, representing 84.4%
fused (including the pediatric of the total platelets transNote: Red cell apheresis refers
contribution expressed as
fused. The transfusion of
to the use of a machine to coladult equivalent units)
whole-blood-derived platelet
lect approximately 2 RBC
equaled 13,912,000.
concentrates continued a
units in one collection procedownward trend, decreasing
dure instead of extracting the
The total number of platelet
41.2% from 2,614,000 units
cells from a single unit of
units provided to patients in
in 2001 to 1,537,000 units in
whole blood.
2004 was 9,875,000; this was 2004 (p<0.0001).

2

Executive Summary

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

2. Key Findings

The results of The 2005 Nationwide Blood Collection
and Utilization Survey
(NBCUS) provide an update
of United States (US) blood
services and related activities
assessed by the three previous
nationwide surveys conducted in 2002, 2000, and
1998. Notable findings from
the 2005 NBCUS and comparisons with the 2002 survey
results were as follows:
• Allogeneic collections remained statistically the
same at 14.8 million units.
• Total WB/RBC transfusions
remained statistically the
same at 14.2 million units.
• Autologous collections declined significantly by 26%
to 458,000 units.
• Red blood cell apheresis
collections increased 202%
for a total of 824,000 units.
• The total number of all
components transfused in
2004 was 29,038,000.
• The total number of transfused units that were irradiated increased by 15.8%
while those that were
leukocyte-reduced increased by 2.7%.

• Transfusion of prestorage
leukocyte-reduced units increased by 7.9%, while the
transfusion of poststorage
leukocyte-reduced units decreased by 45.3%.
• There was a significant
(54%) decline in HPC-M
transplant procedures performed.
• There was a 42.9% decrease in the number of
outdated whole blood (WB)
and red blood cell (RBC)
units.
• The transfusion of wholeblood-derived platelet concentrates continued to decline (–41.2%), while the
use of apheresis platelets
increased by 10%. Overall,
transfusion of platelets remained essentially the same
at 9.9 million units.
• Almost all blood centers
(99.2%) and hospitals
(89.6%) have emergency
preparedness plans for
blood services.
• A total of 135 hospitals reported the cancellation of
elective surgery on one or
more days because of
blood inventory shortages.
This affected 546 patients,

•
•

•

•

•

significantly fewer than in
2001.
The mean age of RBC units
at transfusion was 17 days.
The mean age of wholeblood-derived platelet units
was 3.16 days at transfusion; for an apheresis
platelet unit, the mean was
3.08 days.
The average cost of RBC
units increased by 30.8%.
A statistically significant increase in the cost since
2001 was measured for all
four components assessed.
The WB/RBC collection
rate per thousand US donor
population was 85.6 units
in 2004, compared to 88.0
in 2001. The WB/RBC
transfusion rate remained
the same at 49.6 units per
thousand US population
overall.
The margin between
test-negative allogeneic
WB/RBC units collected
and those transfused in the
US in 2004 was 648,000
units, compared to approximately 920,000 in 2001.

Key Findings

3

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

3. Introduction and Methods

This report presents the results
of the fourth Nationwide
Blood Collection and Utilization Survey (NBCUS), conducted in 2005. The survey
was funded by the Department of Health and Human
Services and conducted by
AABB (Bethesda, MD). Previous surveys were conducted
in 1998, 2000, and 2002 by
the National Blood Data Resource Center, a whollyowned subsidiary of AABB.

Objectives
The objectives of the survey
were to generate national estimates for blood services activities in the United States (US)
in 2004; to provide comparisons with previous measurements; to provide data to validate a real-time national
blood supply and utilization
monitoring system; and to
characterize business practices in the blood collection,
transfusion medicine, and cellular therapy community.
The survey instrument was
designed to capture quantitative data regarding blood col-

lection, processing, transfusion, and final disposition, as
well as other information describing current policies and
practices, and the adoption of
new technologies by the
blood community.

Park, NC). Validation of survey data was achieved by
comparison with 2001 survey
data and by direct contact
with individual respondents as
necessary.

Sampling Frame
Data Collection
The 12-page questionnaire,
cover letter, and postage-paid
return envelope were mailed
in July 2005 to the 2,976
facilities in the sample described below. Survey packets
were addressed to the director
of either the blood center or
the hospital blood bank or
transfusion service. A second
mailing of the survey questionnaire was sent to 2,492
nonresponders six weeks afterwards. Those facilities that
did not respond by the initial
September deadline were
contacted by e-mail, telephone, or both. Data collection concluded on October
31, 2005.

The sampling frame for the
2005 NBCUS was compiled
from two data sources. The
first source was the AABB database, which is a list of all
non-hospital-based blood collecting facilities in the 50
states. This list also contains
hospitals that are members of
AABB. The second source is
the American Hospital Association (AHA) database.

To construct the sampling
frame from the AHA database,
several criteria were used.
Only those hospitals providing general medical and surgical service, children’s general
medical and surgical service,
and obstetrics and gynecology
services were considered eliData coding, keying, and veri- gible population members.
fication were performed by
Two new service categories,
Research Triangle Internaoncology and cardiology,
tional (RTI, Research Triangle were investigated and were

Introduction and Methods

5

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

included in the frame if the
hospitals responded to the
2002 AHA hospital survey. To
ensure comparability with
previous surveys of domestic
institutions, the following
types of treatment facilities
were excluded from the AHA
sampling frame: military and
federal facilities, such as Department of Justice facilities,
Indian Health Service facilities, and non-US facilities.
Veterans Affairs treatment facilities were included in the
sampling frame. Finally, hospitals reporting fewer than
100 inpatient surgeries per
year were also excluded.

hospitals, and cord blood
banks.

Sample Selection

The NBCUS sampling frame
consisted of all the blood centers on the AABB list and all
eligible hospitals on the AABB
list. This initial list of treatment
facilities was supplemented
with hospitals from the AHA
database to ensure a total
sample of 3,000 facilities. The
AHA hospital database was
stratified into six categories
according to annual inpatient
surgical volume. The strata
are defined as follows: 100999 surgeries, 1,000-1,399
To prepare the AABB datasurgeries, 1,400-2,399 surgerbase for sampling, hospitals
ies, 2,400-4,999 surgeries,
were matched to the AHA
database and the AHA identi- 5,000-7,999 surgeries, and
8,000 or more surgeries. Hosfication numbers were aspitals in all strata with more
signed to avoid duplication.
than 2,400 surgeries were
All US hospitals with 100 or
sampled at a rate of 100%. If
more annual surgeries included on the AABB member the treatment facility was a
list were included in the sam- nonrespondent to the 2002
AHA survey and the inpatient
pling frame, regardless of
surgeries were estimated (due
whether they met exclusion
criteria that were used for the to nonresponse) to be fewer
AHA database, because it was than 2,400, then these hospitals were excluded from the
established that these hospisample, but remained as eligitals collect and/or transfuse
blood. Within the AABB data- ble population members.
base, the facilities were cate- Nonrespondents with the
higher surgical volumes were
gorized into four groups:
blood collection centers, hos- kept in the sample. After these
exclusions, the strata for
pitals that collect and trans1,000-1,399 surgeries and
fuse blood, hospitals that
transfuse blood only, and cord 1,400-2,399 surgeries were
sampled at rates of 86% and
blood banks. The final list of
eligible facilities contained a 88%, respectively. The number of hospitals selected in the
total of 4,202 blood centers,

6

Introduction and Methods

100-999 stratum was equal to
the sample size needed to result in a total sample of just
over 3,000. Table 3-1 shows
the sampling frame counts
and the sampling rates by
strata.

Response Rates
Table 3-2 summarizes the
outcome of data collection
efforts—collectively for blood
centers and by surgical volume for hospitals. After eliminating ineligible institutions
(42 hospitals that had ceased
operations, changed status, or
chosen to have an affiliate
make the report), the combined response rate was
58.4%.The response rate for
blood centers was 92.3%
(131/142). The response rate
for sampled hospitals was
56.8% (1,604/2,825). Response rates by surgical volume classes ranged between
50.8% and 66.6%. The total
number of individual survey
responses from the hospital
sample was 1,604.

Edit and Imputation
Procedures
RTI ensured the internal consistency of the 2005 NBCUS
data by developing machine
edit specifications to check
the logic and internal consistency of the data before imputation. Most missing data were
imputed using a model-based

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 3-1. Sampling Frame Counts and Sampling Rates
Sample

Sampling
Rate (%)

1,564

510

32.6

1,000–1,399 surgeries/year

365

314

86.0

1,400–2,399 surgeries/year

682

603

88.4

2,400–4,999 surgeries/year

854

854

100.0

5,000–7,999 surgeries/year

337

337

100.0

≥8,000 surgeries/year

249

249

100.0

142

142

100.0

9

9

100.0

4,202

3,018

71.8

Type of Facility

Total
Population

Hospitals
100–999 surgeries/year

Blood Centers
Cord Blood Banks
Total Facilities

Table 3-2. Response Rates by Type of Facility and Surgical Volume
Type of Facility

No.
Eligible

Hospitals

2,825

1,604

56.8

100–999 surgeries/year

502

255

50.8

1,000–1,399 surgeries/year

312

163

52.2

1,400–2,399 surgeries/year

599

329

54.9

2,400–4,999 surgeries/year

841

483

57.4

5,000–7,999 surgeries/year

332

221

66.6

≥8,000 surgeries/year

239

153

64.0

142

131

92.3

9

3

33.3

2,976

1,738

58.4

Blood Centers
Cord Blood Banks
Total Facilities

Respondents

Response
Rate (%)

Introduction and Methods

7

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 3-3. Raw Sampling Weights
Type of Facility

No. in
Stratum

No. Sampled
in Stratum

Raw Sampling
Weight

Hospitals
100–999 surgeries/year

1,564

510

3.0647

1,000–1,399 surgeries/year

365

314

1.1624

1,400–2,399 surgeries/year

682

603

1.1310

2,400–4,999 surgeries/year

854

854

1.0000

5,000–7,999 surgeries/year

337

337

1.0000

≥8,000 surgeries/year

249

249

1.0000

145

145

1.0000

9

9

1.0000

4,051

3,022

Blood Centers
Cord Blood Banks
Total Facilities

regression method for continuous variables for both hospitals and blood centers. However, for nonresponding blood
centers, key variables were
imputed using data from the
prior NBCUS survey (conducted in 2002) and the Directory of Community Blood
Centers (published by AABB
in 2003) in lieu of regression
imputation. The imputation
models were developed separately for blood centers and
hospitals and included region
and transfusion volume and/or
collection volume strata, as
appropriate. Imputed cases
were flagged to allow the analyst to identify which cases
were imputed.

8

Introduction and Methods

Sampling Weights

in the denominator because
they represent other, unidentiThe final sampling weights for fied ineligible units in the
hospitals were calculated for sampling frame. If these inelieach stratum using a threegibles were removed from the
stage process. In the first
raw weight calculation, resultstage, a raw weight was com- ing data estimates would be
puted as the reciprocal of the overstated. Table 3-3 shows
selection probability for each the number in each stratum,
stratum, to adjust for differthe number sampled, and the
ences in the sampling rates
results of the raw weight calapplied to the strata. The raw culation.
weight for hospitals was calculated as follows:
In the second stage, a weighting class adjustment (WCA)
Raw Weight =
was derived to correct for imNumber in Surgical
balance among the strata due
Volume Stratum
Number Sampled in
to different response rates
Surgical Volume Stratum
from the hospitals in the sample. The numerical adjustment
The “number sampled” in the was computed as the reciprodenominator includes all units cal of the response rate in the
sampled, including those de- stratum. This WCA was calcutermined to be ineligible.
lated as follows:
These ineligible units remain

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

WCA =
Sum of the Raw Weights
for All Sampled Members
of the Weighting Class
Sum of the Raw Weights
for All Responding Members
of the Weighting Class

1.000 was applied to the
blood centers to obtain national estimates, even though
their WCA was calculated to
be 1.0840. Given the high response rate for blood centers,
the WCA was considered to
Table 3-4 displays the number be negligible.
of eligible facilities, the number of responding facilities,
and the computed WCA for
Survey Respondents
each stratum. Only eligible
sample members were inTable 3-2 summarizes the
cluded in this calculation.
outcome of data collection
efforts—collectively for blood
The final sampling weight was centers and by surgical volthen calculated as the product ume for hospitals. The indiof the raw weight and the
vidual sites (regional blood
WCA. The final sampling
centers) of large collectors
weights appear in Table 3-5
such as the American Red
together with the raw samCross Blood Services and
pling weights from Table 3-3 United Blood Services were
and the WCA from Table 3-4. enumerated separately. The
A final sampling weight of
combined survey response

rate was 58.4%, as stated
above. The response rate for
blood centers was 92.3%
(131/142). The overall response from eligible sampled
hospitals that reported for
themselves was 56.8%
(1,604/2,825). Response rates
by surgical volume classes
ranged from 50.8% to 66.6%.
An additional 60 medical
treatment facilities submitted
data through other reporting
institutions (35), a common
practice among centralized
transfusion services and hospital affiliates. Of these, 18
were in the original sample
and 42 were not. Three of the
nine (33.3%) facilities classified as public cord blood
banks also responded. Therefore, the actual number of facilities (including blood cen-

Table 3-4. Weighting Class Adjustments

Type of Facility

No. Eligible in
Weighting Class

No. Eligible
and Responded
in Weighting
Class

Weighting
Class
Adjustment

Hospitals
100–999 surgeries/year

502

255

1.9272

1,000–1,399 surgeries/year

312

163

1.9202

1,400–2,399 surgeries/year

599

329

1.8207

2,400–4,999 surgeries/year

841

483

1.7412

5,000–7,999 surgeries/year

332

221

1.5023

≥8,000 surgeries/year

239

153

1.5621

142

131

1.0840

9

3

3.0000

2,976

1,738

Blood Centers
Cord Blood Banks
Total Facilities

Introduction and Methods

9

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 3-5. Final Sampling Weights

Raw Weight

Weighting
Class
Adjustment

Final
Sampling
Weight

100–999 surgeries/year

3.0647

1.9272

5.9063

1,000–1,399 surgeries/year

1.1624

1.9202

2.2321

1,400–2,399 surgeries/year

1.1310

1.8207

2.0592

2,400–4,999 surgeries/year

1.0000

1.7412

1.7412

5,000–7,999 surgeries/year

1.0000

1.5023

1.5023

≥8,000 surgeries/year

1.0000

1.5621

1.5621

Cord Blood Banks

1.0000

3.0000

3.0000

Blood Centers

1.0000

1.0840

1.0000

Type of Facility
Hospitals

ters, hospitals, and cord blood cessing was 138. Of these,
banks) for which survey data 129 self-reported or centrally
were received was 1,798.
reported for each center individually. Data for two centers
The 2005 response rates for
were aggregated with those of
blood centers and for hospitwo other centers that had astals overall were each within sumed ownership of them.
1% of the 2002 survey rates
Data for nine additional blood
and, thus, were not significenters were imputed from
cantly different. However, the previous surveys in which
actual number of hospitals
those centers had particithat responded to the current pated.
survey was greater than in
2002 by 298 (1,604 vs 1,306).
The additional facilities were Characterization of
Respondents
distributed across all geographic regions.
The majority of blood centers
Figure 3-1 illustrates the distri- self-identified as such. However, 21 centers selected the
bution of responding blood
centralized transfusion service
centers and hospitals among
the 10 geographic regions as option to describe themselves.
This option, added to the surdefined by the US Public
Health Service (USPHS). The vey categories for the first time
total number of blood centers this year, reads “A local or reincluded in the national esti- gional blood center that colmates for collection and pro- lects blood from donors and

10

Introduction and Methods

supplies blood, components,
and crossmatched blood
products to participating facilities (such as a centralized
transfusion service).”
Another classification option
added to the current survey
reads “An independent facility
that collects, processes, manufactures, stores, or distributes
cellular therapy products.”
Three cord blood banks that
are members of AABB selected this option. Hospital respondents for the most part
self-classified as either a transfusion service (1,318) or a
hospital-based blood bank
and transfusion service that
collects blood (278).
The survey also captured the
current corporate status of respondents. Multiple responses
were permitted. The majority
of the blood centers self-

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Figure 3-1. Distribution of blood center and hospital respondents by USPHS region.

classified as independent
(121/129, 93.8%). Also selected were university (2),
government (3), owned by a
corporation or other entity (4),
and affiliated with another institution (5). Hospital respondents were largely independent (649/1,604, 40.5%) or
owned by a corporation or
other entity (605/1,604,
37.7%). Also indicated were
affiliations with another institution (12.3%), government
(10.0%), and university
(3.4%).

Table 3-6. United States Public Health Service Regions
USPHS
Region

States

I

CT, ME, MA, NH, RI, VT

II

NJ, NY, (PR, VI)

III

DE, DC, MD, PA, VA, WV

IV

AL, FL, GA, KY, MS, NC, SC, TN

V

IL, IN, MI, MN, OH, WI

VI

AR, LA, NM, OK, TX

VII

IA, KS, MO, NE

VIII

CO, MT, ND, SD, UT, WY

IX

AZ, CA, HI, NV, (Guam, American Samoa, CNMI,
FSMI, RMI, Palau)

X

AK, ID, OR, WA

Introduction and Methods

11

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Blood Center
Nonrespondents

cilities were both contained
within the same sampling
stratum and located in other
Eleven blood centers did not strata, and several respondents
respond to the survey. Data
reported for treatment facilifor key collection variables
ties that were not in the origiwere available from the previ- nal sample. When groupous survey and other AABB
reporting information was
resources for seven nonavailable, the facilities that
responding centers and were reported the information comused in lieu of statistical imbined the data for all facilities
putation. The remaining four so that it was not possible to
centers are known to have
separate those data to ensure
relatively low collection
that the most appropriate
volumes. Collectively, the
weights were assigned.
non-respondents were
estimated to account for 4.0% For the 2005 survey, a data
of total US blood center colcontrol system was designed
lections.
to collect information on the
following group-reporting
categories:
Limitations of the Survey • The sampled treatment facility reported for itself
The weights described here
only.
account for simple survey re• The sampled treatment fasponse. However, more comcility reported for other
plex issues surrounding resampled treatment facilities
sponse exist in this survey.
only.
• The sampled treatment faSeveral respondents (includcility chose another saming both blood centers and
pled treatment facility to
hospitals) reported for other
report the data.
treatment facilities in the sample. These other treatment fa-

12

Introduction and Methods

• The sampled treatment facility reported for itself and
other sampled treatment facilities.
• The sampled treatment facility reported for itself and
other sampled and
nonsampled treatment facilities.
Useful information was captured using the control system. However, because of the
complex issue of different
types of treatment facilities
(including health systems, individual hospitals, satellite facilities, and affiliated hospitals) that report together,
information necessary to adjust the weights with absolute
accuracy could not be captured. Additional information
describing the relationship between sample members and
those for whom they report, as
well as reporting practices,
will be essential to accounting
for these complex issues. In
other words, stricter definitions of the eligible population
and reporting requirements
are needed for future surveys.

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

4. Blood Collected and Processed
in the United States

Total Collections
The total supply of blood collected in the United States in
2004 was 15,288,000 units
before laboratory testing
(Table 4-1). Blood centers
collected 14,305,000 units, or
93.6% of the total. The balance of 983,000 units (6.4%)
was collected by hospitals.
The total supply in 2004 was
virtually unchanged (–0.2%).
However, the composition of
the supply in 2004 was different than that of 2001. While
traditional allogeneic collections were responsible for the
high number of units collected in 2001, in 2004 these
high numbers were maintained by the increased use of
red cell apheresis technology
in the face of reduced (and, in
the case of autologous and directed donations, significantly
reduced) traditional blood
collections. On a positive
note, the gain seen in blood
donation activity following
the terrorist attacks of September 11, 2001, has not been
lost in the subsequent years.

Whole Blood Collections

Autologous, or self-directed
units, totaled 458,000, a deDonations of WB in 2004 to- crease of 26.0% in comparitaled 14,464,000 and were
son with 2001 (p<0.0003).
reported according to the type Hospitals collected 31.7% of
of donation. Community do- all autologous units.
nations, excluding directed
donations, accounted for
90.8% of the supply; directed Red Blood Cells
donations accounted for
Collected by Apheresis
0.8%; and autologous donations accounted for 3.0%. An- In addition to WB collections,
824,000 RBC units were colother 824,000 units (5.4%)
lected by apheresis, which
were red blood cells (RBCs)
collected by apheresis, which typically yields a double volume of cells. Apheresis proceis discussed below.
dures in 2004 yielded a 202%
Community donations totaled increase in RBC units in com13,890,000, of which 94.4% parison to the 273,000 colwere collected by blood cen- lected in 2001, which is statistically significant (p<0.0001).
ters, and 5.6% by hospitals.
RBCs collected in this manner
The percentage decrease in
contributed 5.4% of the total
community donations beWB/RBC supply in 2004. This
tween 2001 and 2004 was
total included autologous
2.6% (p=0.8036). Directed
donations declined by 31.3%, collections (2.2% of the
to 116,000 units (p=0.0285). apheresis total) as well as
directed donations (0.4% of
Hospitals reported 31.0% of
directed donations, a portion the total).
of which were eventually
The expanded use of this
crossed over to the commutechnology occurred largely
nity supply.
in blood centers, which accounted for 96.7% of reported

Blood Collected and Processed in the United States

13

14

Blood Collected and Processed in the United States

*significantly different from 2001 data.

Collections
WB Allogeneic (excluding
directed)
WB Autologous
WB Directed
RBC Apheresis
Total Supply
Rejected on Testing
Available Supply
Transfusions
Allogeneic (excluding
directed)
Autologous
Directed (to designated
patient)
Pediatric
Total Transfusions
Outdated WB/RBCs

Activity
775
145
36
27
983
15
968
13,169
258
131
57
13,614
337

313
80
797
14,305
255
14, 050
551
13
1
2
568
166

Total

11
329
15

18
38

319

14
10
10
146
3
144

150

95% Cl

Hospitals

13,115

Blood
Centers

60
14,182
503*

270*
132

13,720

458*
116*
824*
15,288
270
15,019

13,890

2004
Combined
Total

0.4
100.0
3.3

1.9
0.9

96.7

3.0
0.8
5.4
100.0
1.8
98.2

90.8

% of Total
Collections/
Transfusions

84
13,898
880

359
95

13,361

619
169
273
15,320
245
15,076

14,259

2001
Total

–29.2
2.0
–42.9

–24.6
39.4

2.7

–26.0
–31.3
200.4
–0.2
10.3
–0.4

–2.6

% Change
2001-2004

Table 4-1. Estimated 2004 Activities of US Blood Centers and Hospitals for Whole Blood (WB) and Red Blood Cells (RBCs)
(expressed in thousands of units)

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 4-2, indicate that 53.7% Non-Red-Blood-Cell
Components Processed
of units released were group
O units, with O-negative
blood accounting for 6.8% of Non-RBC component units
all units. Another 21.6% were collected or processed include platelets from whole
group A units, with most of
blood, platelets from
these being A-positive units.
Groups B and AB comprised apheresis, plasma from
apheresis, cryoprecipitate,
11.8% of units released for
and granulocytes. The total
transfusion. The group and
type of the remaining 13% of number of non-RBC compounits were not reported. Hos- nents produced for transfusion
pitals were less likely to cate- in 2004 was 11,559,000
(apheresis platelets counted as
gorize their distribution data
by group and type. However, platelet doses, not as platelet
because blood centers release concentrate equivalents).
Note: For the current analysis, the greater proportion of units
blood centers were consulted for distribution, they are rePlatelets
to determine the most accusponsible for approximately
rate red cell yield from an
10% of the total for which
Platelet concentrates were deRBC apheresis procedure.
group and type went
rived from 4,202,000 WB
From this input, it was deunreported.
units, an increase of 0.9%
cided that, on average, an
RBC apheresis procedure
yields 1.9 units of RBCs,
rather than 2.0 units. Therefore, a factor of 1.9 was used
Table 4-2. WB/RBC Units Released for Distribution by
in the conversion of RBC
Blood Group and Type by Blood Centers and Collecting
apheresis collection proceHospitals
dures to RBC units produced
Unweighted Units
in 2004. The previous survey
Released
used a factor of 2.0.
units. In 2001, 63 blood centers and 17 hospitals reported
such collections. By 2004,
115 blood centers had
adopted this technology,
while only 30 hospitals reported collecting RBCs by
apheresis. The mean number
of units collected by blood
centers was 6,834 and for
hospitals was 885. The minimum number of units collected by any facility reporting
this activity was 10 and the
maximum was 50,201.

Blood
Group/Type

Whole Blood and Red
Blood Cell Units Released
for Distribution
Blood centers and collecting
hospitals reported the total
number of WB/RBC units released for distribution in
2004, and provided a breakdown of released units by
ABO group and Rh type. The
unweighted results, shown in

(in thousands
of units)

% of
Total

O+

5,880

46.9

O–

846

6.8

A+

2,406

19.2

A–

299

2.4

B+

887

7.1

B–

196

1.6

AB+

327

2.6

AB–

61

0.5

Unknown

1,624

13.0

Total Units

12,527

100.0

Blood Collected and Processed in the United States

15

16

Blood Collected and Processed in the United States
386
214

Total Platelets

FFP/Apheresis Plasma
599

*weighted data, includes 95% confidence interval (CI)
†platelet concentrate equivalent units (apheresis packs); includes splits
‡for transfusion; includes apheresis plasma
§significantly different from 2001 data
⏐⏐for transfusion only

Outdated Non-RBC Components

Cryoprecipitate

32

33

WB-Derived Platelet
Concentrates

⏐⏐

353

1,133

Apheresis Platelets†

Transfusions

Cryoprecipitate

4,265

12,182

Total Platelets

FFP/Apheresis Plasma

3,930

WB-Derived Platelet
Concentrates
‡

8,252

Apheresis Platelets†

Collection Production

Activity

Blood
Centers

480

858

3,875

9,489

1,504

7,985

31

386

1,180

272

909

Total

Hospitals*

50

58

269

655

205

626

21

100

398

104

366

95% Cl

1,079

890
§

4,089

9,875

1,346

898

3,926

10,196

2,614
1,537

7,582
§

1,067

4,437

12,898

4,164

8,734

2001
Total

8,338 (1,390)

1,164

4,651

13,362

4,202

9,161 (1,527)

2004
Total

Table 4-3. Estimated 2004 Activities of US Blood Centers and Hospitals for Non-Red-Blood-Cell Components
(expressed in thousands of units)

–19.8

–0.9

4.1

–3.1

–41.2

10.0

8.9

4.8

3.6

0.9

4.9

% Change
2001-2004

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

from the 2001 volume (Table
4-3). Platelets were prepared
from 30.3% of all allogeneic
WB collected, essentially unchanged from 2001. Blood
centers processed 3,930,000
units (93.5%), while hospitals
produced 272,000 (6.5%).
An estimated 1,164,000
apheresis platelet procedures
were completed, yielding
1,527,000 platelet doses. This
volume indicates a split rate
of 31.2% overall. For comparison with production of WBderived platelets, it is assumed
that the number of platelets in
each apheresis collection is
equivalent to six units of
platelet concentrates, yielding
the equivalent of 9,161,000
platelet concentrate units. This
was statistically unchanged in
comparison with 2001. Blood

centers collected 90.1% of
96% coming from blood cenapheresis platelets, while
ters.
hospitals were responsible for
9.9%.
Cryoprecipitate
A total of 1,164,000 units of
cryoprecipitate were prepared.
This was an increase of 9.1%
A total of 4,651,000 units of
plasma from apheresis proce- over 2001, but not statistically
significant (p=0.0730). Blood
dures were produced for
transfusion. This volume is an centers accounted for 97.3%
of cryoprecipitate produced.
increase of 4.8% from the
2001 volume but statistically
unchanged (p=0.6941). Blood
Granulocytes
centers produced 91.7% of
the plasma (4,265,000 units) Granulocytes, which are preand hospitals produced the re- pared from both apheresis and
maining 8.3% (386,000 units). whole-blood-derived buffy
A total of 111,000 plasmacoat units, totaled 13,000
pheresis procedures were re- units produced. Blood centers
ported. In addition, 7,804,000 reported 94.4% of this total,
units of plasma were prowhich includes 6,050 nonduced that were intended for directed apheresis donations,
further manufacture, with
and 334 directed apheresis

Plasma

donations.

Blood Collected and Processed in the United States

17

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

5. Blood Transfused in the United States

Whole Blood and Red
Blood Cells Transfused

by patients in 2004. Autologous units accounted for 1.9%
of all units transfused in 2004.
Transfusions of whole blood
Autologous transfusions,
(WB) and red blood cells
which declined steadily be(RBCs) of all donation types
tween 1992 and 1999 and
totaled 14,182,000 units, a
were statistically unchanged
2% increase from 2001,
from 1999 to 2001, declined
which is not statistically signif- significantly again (p<0.0001)
icant (p=0.5379) (see Table
between 2001 and 2004.
4-1). Allogeneic units, includ- Only 487 units (0.04%) were
ing pediatric units expressed
reported to have been crossed
as adult equivalent units, ac- over to the community supply
counted for 98% of units
in 2004.
transfused, or 13,912,000
units. There was a small inDirected donations, the donacrease (2.7%, not statistically tion of allogeneic blood for a
significant) in the transfusion designated patient other than
of allogeneic units (commuthe donor, accounted for
nity, directed, and pediatric
132,000 units transfused. Ancombined) in comparison
other 144,000 units were rewith 2001. The percent of
ported to have been crossed
available allogeneic units uti- over to the community suplized was 97.7%, in contrast
ply. The overall utilization of
to 93.7% and 92.5% in 2001 directed units was greater
and 1999, respectively. Blood than 100% vs 88.2% in 2001.
centers claimed responsibility Although it appears that there
for 4.0% (568,000) of transare more directed units transfused units, primarily through fused and outdated than colcontractual arrangements with lected, many red cell apheresis
hospitals.
procedures are directed donations and are not attributed as
The number of autologous
such in the collection figures.
units transfused, 270,000, represented 59% of the 458,000 In the last survey, an attempt
units donated preoperatively was made to capture more

detailed information regarding
the usage of WB/RBC units—
specifically, the number of
units ordered for surgical procedures (total sent to operating and recovery rooms). The
data were not weighted. In
2004, 261 (16.3%) hospitals
answered the question, compared with 179 (15.7%) in
2001. The responses totaled
297,000 units ordered, as
compared with 350,000
reported in 2001. This total
was then calculated as a percentage of the total units
transfused at the responding
institutions. The overall percentage of transfused WB/
RBCs ordered for surgical
procedures at the reporting
hospitals was 27.9%
(compared with 39.5% in
2001).

Whole Blood and Red
Blood Cell Recipients
The 2005 NBCUS captured
the number of patients who
received WB/RBCs of each
donation type. The reported
number of recipients of
allogeneic units (including
directed) was 1,688,345 per

Blood Transfused in the United States

19

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

4,569,584 units transfused by
1,073 medical treatment facilities. This represents 2.7 units
per recipient, a small decline
from 2.9 units per recipient in
2001. Autologous recipients
received an average of 1.5
units each compared to 1.6
units in 2001. Finally, for recipients of pediatric units, the
ratio was 2.3 units per recipient (2.4 units per recipient in
2001).

cated the volumes of transfused plasma that had been
processed by a variety of
methods as shown in Figure
5-1. The weighted results, for
which overlap is possible, are
as follows:
• FFP represented 82.5% of
plasma transfused
(3,398,000 units).
Note: The AABB standard re- • Plasma frozen within 24
quiring testing for bacterial
hours after phlebotomy
contamination of platelets was
comprised 14.7% of
implemented on March 1,
plasma transfused (604,000
Extrapolating the ratio of 2.7
2004.
units); this amount is signififor WB/RBC recipients to total
cantly higher than in 2001
WB/RBCs transfused yields a The combined total of fresh
(p<0.05).
national estimate of 5.3 milfrozen plasma (FFP), and
• Jumbo plasma accounted
lion total WB/RBC recipients apheresis plasma resulted in
for 1.8% (73,000 units) of
in 2004. This represents an
4,089,000 units transfused, an
plasma transfused.
increase in recipients of 8.6% increase of 4.1% over the
• Pediatric-sized plasma units
(4.9 million recipients in
2001 volume of 3,926,000
represented 0.9% (39,000
2001).
(not statistically significant).
units) of plasma transfused.
Reporting institutions indiThe transfusion of wholeblood-derived platelet concentrates continued on a significantly downward trend
(p<0.001), declining 41.2% to
1,537,000 from the previous
survey estimate of 2,614,000.
This has been the trend in
platelet usage since 1994.

Non-Red-Blood-Cell
Components Transfused
National estimates for
non-RBC components transfused in 2004 are presented in
Table 4-3.
An estimated total of
9,875,000 platelet units were
transfused to patients in 2004,
a decrease of 3.2% in comparison with 2001; this decrease is not statistically significant. The transfusion of
apheresis platelets increased
by 10.0% from 7,582,000 to
8,338,000 platelet concentrate equivalent units; this increase is not statistically
significant.

20

Donor Retested
0.09%
Plasma Frozen
Within 24 Hours
after Phlebotomy
14.67%

Solvent/DetergentTreated
0.00%

Jumbo
1.77%
Pediatric Size
0.94%

FFP
82.53%

Figure 5-1. Types of plasma transfused in 2004.

Blood Transfused in the United States

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

(n=1,327). Blood centers reported the transfusion of
44,000 apheresis platelet
doses (n=20) and 10,000
whole-blood-derived platelet
concentrates (n=17). The
Donor retested and solvent/
combined totals were
detergent-treated plasma
725,000 doses of apheresis
showed marked and signifiplatelet products and 194,000
cant decreases in use and
whole-blood-derived platelet
each represented less than 1% Platelet Dosage
concentrates. This represents
of the plasma transfused in
Institutions
reporting
platelet
a ratio of 3.7:1 in contrast to
2004—3,639 and 19 units
transfusions were requested to the 2.2:1 ratio reported in
reported, respectively. The
indicate the number of thera- 2001.
manufacturer of solvent/
peutic doses of each type of
detergent-treated plasma
platelet product (see Figure
Hospitals reporting wholeceased production in 2002.
5-2). These data were not
blood-derived platelet conweighted.
centrate doses also indicated
Cryoprecipitate transfusions
the most common dosage of
decreased slightly by 0.9% to
Hospitals
reported
the
transfuthat product used in their in890,000 units. Only an addision of 681,000 doses of
stitutions (see Figure 5-3).
tional 15,000 units were reapheresis
platelets
(n=1,393)
Similar to 2001 results, data
portedly issued for fibrin sealand 184,000 whole-bloodshow that 50.3% of hospitals
ant, compared with 44,000
derived
platelet
concentrates
reported six concentrates,
units in 2001, a 65.9% decline (p=0.005), following a
61.4% decline in the 2001
survey (from 114,000 units).
Other uses of cryoprecipitate
were not assessed.
Whole-BloodThe median volume of plasma
transfused during a single
transfusion episode was 300
mL (n=1,441).

Granulocytes, prepared from
both apheresis and wholeblood-derived buffy coat
units, resulted in a total of
2,174 units transfused. This
represents a 42% decrease
from 2001 levels (3,744 units
transfused).
The total number of units
transfused in the United States
in 2004, both RBC and
non-RBC components, was
29,038,000. The increase of
119,000 (0.4%) in comparison with 2001 is not statisti-

cally significant. With the exception of the significantly
lower number of transfusions
of whole-blood-derived
platelet concentrates, transfusion activity remained consistent between 2001 and 2004.

Derived Platelets
21%

Apheresis Platelets
79%

(n=919,000)

Figure 5-2. Therapeutic platelet doses transfused in 2004.

Blood Transfused in the United States

21

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

60

Percent of Hospitals

50.3

48.5

50

2001

2004

40

30
22.9
20

17.9
16.6
13.3

13.1
10.1

10

0

≤5
<=5

6

8

10

Platelet Concentrate Dosage (in Units)

Figure 5-3. Most common platelet concentrate dosage reported by hospitals.

while 13.1% reported 10,
10.1% reported eight, and
22.9% reported five or fewer
(n=731). In this survey, the
use of five or fewer concentrates was increasingly popular in all hospitals except
those performing fewer than
1,000 surgeries per year.

Outdated Units

(p<0.001). Blood centers were
responsible for 47.8% of all
outdates. Allogeneic, nondirected RBC outdates were
shared evenly between blood
centers and hospitals in 2004,
while hospitals were responsible for more of the directed
and autologous outdates.
Most non-RBC components,
with the exception of wholeblood-derived platelets, were
outdated by hospitals.

The national estimate for all
whole blood and blood component units outdated by
blood centers and hospitals in
2004 was 1,551,000 units.
Significantly fewer units outdated in 2004 than in 2001

Outdated WB and RBCs totaled 503,000 units, of which
266,000 were allogeneic,
nondirected RBCs. The remaining outdates were:
autologous (203,000), whole
blood (24,000), and directed

22

Blood Transfused in the United States

(10,000) units. The percentage
of outdated WB/RBCs contributed by each collection type
is illustrated in Figure 5-4. The
mean number of units outdated by blood centers was
923 while the mean for hospitals was 33, a reduction for
both medical treatment facilities (3,100 and 114, respectively, in 2001). As shown in
Table 5-1, outdated WB/RBCs
accounted for 3.2% (503,000)
of all WB/RBC units processed
in 2004.
Survey respondents provided
a breakdown of ABO group
and Rh type for approximately
half of all WB/RBCs outdated,
as shown in Figure 5-5. The

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

identified by ABO group or
Rh type.

Whole Blood
4.8%

The total number of WB/RBC
outdated units was significantly fewer than the 2001 total by 42.9% (p<0.0001). The
percentage of outdated
WB/RBCs, 3.2%, was lower
than the 5.8% and 5.3% reported in 2001 and 1999, respectively. The percentages of
directed and autologous units
outdates (9% and 44.2%, respectively) were unchanged
from 2001.

Directed
2.0%

Allogeneic (nondirected)
52.9%

Autologous
40.4%

(n=503,000 units)

As has been the case in previous surveys, whole-bloodderived platelet concentrates
Figure 5-4. WB/RBC unit outdates by collection type.
accounted for the greatest percentage of total outdates,
48.2% (762,000/1,582,000).
These 762,000 unused units
represented a 29% decline in
largest portion of outdated
O-positive and B-positive
the volume of outdates reunits was AB-positive; when
units accounted for an addiported in 2001. Outdated
combined with A-positive
tional 16%; Rh-negative units platelets from WB accounted
units, this accounted for
also accounted for 16% of
for 18.1 % of all whole-bloodslightly more than one-third of outdated units. An additional derived platelets processed in
the total WB/RBC outdates.
73,000 (31.3%) units were not 2004.

Table 5-1. Outdated Components as a Percentage of the Total Volume of Each Type
Processed for Transfusion in 2004

WB/RBCs
Outdated
Processed
Percent
Outdated

Whole-BloodDerived
Apheresis
Platelets
Platelets

Plasma

Cryoprecipitate

All
Components

503,000

762,000

212,000

75,000

31,000

1,582,000

15,571,000

4,202,000

1,527,000

4,651,000

1,164,000

27,115,000

3.2%

18.1%

13.9%

1.6%

2.7%

5.8%

Blood Transfused in the United States

23

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Intraoperative Autologous
Blood Recovery

O+
9%

Intraoperative blood recovery
refers to blood recovery and
return procedures conducted
during surgery. Such procedures may result in a reduction of the demand for alloB+
AB7%
geneic blood in some patients.
5%
BIn 2004, a total of 95,000 of
3%
these procedures were conA4% Oducted nationally, an amount
AB+
4%
estimated on the basis of data
21%
reported by 150 hospitals. In
(n=234,000 units)
2001 the estimated number
of these procedures was
180,000. It is possible that the
Figure 5-5. Percentage of WB/RBC outdates by blood
prevalence of this procedure
group and type.
went underreported in the
2005 survey. Data regarding
the use of intraoperative
blood recovery may have
been maintained by different
Apheresis platelets contribThe number of outdated
hospital groups (eg, operating
uted 212,000 units, or 13.4% cryoprecipitate units was
room personnel) than the
of total outdates. This amount 31,000, 2.7% of the total
blood bank staff; therefore, all
represented 13.9% of aphere- cryoprecipitate processed.
data may not have been resis platelets processed, which Apheresis platelets, plasma,
ported to those completing
is significantly lower than the and cryoprecipitate combined the survey. Future surveys will
number reported in 2001.
accounted for 20% of all out- explore how to better ascerdated units.
tain the procedure’s actual
Outdated plasma units totaled
prevalence. In 2004, this ac75,000, only 1.6% of the
Overall, the proportion of out- tivity is more likely to be replasma units processed for
dated units was much lower
ported by hospitals performtransfusion. FFP contributed
in 2004 than in previous
ing more than 2,400 surgeries
66,000 units to the total and
years, suggesting that product per year than by smaller hosapheresis plasma accounted
utilization has become more pitals and blood centers.
for 9,000 units.
efficient at meeting specific
demand.
Unknown
31%

24

Blood Transfused in the United States

A+
16%

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

6. Component Modification

Irradiation

of component units produced Leukocyte Reduction
in 2004 were irradiated. Of
Irradiation of blood compothese, 1,473,000 were whole Blood components are leukonents is intended to prevent
blood (WB) or red blood cell cyte-reduced (LR) to remove
graft-vs-host disease and is
leukocyte-associated infec(RBC) units and 1,184,000
performed primarily in hospi- were all other components
tious agents and to avoid
tals. The survey measured the combined. The percentage of alloimmunization in transfunumber of facilities in 2004
sion recipients. Leukocyte reall units irradiated in 2004,
that had a policy to irradiate
10.0, was a decrease of 0.5% duction may occur during
all components before transfu- from the percentage irradiated collection or at some time before components are placed in
sion. Of 1,723 survey respon- in 2001 (10.5%).
dents, 29 (1.7%) indicated
that there was such a policy
in place at their institution.
These 29 respondents were
hospitals, and 23 (79.3%)
Table 6-1. Blood Components Modified by Irradiation in
perform 2,400 or more surAll Facilities in 2004
geries per year. Of the 129
% of Total
blood centers, 128 reComponents
sponded negatively and 1
Irradiated
Blood Component
Irradiated
response was missing.
Table 6-1 summarizes the
types and numbers of blood
component units irradiated in
blood centers and hospitals
in 2004 (including all facilities that irradiate some components, not just the 29 hospitals that irradiate all
components). A total of
2,657,000 component units,
including pediatric aliquots,
were irradiated; 736,000 in
blood centers and 1,922,000
in hospitals. Overall, 10.0%

All Facilities
WB/RBCs

1,473,000

9.8

Other Component Units

1,184,000

10.3

Total Components

2,657,000

10.0

WB/RBCs

413,000

2.9

Other Component Units

323,000

3.0

Total Components

736,000

3.0

1,060,000

7.1

861,000

7.5

1,922,000

7.2

Blood Centers

Hospitals
WB/RBCs
Other Component Units
Total Components

Component Modification

25

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 6-2. Blood Components Modified by Prestorage and Poststorage Leukocyte
Reduction in All Facilities in 2004
Poststorage
LeukocyteReduced
Units

Prestorage
LeukocyteReduced
Units

Blood Component

Total
LeukocyteReduced
Units

% of Total
Units
Leukocyte
Reduced

All Facilities
WB/RBCs

10,346,000

885,000

11,231,000

74.8

996,000

168,000

1,165,000

27.7

1,491,000

11,000

1,502,000

98.4

16,000

1,000

17,000

0.3

12,849,000

1,066,000

13,915,000

52.4

WB-Derived Platelets
Apheresis Platelets
Other Component Units
Total Components

inventory and is considered
prestorage LR. Leukocyte
reduction is considered poststorage LR when it occurs
after the unit is placed into
inventory, primarily by filtration at the patient’s bedside.
In 2004, a total of 13,915,000
units, including pediatric
aliquots, were LR compo-

nents—12,849,000 (92.3%)
prestorage LR units and
1,066,000 (7.7%) poststorage
LR units (Table 6-2). Blood
centers produced 12,770,000
units (91.8%) and hospitals
produced 1,145,000 units
(8.2%). Of those units produced by blood centers,
12,094,000 (94.7%) were

prestorage LR units, while
only 676,000 (5.3%) were
poststorage LR units.
The components most frequently undergoing leukocyte
reduction were WB/RBCs
(74.8%) and apheresis platelets (98.4%). The 9.6% increase in LR components

Table 6-3. Number of Irradiated and Leukocyte-Reduced Blood Component Units
Processed in All Facilities in 2004 and 2001 (expressed in thousands of units)
Blood Centers

All Facilities

Modification

2001

2004

%
Change

2001

2004

%
Change

2001

2004

Total, All Units

11,513

13,506

17.3

3,935

3,066

–22.1

15,448

16,572

7.3

644

736

14.3

2,103

1,921

–8.7

2,746

2,657

–3.2

10,870

12,770

17.5

1,832

1,145

–37.5

12,702

13,915

9.6

10,413

12,094

16.1

743

755

1.6

11,157

12,849

15.2

456

676

48.2

1,089

390

–64.2

1,545

1,066

–31.0

Irradiated
LeukocyteReduced
Total
prestorage
poststorage

26

Hospitals

Component Modification

%
Change

139,000

Other Component Units

22,000
6,000

Apheresis Platelets

Other Component Units

133,000

Other Component Units
2,168,000

514,000

Apheresis Platelets

Total Components

361,000

1,160,000

WB-Derived Platelets

WB/RBCs

Hospitals

142,000

52,000

WB-Derived Platelets

Total Components

62,000

WB/RBCs

Blood Centers

2,310,000

536,000

Apheresis Platelets

Total Components

413,000

1,222,000

WB-Derived Platelets

WB/RBCs

All Facilities

Blood Component

Irradiated
Units

11,088,000

333,000

1,088,000

712,000

8,955,000

317,000

10,000

52,000

68,000

187,000

11,405,000

343,000

1,140,000

780,000

9,142,000

Prestorage
LeukocyteReduced Units

616,000

4,000

15,000

146,000

451,000

3,000

0

0

20

3,000

619,000

4,000

15,000

146,000

454,000

Poststorage
LeukocyteReduced Units

11,704,000

337,000

1,103,000

858,000

9,406,000

320,000

10,000

52,000

68,000

190,000

12,024,000

347,000

1,155,000

926,000

9,596,000

Total
LeukocyteReduced Units

9.8

2.7

37.0

23.5

8.2

0.6

0.1

1.6

3.4

0.4

10.5

2.8

38.6

26.9

8.6

% of Total
Units
Irradiated

53.0

6.8

79.4

55.8

66.3

1.4

0.2

3.7

4.4

1.3

54.4

7.0

83.1

60.3

67.7

% of Total Units
LeukocyteReduced

Table 6-4. Estimated Number of Blood Component Units Modified by Irradiation or Leukocyte Reduction and Transfused by All
Facilities in 2004
THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Component Modification

27

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

since 2001 can be attributed
to LR WB/RBCs (80%) and
apheresis platelets (20%).
Table 6-3 summarizes the
changes that occurred between 2001 and 2004 in the
numbers of irradiated and LR
component units. During the
period, the number of units irradiated in blood centers increased by 14.3%, while the
number irradiated in hospitals
declined by 8.7%.

poststorage LR units continued to decline in hospitals by
64.2%, while increasing in
blood centers by 48.2%.
Poststorage LR units declined
overall, however, by 31.0% in
2004.

Transfusion of Irradiated
and Leukocyte-Reduced
Components

ters and 11,704,000 (97.3%)
by hospitals. Of the total units
transfused, 95% were prestorage LR units and 5% were
poststorage LR units. Substantial proportions of all RBCs
and platelets transfused in
2004 were LR units: 67.7% of
WB/RBCs; 60.3% of wholeblood-derived platelets, and
83.1% of apheresis platelets.

Table 6-5 summarizes the
changes that occurred between 2001 and 2004 in
numbers of irradiated and LR
component units transfused.
The number of irradiated units
increased by 15.8%, while LR
units increased by 2.7%. At
the same time, a continuing
shift occurred in the mix of LR
units transfused. Transfusion
of prestorage LR units increased by 7.9%, while transIn
2004,
some
12,024,000
fusion of poststorage LR units
Prestorage LR component
transfused component units
declined by 45.3%, a pattern
units increased by 16.1% in
were
leukocyte-reduced—
also observed in 1999 and
blood centers, and by 1.6% in
320,000 (2.7%) by blood cen- 2001.
hospitals. At the same time,
Table 6-4 summarizes the
types and numbers of irradiated and LR blood component
In contrast to 2001, when
units transfused during 2004.
hospitals leukocyte-reduced
73% of units after storage, in A total of 142,000 irradiated
units were reported as trans2004 only 34.0% (390,000
fused by blood centers and
units) were processed after
storage. Another 755,000 total 2,168,000 by hospitals. In
total, 10.5% of all transcomponents (66.0%) were
leukocyte-reduced by hospi- fused component units were
irradiated.
tals before storage.

Table 6-5. Number of Irradiated and Leukocyte-Reduced Component Units Transfused in
2004 and 2001 (expressed in thousands of units)
Total Units
2004

2001

Total, All Units

14,192

13,576

616

4.5

2,168

1,873

295

15.8

12,024

11,703

321

2.7

11,405

10,572

833

7.9

619

1,131

–512

–45.3

Irradiated
Leukocyte-Reduced Total
prestorage
poststorage

28

Increase/
Decrease

Modification

Component Modification

% Change

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

7. Current Issues in Blood Collection and
Screening

Screening Test Losses
The overall number of donated whole blood units with
positive results on infectious
disease screening tests was
270,000, or 1.8% of allogeneic collections, including
directed donations. The number of units collected by
blood centers and discarded
for this reason was 255,000
(1.9%) (n=138 blood centers
reporting). Among hospitals,
screening test loss was 1.8%
of donations or 15,000 units
(n=258). Screening test loss
was lower in hospitals than
reported in 2001, when the
test loss totaled 2.5% of
donations.

perienced testing losses significantly lower than the mean
in 2004. The proportions of
units lost per individual
screening test were not assessed by the 2005 NBCUS.

Therapeutic Phlebotomy

Blood centers reported a total
of 28,000 procedures performed at 33 centers. The
number of procedures per
center ranged from 31 to
3,685, with a median of 471
procedures. Hospital data
were weighted to achieve a
national estimate of 112,000.
The median for hospitals was
41 procedures, with a range
of 1 to 1,839. Together, hospitals and blood centers nationwide are estimated to have
performed 140,000 therapeutic phlebotomy procedures in
2004. This is not significantly
different from the previous
survey findings. The larger the
surgical volume of the hospital, the more likely the hospital was to offer this service,
with the most procedures being performed at hospitals
with 8,000 or more surgeries
per year.

Therapeutic phlebotomy procedures are performed for the
benefit of the donor as a patient rather than for blood collection purposes. In August
2001, the US Food and Drug
Administration (FDA) approved the use of blood
collected from hereditary
hemachromatosis patients for
transfusion. The FDA has allowed this use provided that
1) the blood collecting facility
As observed in previous surhas a variance from the FDA,
veys, test losses varied between reporting blood centers 2) the patient meets all regular
in different United States Pub- donor eligibility requirements,
and 3) the therapeutic phlelic Health Service regions.
botomy is performed at no
The highest average losses
were experienced by centers cost to the patient. Hereditary Disaster Planning
hemachromatosis is only one
in regions II, III, and VI. ReIn follow-up to the terrorist atgion VI reported test loss that of the indications for therapeutic
phlebotomy.
tacks of September 11, 2001,
was significantly higher than
the United States instituted
the mean. Regions IV, V, VII,
many changes in approaches
VIII, and IX (see Table 3-6) ex-

Current Issues in Blood Collection and Screening

29

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Facilities were also asked to
report the number of days of
group O RBC uncrossmatched
inventory that was considered
The survey also asked how
to be “ideal.” The overall
many days of group O red
mean of 5.5 days again varied
blood cell (RBC) inventory
were considered to be “criti- between hospital and blood
center respondents. The hoscally low.” The response to
pital mean was 5.5 days of
Medical facilities were asked this question varied widely
group O RBC uncrossmatched
whether they have an “emer- depending on the type of fagency preparedness plan” for cility, its location, and the ser- inventory, whereas the blood
centers reported a mean of
blood. Almost all (99.2%) of
vices provided. The system4.7 days as ideal. Smaller hosthe blood centers responded
wide mean of 2.4 days of
group O RBCs was driven by pitals were more likely to reaffirmatively. Among the
port higher mean numbers of
the hospitals, which had a
98.7% of hospitals that remean of 2.5 days as the criti- days as ideal than were larger
sponded to the question,
cally low inventory level. The hospitals (see Figure 7-1). This
89.6% had emergency preblood center mean of 1.7 days may reflect the locations of
paredness plans. Hospitals
larger hospitals—urban and
most likely reflected closer
across all surgical volume
proximal to blood centers and
proximity to the supply and
strata were equally likely to
blood donating populations.
have a plan in place. None of greater control over access.
the cord blood banks reported
to critical national functions,
including blood safety and
supply. Several new questions
were developed for the 2005
NBCUS, the first survey to be
distributed nationally since
the attacks.

having emergency plans for
blood.

8

Number of Days

7
6
5
4
3

6.68

6.08

5.48

5.49

1,4002,399

2,4004,999

2

4.77

4.65

5,0007,999

≥ 8,000

1
0
100-999

1,0001,399

Hospitals by Annual Surgical Volume

Figure 7-1. Ideal number of days of group O red blood cells in uncrossmatched inventory.

30

Current Issues in Blood Collection and Screening

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

8. Current Issues in Blood Transfusion

Blood Inventory
Shortages
The current survey, as well as
previous versions, asked hospitals to indicate the number
of days in the past year that
elective surgery was postponed because of blood inventory shortages, as well as
the number of days that they
were unable to meet other
blood requests. The results are
based on actual (unweighted)
responses.
A total of 135 hospitals (8.4%)
reported that elective surgery
was postponed on one or
more days in 2004 because of
blood inventory shortages. Table 8-1 provides a character-

ization of cancellation reports
in 2004 in comparison with
previous survey years. The
percentage of responding hospitals reporting any shortage
was a statistically significant
decrease from the 2001 data
(p<0.0001), but comparable
to the percentages from 1997
and 1999 data. The range (1
to 39) of days postponed was
narrower in 2004 than in the
recent surveys, but the mean
number of days for all hospitals responding was 3.39, noticeably higher than in years
past. This suggests that while
shortages were less frequent,
when they did occur they
were more acute. There were
no significant differences in
the mean number of days of

postponement among hospitals when grouped by surgical
volume or by region of the
country.
Hospitals indicated separately
that the total number of surgical procedures that were postponed was 546 (n=110) compared with 952 in 2001
(p=0.026). This is a decrease
of 42.6%. The number of surgeries postponed varied significantly between hospitals by
surgical volume (p=0.003).
Hospitals that performed
1,400 to 2,399 surgeries per
year appeared to have the
greatest proportion of surgeries postponed among those
reporting.

Table 8-1. Cancellation of Elective Surgeries by US Hospitals, 1997-2004
Year

>0 Days
(%)

Range
of Days

Median
No. of Days

Mean
No. of Days

No. of Patients
Affected

1997

8.60

1–21

2.0

0.44

Not determined

1999

7.40

1–150

2.0

0.32

568 (n=83)

2001

12.70

1–63

2.0

0.55

952 (n=116)

2004

8.40

1–39

2.0

3.39

546 (n=110)

Current Issues in Blood Transfusion

31

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Hospitals also indicated the
number of days in which
nonsurgical blood requests
were not met. Of responding
hospitals, 16.0% (257/1604)
experienced at least one day
in which nonsurgical blood
needs could not be met vs
18.9% (202/1,066) in 2001
and 16.2% (273/1,588) in
1999. The total number of
days reported was 4,953 and
the range was 1 to 366 (2004
was a leap year). There is a
significant difference between
the mean number of days of
unmet nonsurgical needs for
all respondents between 2001
(2.1) and 2004 (19.27)
(p<0.001). Eight hospitals reported 365 or 366 days in
which nonsurgical blood requests were not met in 2004,
whereas only one hospital reported an entire year of unmet
need in 2001. This most likely
accounted for the large increase in the mean between
2001 and 2004. The hospitals
reporting this condition were
evenly distributed throughout
the country, and only one of
these hospitals had reported
any days of unmet need in
2001.

Crossmatch Procedures
Transfusing facilities were requested to report the total
number of crossmatch procedures, as well as the percentage of procedures performed
serologically, using
microplate technology, and

32

using gel media, if applicable. age of an RBC unit at the time
The data were not weighted. of transfusion. In this survey,
488 hospitals responded, up
A total of 11,221,000 crossfrom 293 in the last survey.
match procedures were perThe data were not weighted.
formed in 2004, compared to The minimum average age re10,506,000 in 2001. This in- ported was 1.5 days, the maxcrease was not statistically sig- imum was 42 days, and the
nificant. Of the total volume
median was 15.0 days. When
of crossmatches reported,
analyzed by United States
microplate technology acPublic Health Service
counted for less than 1%, gel (USPHS) region, however,
media accounted for 18.9%, there was one statistically sigand serologic methods
nificant difference from the
accounted for 77.8%.
national mean of 17.0: the
mean average age of RBC
In order to calculate the cross- units transfused in region VI
match-to-transfusion (C:T) ra- (NM, OK, AR, TX, LA) was
tio, the total number of allo14.5 days (p=0.0013).
geneic WB/RBC units
(unweighted, not imputed)
transfused by the responding Platelet Product Age
medical treatment facilities
This was the first survey to ask
served as the denominator
about the average age of
(7,250,000). The overall C:T
ratio was 1.55 procedures per transfused platelets. A total of
381 hospitals provided data
unit in 2004.
on the average age of wholeblood-derived platelets at
When analyzed by surgical
volume, the trend established transfusion. The minimum average age was 0.5 day old and
in the 2001 data report continued to be seen in the 2005 ranged to a maximum of 6
NBCUS—C:T ratios increased days with the median being 3
with increasing hospital size. days. The mean age was 3.16
days. Quite a few more hospiThe C:T ratios ranged from
1.30 in hospitals performing 0 tals (918/1,604 or 57%) responded with data on an age
to 999 surgeries per year to
for apheresis platelets at the
1.63 in hospitals performing
more than 8,000 surgeries per time of transfusion. The minimum average age was again
year.
0.5 day, the maximum was 6
days, and the median was 3
days. The mean age of
RBC Unit Age
apheresis platelets at
The 2005 survey attempted to transfusion was 3.08 days.
identify trends in the average

Current Issues in Blood Transfusion

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

4
Whole-Blood-Derived Platelets
Apheresis Platelets
3.5

Number of Days

3

2.5

2

1.5

1

0.5

0
100-999

1,000-1,399

1,400-2,399

2,400-4,999

5,000-7,999

≥ 8,000

Hospitals by Annual Surgical Volume

Figure 8-1. Mean platelet age when transfused.

Number of Reactions

80.0
67.5

70.0
60.0
50.0
36.3

40.0
30.0

20.9

20.0
10.0

6.0

9.5

11.8

1,0001,399

1,4002,399

0.0
100-999

2,4004,999

5,0007,999

≥8,000

Hospitals by Annual Surgical Volume

Figure 8-2. Mean number of adverse reactions.

Current Issues in Blood Transfusion

33

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

When analyzed by the number of surgeries per year, data
from responding hospitals
showed a significant trend toward higher mean platelet age
in the largest stratum of hospitals, for both whole-bloodderived (p=0.0065) and
apheresis (p=0.0006) platelets.
See Figure 8-1. Means ranged
from 2.7 to 3.5 days in
whole-blood-derived platelets
and from 2.5 to 3.4 days in
apheresis platelets. There

34

were no significant differences The larger the surgical volume
in platelet age at transfusion
of the hospital, the more
by USPHS region.
transfusion-related adverse reactions reported (see Figure
8-2). Of these adverse events,
Adverse Reactions
160 were reported as transfusion-related acute lung injury.
Approximately 1,322 medical These were more likely to octreatment facilities reported a cur in hospitals performing
total of 32,128 transfusionmore than 8,000 surgeries per
related adverse reactions in
year. Adverse events related
2004. These were defined as to ABO incompatibility were
events that required diagnos- quite rare, with only 52
tic or therapeutic intervention. occurrences reported in 2004.

Current Issues in Blood Transfusion

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

9. Component Costs

Hospitals were requested to
report the minimum, maximum, and average costs paid
per unit in 2004 for each of
four specific components.
These cost data were not imputed or weighted. The mean
of the reported results for each
variable is presented in Table
9-1 and compared with the
2001 value. Table 9-2 displays the average cost of each
component by region of the
country and provides a statistical comparison with the national average. Average component costs are stratified by
hospital surgical volume in
Table 9-3.

Red Blood Cells
The mean of the average
amount paid nationally for a
unit of red blood cells (RBCs)
that was O positive, leukocytereduced, not irradiated, and
not cytomegalovirus negative,
was $201.07 (Table 9-1). This
was higher than the average
in 2001 ($153.68); the difference was statistically significant (p<0.0001). The values
reported by hospitals for this
component ranged from a

minimum of $5.00 to a maximum of $595.00. When
analyzed by United States
Public Health Service
(USPHS) region, the mean of
the average value reported
was significantly different
from the national average in
every region except III, VIII,
and X (Table 9-2). The average cost was significantly
higher than the national mean
in the Northeastern and
Southwestern states (I, II, and
IX). Significantly lower means
were found in the Southeastern and Central states (IV, V,
VI, and VII; p<0.0001). When
analyzed by surgical volume,
the only stratum that differed
statistically from the overall
mean was that of hospitals
with 1,400-2,399 annual inpatient surgeries, with a mean
of $205.45 (p=0.0461) (Table
9-3).

Fresh Frozen Plasma
The mean cost of a unit of
fresh frozen plasma (FFP),
type AB, with a volume of
approximately 250 mL, averaged $56.29 nationally (Table
9-1). The higher cost than the

2001 mean of $52.37 was
statistically significant
(p<0.0001). The values reported by hospitals for this
component ranged from a
minimum of $20.00 to a maximum of $259.77. When analyzed by USPHS region, the
mean of the average values
reported was significantly
higher in the regions including New York, and the Mountain and Western states (II,
VIII, IX, and X). The means reported by regions IV and VII
(which include Southeastern
and Midwestern states) were
significantly lower than the
national mean (Table 9-2).
Hospitals in the largest surgical volume stratum (≥8,000)
reported a mean cost for FFP
that was significantly less than
the overall mean, $52.19
(p=0.0080) (Table 9-3). Not
surprisingly, smaller facilities
(100-999 surgeries) reported
significantly higher FFP costs
(p=0.0257).

Component Costs

35

36

Component Costs
461.44

52.36

Whole-Blood-Derived Platelet
Concentrate, not leukocyte-reduced,
not irradiated, 3 days remaining
before expiration

Apheresis Platelets with 3-4 × 1011
platelets, leukocyte-reduced

50.62

143.03

2001

FFP, type AB ~ 250 mL

RBCs, O positive, leukocyte-reduced,
not irradiated, not cytomegalovirus
negative

Component

494.06

59.49

54.68

193.20

2004

Minimum
Amount Paid

+7.1

+13.6

+8.0

+35.1

%
Change

489.08

57.54

54.55

164.13

2001

527.47

64.21

57.97

210.90

2004

Maximum
Amount Paid

+7.8

+11.6

+6.3

+28.5

%
Change

Mean Dollar Values

Table 9-1. Mean Dollars Paid per Selected Component Unit by Hospitals in 2004 vs 2001

475.48

54.90

52.37

153.68

2001

510.05

63.67

56.29

201.07

2004

Average
Amount Paid

+7.3

+16.0

+7.5

+30.8

%
Change

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

156
271
264
160
96
54
148
45

III

IV

V

VI

VII

VIII

IX

X
1,381

108

II

All Hospitals

79

No.

I

USPHS Region

201.07

199.92

218.12

204.98

189.49

189.81

190.78

186.24

202.67

239.06

233.63

Avg

0.8159

<0.0001

0.3906

0.0007

<0.0001

<0.0001

<0.0001

0.5385

<0.0001

<0.0001

p Value

RBCs

Table 9-2. Average Component Cost by USPHS Region

56.29

66.08

65.74

65.83

48.79

55.56

55.53

50.04

53.86

63.12

58.24

Avg
0.3207

p Value

0.0002

<0.0001

<0.0001

<0.0001

0.5853

0.4645

<0.0001

0.0716

<0.0001

FFP

63.67

86.85

66.75

78.94

54.08

59.88

62.40

63.12

66.12

56.38

90.10

Avg

0.0052

0.7436

0.1474

0.1010

0.2649

0.5975

0.8298

0.4174

0.0583

<0.0001

p Value

Whole-BloodDerived Platelets

Mean Dollar Values

510.05

539.09

527.52

567.07

456.30

529.38

494.48

490.94

509.44

551.33

516.61

Avg

0.0060

0.0033

<0.0001

<0.0001

0.0009

0.0003

<0.0001

0.8894

<0.0001

0.4297

p Value

Apheresis Platelets

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Component Costs

37

38

Component Costs
214
139
284
423
189
132

1,000–1,399

1,400–2,399

2,400–4,999

5,000–7,999

≥8,000

No.

100–999

Annual Surgical
Volume

196.71

199.01

198.38

205.45

204.99

202.47

Avg

0.1796

0.4497

0.1384

0.0461

0.2101

0.5834

p Value

RBCs

Table 9-3. Average Component Cost by Surgical Volume

52.19

55.17

55.63

57.14

57.91

59.06

Avg

FFP

0.0080

0.3752

0.4470

0.4071

0.2813

0.0257

p Value

57.17

60.64

61.39

66.96

69.65

72.01

Avg

0.0626

0.3306

0.2839

0.2535

0.1423

0.0263

p Value

Whole-BloodDerived Platelets

Mean Dollar Values

499.30

510.25

509.71

514.92

515.64

508.11

Avg

0.0986

0.9982

0.8875

0.2919

0.4082

0.6884

p Value

Apheresis
Platelets

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Whole-Blood-Derived
Platelets
The national mean cost for a
unit of whole-blood-derived
platelet concentrate (individual concentrate, not pooled),
not leukocyte-reduced or irradiated, with 3 days remaining
before expiration, was $63.67
in 2004 (Table 9-1). This increase of 16.0% over the
2001 cost was statistically significant (p<0.0001). The reported range varied from
$23.00 to $396.95. The
higher mean costs for this
component in the New England and Northwestern
states—regions I and
X—(compared with the national mean) were statistically
significant (Table 9-2). The
hospitals in the smallest surgi-

cal stratum (100-999) again
reported significantly higher
costs (p=0.0263).

Apheresis Platelets
The mean cost of a unit of
leukocyte-reduced platelets
collected by apheresis (3-4 ×
1011 platelets) was $510.05 in
2004 in comparison with
$475.48 in 2001, a statistically significant increase of
7.3% (p<0.0001). Reported
values ranged from $114.00
to $1,200.00. The mean cost
was lower in the Southeastern
and Midwestern states (regions IV, V, and VII). Regions
that were significantly higher
than the mean were II, VI,
VIII, IX, and X (which include
New York and New Jersey as

well as the Mountain, South
Central, and Western states as
listed in Table 3-6). When
stratified by surgical volume,
there were no statistically significant differences in the cost
of this component.

Reimbursement
The Centers for Medicare and
Medicaid Services (CMS) hospital outpatient prospective
payment system (OPPS) reimbursement rates for the four
components assessed are reported in Table 9-4. While
costs for components increased between 2001 and
2004, outpatient reimbursement decreased for all components during the same period. The reimbursement rate

Table 9-4. CMS Hospital Outpatient Prospective Payment System Rates for Selected Blood
Components
Reimbursement Code
Blood Component

CPT/
HCPCS

Red Blood Cells (leukocyte-reduced)

P9016

Fresh Frozen Plasma (frozen within 24
hours after phlebotomy)

Reimbursement Rate
%
Change

2001*

2004†

0954

$142.17

$119.26

–16.1

P9017

0955

$113.30

$ 95.00

–16.2

Whole-blood-derived platelets

P9019

0957

$ 49.18

$ 41.44

–15.7

Apheresis platelets (leukocyte-reduced)

P9035

9501/1014

$448.87

$408.81

– 8.9

APC

*Department of Health and Human Services. Update of calendar year 2001 hospital outpatient payment rates and
coinsurance amounts effective April 1, 2001. [Available at http://www.cms.hhs.gov/hospitaloutpatientpps/
downloads/apr001a.pdf.]
†Department of Health and Human Services. Medicare program; hospital outpatient prospective payment system; payment
reform for calendar year 2004; interim final rule. (January 6, 2004). Fed Regist 2004;69:819-44.
CMS = Centers for Medicare and Medicaid Services; CPT = current procedural terminology; HCPCS = health-care common
procedure coding system; APC = ambulatory patient classification.

Component Costs

39

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

for a unit of leukocyte-reduced
RBCs decreased 16.1% between 2001 and 2004, while
the mean cost to hospitals increased 30.8%. The smallest
decrease was seen in a unit of
leukocyte-reduced apheresis
platelets, which was reimbursed at a rate of $408.81,
8.9% less than the 2001 rate.

payers, besides Medicare,
pay for blood under varying
mechanisms that are not included in this report.

Summary

In summary, the mean cost of
an RBC unit increased by
30.8% between 2001 and
CMS OPPS rates are reported 2004. Costs for FFP and
apheresis platelets increased
here because they are the
by less than 10% in the same
only simple measure of
Medicare reimbursement for period. Whole-blood-derived
individual blood components. platelets increased in cost by
However, most Medicare re- 16% during the period. The
imbursement for blood is part 2005 survey was the first to
of the diagnosis-related group measure a statistically signifi(DRG) payment made for hos- cant increase in the cost of
pital inpatient services. Other every component assessed.

40

Component Costs

CMS OPPS reimbursement
rates for all components assessed were from 8.9% to
16.2% lower in 2004 than in
2001. Average costs for the
components assessed were
generally higher in regions I
and II (Northeastern states), as
well as VIII, IX, and X (Mountain and Western states), and
lower in regions IV, V, VI and
VII (Southeastern and Central
states). Hospitals with 8,000
or more surgeries per year
typically pay less than the national mean for blood components; this likely reflects more
favorable supplier agreements
based on volume purchases.

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

10. Cellular Therapy Products

Because of ongoing interest in
hematopoietic transplantation
and novel cellular therapies,
the 2005 NBCUS captured
data on collection, processing, and infusion of five types
of cellular therapy products:
hematopoietic progenitor cells
collected by apheresis (HPC-A),
HPCs derived from marrow
(HPC-M), HPCs from cord
blood (HPC-C), lymphocytes,
and cell products generated in
culture.
The survey questions were
nearly identical to those in
previous surveys so that comparisons could be made and
trends identified. This was the
first year that public cord
blood banks were included in
addition to blood centers and
hospitals. The majority of independent cord blood banks
are private in that the HPC-C
units are collected, stored,
and processed at the private
expense—and for the future
potential use—of a family
member. However, there are
some public or allogeneic
cord blood banks and attempts were made to include
them in the survey. Nine cord
blood banks, all AABB mem-

Cellular Therapy Product Terminology*
Previous

Current

Term

Abbr.

Term

Abbr.

Peripheral
Blood
Progenitor
Cells

PBPC

Hematopoietic
Progenitor
Cells-Apheresis

HPC-A

Bone
Marrow

BM

Hematopoietic
Progenitor
Cells-Marrow

HPC-M

Cord Blood

CB

Hematopoietic
Progenitor
Cells-Cord

HPC-C

*Circular of Information for the Use of Cellular Therapy Products (2003).

therapy products collected in
2004, exceeding the collection of HPC-M, lymphocytes,
and cell products generated in
culture. The majority of
HPC-A products were autologous; however, there was a
Collections
26.1% decline in collections
since 2001. The change was
Autologous and allogeneic
seen mostly at the blood cencellular therapy product col- ters, which showed a 60% delections are illustrated in Tacline in autologous HPC-A
bles 10-1 and 10-2, and Figcollections in comparison to
ure 10-1. Collection of HPC-A 2001 with a single large blood
and HPC-C products made up center no longer participating
the largest cohort of cellular
bers thought to have some
public donation activity, were
surveyed and three responded; any unique responses are noted.

Cellular Therapy Products

41

42

Cellular Therapy Products
21
5
1
3
0

HPC-A

HPC-M

HPC-C*

Lymphocytes

Cells Generated in
Culture

*autologous

1,566

0

63

24

20

1,459

Products
Collected

HPC-C products represent private cord blood collections

All Products

No.

Product Type

Blood Centers

5

5

3

30

95

No.

13,497

105

502

91

175

12,624

Products
Collected

Hospitals

Table 10-1. Autologous Cellular Therapy Product Collections Performed

1

No.

2,234

2,234

Products
Collected

Cord Blood Banks

17,297

105

565

2,349

195

14,083

Products
Collected

–15.4

–73.1

248.8

442.5

–49.5

–26.1

% Change
2001-2004

All Facilities

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

19
5
4
12
1

HPC-A

HPC-M

HPC-C

Lymphocytes

Cells Generated in
Culture

All Products

No.

Product Type

4,492

81

273

3,331

63

744

Products
Collected

Blood Centers

1

36

9

49

60

No.

14,681

185

750

10,590

602

2,554

Products
Collected

Hospitals

Table 10-2. Allogeneic Cellular Therapy Product Collections Performed

1

No.

477

477

Products
Collected

Cord Blood Banks

19,650

266

1,023

14,398

665

3,298

Products
Collected

–32.1

2,855.6

40.7

–37.8

–21.7

–21.7

% Change
2001-2004

All Facilities

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Cellular Therapy Products

43

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Cells Generated in Culture
1%
Lymphocytes
4%

HPC-A
48%

HPC-C
45%

HPC-M
2%

Figure 10-1. Collection of all cellular therapy products.

are used as the first line of
treatment before transfusion
with HPC-A; 3) newer
nonmyeloablative protocols
(mini-transplants—always
allogeneic) often involve less
processing and, therefore,
fewer collections; and 4) collection efficiency is increased
as a result of improved mobiChanges in medical practice
that may contribute to the de- lization regimens.
crease in the collection totals
include: 1) the marrow trans- Private (or autologous) HPC-C
plant protocol is no longer
collections (collections inrecommended for treatment
tended for the use of the famof breast cancer patients; 2)
ily from whom they were colnew drugs (including
lected and whose collection
rituximab, an anti-CD20 for
and storage costs are paid by
treatment for non-Hodgkin’s
the family) increased signifilymphoma, and imatinib
cantly. This probably results
mesylate for treatment of
from the inclusion of cord
chronic myelocytic leukemia) blood banks—
in cellular therapy activities.
Allogeneic HPC-A collections
were down 21.7%, from
4,211 to 3,298, although not
back to pre-2001 levels. In
this case, the decrease can be
accounted for by changes in
hospital collection volumes.

44

Cellular Therapy Products

some of which have had
mixed public and private
activity—in the survey. But
overall HPC-C collections
were lower than in the previous survey, a consequence of
a large decrease in allogeneic
collections by blood centers,
with the aforementioned large
blood center no longer participating in cellular therapy activities. It is also possible that
there is underreporting of hospital HPC-C collections when
the collection takes place in
the obstetrics ward and is not
reported to the hospital blood
bank. Increased collections
were also noted for both
autologous and allogeneic
lymphocytes and allogeneic
cells generated in culture.
HPC-M collections continued
the declining trend—perhaps
because HPC-A collection is
so much easier on the donor.

Processing
Processing activity for cellular
therapy products is displayed
in Table 10-3 and Figure
10-2. Processing of HPC-C
units rose by 161.6% since
the last survey, with 32,125
collections processed, representing more than 100% of
total cord blood collections
reported. Again, the inclusion
of cord blood banks contributed much to the increase in
this area of activity.
Underreporting of cord blood
collections is most likely re-

12
6
9
6
1

HPC-A

HPC-M

HPC-C

Lymphocytes

Cells Generated in
Culture

All Products

No.

Product Type

8,278

81

765

6,106

88

1,238

Products
Processed

Blood Centers

Table 10-3. Cellular Therapy Products Processed

6

34

21

50

75

No.

18,204

312

371

3,409

666

13,446

Products
Processed

Hospitals

2

No.

22,610

22,610

Products
Processed

Cord Blood Banks

49,092

393

1,136

32,125

754

14,684

Products
Processed

42.8

–31.7

80.0

161.6

–44.2

–24.9

% Change
2001-2004

All Facilities

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Cellular Therapy Products

45

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Infusion
Cells Generated in
Culture
1%

Issue/infusion activity (Table
10-4 and Figure 10-3) deLymphocytes
2%
creased in comparison to
2001 for all cell therapy product types except HPC-C
HPC-A
30%
(67.3% increase). The declines in infusions of HPC-M
(–54%; p=0.0144) and, predictably, the number of recipients of HPC-M (p=0.0089)
were substantial. There was
HPC-M
also a substantial but not sig2%
nificant decrease in infusions
HPC-C
of other cells generated in
65%
culture (–71.6%). However,
in this survey, two new categories were included under
Figure 10-2. Processing of all cellular therapy products.
the infusion section: “Other
Cellular Products” and “Tissue Products.” Other cellular
products accounted for 5% of
the infusions and 6% of the
recipients. Tissue products accounted for 24.5% of the infusponsible for the high percent- other transplant protocols to
sions and 22.7% of the recipiage of cord blood processing enhance effectiveness of those ents, mostly autologous. The
activities reported. There is al- protocols, and this may con- number of recipients of celluways less than a 1:1 ratio of
tribute to the increase seen.
lar therapy products decollected to processed units,
Most of the increase in procreased for all products exbecause of inadequate
cessing was seen at blood
cept HPC-C. The increase in
collection, contamination,
centers. In 2001 there were
HPC-C recipients (65.1%) reetc. Cord blood banks that re- only two blood centers proflected primarily allogeneic
ported processing but not col- cessing lymphocytes, whereas infusions.
lecting cord blood might be
six centers reported this activprocessing collections from
ity in 2004. Processing activity
several hospitals, perhaps not for all other cell therapy prod- Characterization of
all of them being represented ucts decreased in 2004. New Reporting Facilities
on this survey or having renonmyeloablative protocols
ported on this section.
that call for unprocessed
The relative proportions of
HPC-A and HPC-M may accollection, processing, and inAnother product type for
count for some of this change, fusion activities performed by
which processing activity rose because these protocols often blood centers and hospitals
is lymphocytes (80%). Lymutilize donor lymphocyte infu- are shown in Tables 10-1,
phocytes are given as part of sion.
10-2, 10-3, and 10-4. HPC-A

46

Cellular Therapy Products

17

9

10

7

1

1

0

HPC-A

HPC-M

HPC-C

Lymphocytes

Cells Generated
in Culture

Other Cellular
Products

Tissue Products

All Products

No.

Product Type

1,273

0

18

1

96

263

89

806

Infusion
Episodes

Blood Centers

10

7

5

39

26

55

94

No.

11,576

3,095

625

146

398

312

526

6,474

Infusion
Episodes

Hospitals

1

No.

All Facilities

64

12,913

3,095

643

147

494

639

615

7,280

–4.8

–71.6

–29.1

67.3

–54.0

–31.5

Infusion Infusion % Change
Episodes Episodes 2001-2004

Cord Blood Banks

Table 10-4. Cellular Therapy Products Issued and/or Infused

4,949

7

55

146

46

4

105

4,586

Autologous

5,862

2,391

586

1

334

620

466

1,464

Allogeneic

10,811

2,398

641

147

380

624

571

6,050

Total

No. of Recipients

4.2

–51.2

–23.4

65.1

–55.7

–23.5

% Change
2001-2004

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Cellular Therapy Products

47

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Tissue Products
24%

Other Cellular Products
5%

HPC-A
56%

Cells Generated in Culture
1%
Lymphocytes
4%

HPC-C
5%
HPC-M
5%

Figure 10-3. Infusion of cellular therapy products.

and HPC-M collection, processing, and infusion activities continue to be more
common in hospitals than in
blood centers. The number of
hospitals collecting
autologous HPC-A products
exceeds the number collecting allogeneic HPC-A products (95 vs 60), both reduced
from the previous survey.
Fewer blood centers as well
as hospitals reported collecting cord blood in 2004 than
in 2001. This may be in response to the number of cord
blood banks that have entered the market and the exit
of a major blood center from
the cellular therapies market.

8,000
7,453

Autologous
Allogeneic

7,288

7,000
6,070

Number of Products

6,000

5,000

4,000
3,027
3,000

1,830

2,000

1,194
1,000

744
133 110

304
25

0

0
0-999

1,000-1,399

1,400-2,399

2,400-4,999

5,000-7,999

Hospitals by Annual Surgical Volume

Figure 10-4. Cellular therapy product collections by hospitals (n=101).

48

Cellular Therapy Products

> 8,000

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Table 10-5. Cellular Therapy Product Collections by Hospitals, by Surgical Volume
Surgeries
per Year

No. of
Facilities

% of
Facilities

Collections
Autologous

Allogeneic

Total

0-999

3

2.4

133

110

243

1,000–1,399

2

1.6

25

0

25

1,400–2,399

8

6.5

1,830

7,453

9,283

2,400–4,999

29

23.6

1,194

304

1,498

5,000–7,999

26

21.1

3,027

744

3,771

≥8,000

55

44.7

7,288

6,070

13,358

123

100.0

13,497

14,681

28,178

All Hospitals

The reporting in this survey
may better reflect the areas of
responsibility, with the hospital responsible for the HPC-C
collection (possibly under the
blood center’s protocol) and
the blood center or cord
blood bank responsible for
processing and storage.
Of hospitals reporting collection activity for cellular therapy products, those with
higher surgical volumes gen-

erally were more likely to
have the higher levels of activity, as seen in previous surveys (Figure 10-4 and Table
10-5). The high number of
allogeneic collections seen in
the category of 1,400 to 2,399
surgeries per year can be accounted for primarily by one
large facility.

lection (45.4%) and processing (65%) activities among
surveyed medical facilities,
but only a very small amount
of the infusion activity (3.3%).
The majority of the cord
blood activity is currently associated with the collection
and storage of HPC-C from
unrelated donors to provide a
bank from which potential
As seen in previous surveys,
transplant recipients can idencord blood products represent tify a suitably matched cellua sizeable proportion of collar therapy product.

Cellular Therapy Products

49

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

11. Historical Perspectives

The 2005 NBCUS has allowed for the extension of the
findings of previous nationwide surveys conducted by
the National Blood Data Resource Center in 2002, 2000,
and 1998, and earlier assessments conducted by the Center for Blood Research and the
National Heart, Lung, and
Blood Institute.

period from 1997 to 2001. In
2004, autologous collections
comprised only 0.46 million
units, or 3.0% of total collections, the lowest amount that
has been measured by contemporary surveys.

Figure 11-2 illustrates the
trends in allogeneic WB/RBC
collections and transfusions
over time, as well as the margin between the two curves,
Time Trends
which is discussed below in
the section on Blood Supply
Whole blood (WB) and red
Adequacy. The steep upward
blood cell (RBC) collections
rise of the transfusion curve
for the past 15 years are illus- observed since 1994 reached
trated in Figure 11-1. Total
a maximum of 13.9 million in
collections, which dropped to 2004. The total increase in
a decade low of 12.6 million allogeneic transfusions over
units in 1997, reached 15.3
the past decade was 31.1%.
million in 2001, largely beThe increase in allogeneic
cause of an increased allocollections over the past degeneic donation rate followcade was only 20.3%,
ing the September 11 terrorist however.
attacks. In 2004, total collections remained virtually unThe utilization of platelets
changed at 15.3 million units overall was statistically un(an actual decline of 0.2%).
changed between 2001 and
2004, declining by 3.1%. The
Autologous donations, which continued shift in the mix of
declined dramatically beplatelet types, first observed in
tween 1992 and 1997, ap1999, is notable. Figure 11-3
peared to level off at approxi- illustrates the increased use of
mately 0.6 million units in the apheresis platelets (+10.0%)

and the decline in transfusion
of whole-blood-derived platelet concentrates (–41.2%).

Blood Supply Adequacy
The available supply of both
WB/RBCs and non-RBC components was sufficient to meet
overall transfusion demands
in 2004. Yet, despite the increase in collections achieved
since 1997, there is cause for
concern regarding the adequacy of the United States
(US) blood supply, both currently and in the future.
The shrinking margin between
allogeneic WB/RBC supply
and demand depicted in Figure 11-2 is one reason for
concern. In 1989, allogeneic
collections totaled 13.6 million, with a margin of 1.9 million, 13.8% of supply. By
1997, the margin had decreased to 862,000, 7.2% of
supply. In response to a
sharply increasing demand for
RBCs, blood centers successfully increased allogeneic collections in 1999 to 13.2 million, increasing the margin to
9.1% in spite of an 8.1% in-

Historical Perspectives

51

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Figure 11-1. Allogeneic, autologous, and total whole blood and red blood cell collections,
1989-2004.

Figure 11-2. Allogeneic whole blood and red blood cell collections and transfusions,
1989-2004.

52

Historical Perspectives

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

12,000

Units Transfused (in thousands)

Apheresis platelets*

Whole-blood-derived platelet concentrates

10,000
2,614

8,000
3,396

1,537

3,036

6,000
4,000

7,582
5,640

6,017

1997

1999

8,338

2,000
0
2001

2004

*expressed as platelet concentrate equivalent units.

Figure 11-3. Trends in platelet transfusion.

A similar analysis can be performed using the available
allogeneic supply. The available allogeneic supply is composed of those units that have
passed all laboratory tests and
are available for transfusion
(not illustrated, but mentioned
in the Executive Summary). In
The 2004 data, however, indi- 2004, the available supply of
cate a subsequent reduction
screened allogeneic WB/RBC
in the margin to 6% of supply. 14,560,000 units exceeded
This was the result of a level- transfusions of allogeneic
ing of allogeneic units transWB/RBCs (13,912,000) by
fused to 13.9 million and a re- 648,000 units. This margin
duction in allogeneic supply
was only 4.5%, in comparito 14.8 million, yielding a
son with 6.3% in 2001.
margin of 0.9 million units.
Thus, as illustrated in Figure
In summary, 2004 NBCUS
11-2, the margin between
data indicate some leveling of
supply and demand over the the steep increase in the de15-year period of 1989 to
mand for RBCs previously ob2004 was reduced by 52.6%. served. Collections, however,
remained constant in comparcrease in demand. Collections
increased significantly
(p<0.0001) in 2001 largely in
response to the extraordinary
events of September 11, temporarily boosting the margin
to nearly 1.2 million, or 8.0%
of supply.

ison with 2001, resulting in
further contraction in the margin between supply and demand. The 2004 margin was
the smallest ever measured by
this series of contemporary
national collection and
transfusion surveys.
Through efficient product
management, evidenced by
reduced outdates and low
numbers of hospitals reporting
unmet need, the blood community has maintained adequate supply to meet blood
product needs. The community must remain vigilant,
however, because the margin
is smaller than it has ever
been. We must be able to ensure that the blood supply is
always sufficient for the country’s needs.

Historical Perspectives

53

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

Figure 11-4. Trends in estimated rates of blood collection and transfusion in the United
States, 1980-2004.

US Population Trends

total population of all ages for
that year. Each rate includes
Figure 11-4 illustrates the
the 95% confidence interval
trends in the estimated rates of for the estimate. Population
WB/RBC collection and trans- figures were obtained from
fusion in the US from 1980 to the US Bureau of the Census.
2004. The rate of collection,
the upper line, was calculated Blood collection per thousand
from the national estimate of population of donor age was
total WB and RBC units
85.6 units in 2004 compared
collected per thousand popu- with 88.0 units in 2001 and
lation aged 18 to 65 for a
80.8 units in 1999. In comgiven survey year. The rate of parison with 2001, this was a
transfusion, the lower line,
decline of 2.7%, primarily rewas calculated from the nasulting from the temporary eftional estimate of WB/RBC
fect of additional donations
units transfused per thousand experienced following the

54

Historical Perspectives

September 11, 2001 terrorist
attacks. The US WB/RBC
transfusion rate in 2004 was
49.6 units per thousand population, statistically unchanged
from the rate in 2001 (50.0),
which was the highest rate
observed in 15 years.

Note: Blood collection per
thousand total population in
2004 was 53.5. The age-adjusted value of 85.6 was used
in Figure 11-4 for consistency
with historical methods.

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

12. Acknowledgments

AABB acknowledges the following individuals who contributed their time and considerable
expertise to this project:
AABB
Russell Cotten
Nina Hutchinson
Diane Killion, JD
Kathy Loper, MHS, MT(ASCP)
Laurie Munk, MLS
Philip Schiff, JD
Ashley Smith
Theresa Wiegmann, JD

DHHS
LCDR Richard Henry
Jerry Holmberg, PhD
CAPT Lawrence McMurtry

RTI
Dhuly Ahsan
Kurtida Amin
Kimberly Ault, PhD
Marjorie Hinsdale
Sarah Rosquist
Colleen Waters
Behnaz Whitmire

Acknowledgments

55

THE 2005 NATIONWIDE BLOOD COLLECTION AND UTILIZATION SURVEY REPORT

13. References

National Blood Data Resource Center. Report on Blood Collection and Transfusion in the
United States in 2001. Bethesda, MD: AABB, 2003.
Sullivan MT, Wallace EL. Blood collection and transfusion in the United States in 1999.
Transfusion 2005;45:141-8.
Read EJ, Sullivan MT. Cellular therapy services provided by blood centers and hospitals in the
United States, 1999: An analysis from the Nationwide Blood Collection and Utilization
Survey. Transfusion 2004;44:539-46.
Sullivan MT, McCullough J, Schreiber GB, Wallace EL. Blood collection and transfusion in the
United States in 1997. Transfusion 2002;42:1253-60.
Wallace EL, Churchill WH, Surgenor DM, et al. Collection and transfusion of blood and blood
components in the United States, 1994. Transfusion 1998;38:625-36.
Wallace EL, Churchill WH, Surgenor DM, et al. Collection and transfusion of blood and blood
components in the United States, 1992. Transfusion 1995;35:801-12.
Wallace EL, Churchill WH, Surgenor DM, et al. Collection and transfusion of blood and blood
components in the United States, 1989. Transfusion 1993;33:139-44.
Surgenor DM, Wallace EL, Hao SH, et al. Collection and transfusion of blood and blood
components in the United States, 1982-88. N Engl J Med 1990;332:1646-51.

References

57