E D U C AT I O N A N D A D M I N I S T R AT I O N

Development of a validated exam to assess physician transfusion

medicine knowledge

Richard L. Haspel,1 Yulia Lin,2 Patrick Fisher,3 Asma Ali,3 and Eric Parks3 for the
Biomedical Excellence for Safer Transfusion (BEST) Collaborative

A
number of studies have shown a high rate of
BACKGROUND: There is evidence that physicians lack inappropriate blood product use.1-3 These find-
adequate transfusion medicine knowledge. To design ings are likely in part due to inadequate knowl-
needs-based educational interventions to address this edge of transfusion medicine.4 While medical
gap, a validated assessment tool is required. Previously schools have transfusion medicine curricula, the amount
published exams have not been created or validated of time spent teaching this material can vary greatly.5,6 In
using rigorous psychometric methods. one study from the United States, while 83% of the 86
STUDY DESIGN AND METHODS: A modified Delphi surveyed administrators reported that their school had
method was used to achieve consensus regarding the transfusion medicine lectures, almost half of this teaching
essential knowledge and skills for physicians who trans- amounted to only 1 or 2 hours.5 Only 29% reported small
fuse blood products. To ensure content validity, group sessions related to transfusion medicine. In addi-
members of an international organization of transfusion tion, there are no published data that have determined the
medicine experts (Biomedical Excellence for Safer efficacy of these different curricula.
Transfusion [BEST] Collaborative) participated in the A validated transfusion medicine exam would enable
exam design process. An exam, based on the most a better assessment of educational methods, trainee
highly rated topics, was created and administered to knowledge, and areas for improvement. While transfusion
individuals with a priori expected basic, intermediate, medicine exams have been published, they were not
and expert levels of transfusion medicine knowledge. designed or validated using accepted psychometric
Rasch analysis, a psychometric technique used in high- techniques.7-10 As noted in an editorial on the develop-
stakes medical licensure and board testing, was used ment of one of the more recent of these exams, “several
to determine exam accuracy and precision. critical points about the experimental design and conduct
RESULTS: Thirty-six topics achieved ratings sufficient of the study raise questions about the quality of the results
to be considered for inclusion in the exam (content and/or data and whether all conclusions and recommen-
validity index > 0.8). A 23-question exam was adminis- dations are truly warranted.”11 In this study, we used
tered to 49 individuals. Mean scores for individuals with
expected basic, intermediate, and expert knowledge ABBREVIATIONS: ASCP = American Society of Clinical
were 42, 62, and 82%, respectively (p < 0.0001). The Pathology; CVI = content validity index; PGY = Postgraduate
exam achieved good fit with the Rasch model. Year; TACO = transfusion-associated circulatory overload.
CONCLUSION: A validated exam has now been
From the 1Department of Pathology, Beth Israel Deaconess
created to accurately assess transfusion medicine
Medical Center and Harvard Medical School, Boston,
knowledge. This exam can be used to determine knowl-
Massachusetts; the 2Department of Clinical Pathology,
edge deficits and assist in the design of curricula to
Sunnybrook Health Sciences Centre, and the Department of
improve blood product utilization.
Laboratory Medicine and Pathobiology, University of Toronto,
Toronto, Ontario, Canada; and the 3American Society for
Clinical Pathology (ASCP), Chicago, Illinois.
Address reprint requests to: Richard L. Haspel, MD, PhD,
Beth Israel Deaconess Medical Center, 330 Brookline Avenue,
Yamins 309, Boston, MA 02215; e-mail:
[email protected].
Received for publication May 6, 2013; revision received
June 25, 2013, and accepted July 26, 2013.
doi: 10.1111/trf.12425
TRANSFUSION 2014;54:1225-1230.

Volume 54, May 2014 TRANSFUSION 1225

HASPEL ET AL.

rigorous analytical methods to develop a validated trans- ASCP oversees both the Pathology Resident In-Service
fusion medicine exam. Exam (RISE) and certification exams for laboratory tech-
nologists.14 As such, the organization has decades of expe-
rience in exam design and psychometrics. Following
MATERIALS AND METHODS
guidelines for the Pathology RISE, BEST members submit-
Identifying topics for the exam ted questions for inclusion in the exam. To focus on the
Exam content was determined using a modified Delphi most pertinent issues and limit exam length, questions
method that is a structured approach for achieving con- were primarily based on topics with a CVI of greater
sensus among experts.12 The experts were members of the than 0.9 although topics with a CVI of greater than 0.8
Biomedical Excellence for Safer Transfusion (BEST) Col- were also included. All submitted questions required a
laborative, an international research organization that reference supporting the correct answer. These questions
works to explore ways to improve transfusion practice were vetted by the authors and refined to create a
(http://www.bestcollaborative.org). The group, represent- 23-question pilot exam. Question selection was based on
ing a diversity of interests and high levels of accomplish- quality and to ensure diversity in the topics covered.
ment, has scientific and blood center members from 10
countries. While there is some turnover, during the survey
period described below, approximately 80% of the mem- Piloting the exam
bership remained constant. The exam was piloted on individuals from three different
BEST members were presented with an initial survey hospitals (one in Canada and two in the United States).
that asked the following open question: The following individuals were included and divided into
“What knowledge or skills related to transfusion three a priori categories:
medicine are absolutely essential for physicians who are
1. Expected basic knowledge: Postgraduate Year-1
not transfusion medicine specialists (e.g., internists, car-
(PGY-1) internal medicine and pathology residents.
diologists, surgeons, neurologists) but whose practice
2. Expected intermediate knowledge: non–transfusion
includes the transfusion of blood products? (List as many
medicine specialists with an interest in transfusion
as 10 items).”
medicine (e.g., members of hospital transfusion com-
To provide specific responses the following additional
mittees, hematologists, anesthesiologists, critical care
instructions were given:
physicians); PGY-2 and above clinical pathology or
“Please be as specific as possible and avoid broad
anatomic and clinical pathology residents or fellows.
subject headings. Examples: Too non-specific (avoid if
3. Expected expert knowledge: transfusion medicine
possible): ‘Adverse reactions’; ‘Immunohematology’; ‘Indi-
physicians.
cations for red cell transfusion’; Specific (preferred): ‘Post-
transfusion purpura’; ‘Testing for a biphasic hemolysin’; The exam was administered through SurveyMonkey.
‘Red cell transfusion threshold for (a specific indication).’ ” Although several demographic questions were asked, no
After removal of redundant and nonspecific identifying information was requested. As such, the exam-
responses, this list of topics was then collated for a second inees remained anonymous.
survey in which BEST members were asked to rate the
randomly presented topics on a scale of 1 (very low impor-
tance) to 6 (very high importance/essential) for inclusion Statistical analysis of pilot data
on an exam to be administered to all physicians who Classical test theory involves direct comparison of scores
transfuse blood products. As the results from this survey (i.e., percent correct) among examinees as an indication of
would be used to determine content for a physician exam, accuracy (i.e., experts should have higher mean scores).15
only results from BEST physician members were used to For this purpose, we determined mean scores for each
calculate a content validity index (CVI) for each rated expected expertise group and analyzed the data using a
item. The CVI is equal to number of respondents rating a one-way analysis of variance (ANOVA; Microsoft Excel,
topic a 4, 5, or 6 divided by the total number of respon- Seattle, WA).
dents. A CVI of greater than 0.8 was considered appropri- To provide a more detailed assessment of accuracy
ate for inclusion on the exam.13 All surveys were and reliability, Rasch psychometric analysis was per-
administered through SurveyMonkey (surveymonkey formed.15-17 This approach provides greater information
.com) and were anonymous. than classical test theory in regard to specific item diffi-
culty and student ability. Rasch analysis is routinely used
in high-stakes testing such as for medical certification and
Developing exam questions licensure.17
The exam questions were developed in conjunction with The Rasch model assumes a logarithmic relationship
the American Society of Clinical Pathology (ASCP). The between student ability and item difficulty:

1226 TRANSFUSION Volume 54, May 2014

 TRANSFUSION MEDICINE EXAM

Probability = 1/(1 + exp( −(ability − difficulty ))).

TABLE 1. Demographics of survey respondents
Using the above formula, for a question with a set Open-question Rating
Characteristic survey survey
difficulty, one can predict how often individuals with a
Respondents 34 27
particular ability should select the correct answer. For Response rate (%) 48 36
example, an individual with the highest ability would be Countries represented 10 10
Degrees (%)
expected to get the answer correct almost 100% of the
MD 55 63
time while an individual with average ability would be MD/PhD 30 27
expected to get the question right only 50% of the time. PhD 15 0
Specialty (%)
Rasch analysis determines how well, from exam adminis-
Transfusion medicine 66 78
tration data, each question and each individual fit the Hematology 22 15
model. Other 12 7
Work setting (%)
Fit can be determined using chi-square analysis for
Blood center 28 33
the comparisons of the acquired data to the Rasch model. Hospital blood bank 41 30
This value is related to the measured variance divided by Hospital transfusion service 24 26
Other 7 11
the predicted variance adjusted for the number of obser-
Experience in years 24 [0-40] 24 [7-40]
vations. As such, a question fit value of 1.3 indicates there (mean [range])
is 30% more variance than predicted by the Rasch model.16
While recommendations vary for acceptable question fit
scores, there is general agreement that questions with a fit
value of more than 1.5 have more variance than expected presented and deemed acceptable by the BEST member-
and should be reevaluated.18 While not necessarily indi- ship at the October 2010 BEST meeting.
cating an item needs to be removed, the basis for the
misfit should be determined during the question reevalu-
ation process. Exam development, piloting, and validation
Fit values for examinees can also be determined. In An exam consisting of 23 questions was created with input
addition, an overall examinee reliability measure can be from BEST Collaborative members and the ASCP. The
calculated and interpreted in a similar fashion to topic for each question and the answer reference are
Cronbach’s alpha.19 The Rasch model’s value for reliability shown in Table 3. All of the questions were based on topics
tends to be an underestimate when compared to with at least a CVI of more than 0.8 and 15 of the 23 ques-
Cronbach’s alpha.20 tions (65%) were based on topics with a CVI of more
than 0.9. Twelve of the 16 topics with a CVI of more
than 0.9 were included in the exam.
RESULTS To pilot the exam, 101 individuals were sent an e-mail
invitation. Forty-nine individuals completed the exam for
Determining topics for exam inclusion an overall response rate of 49%. Demographics of the
Of the 71 BEST members e-mailed, 34 responded (48%) to respondents are shown in Table 4. The group of individu-
the first open-question survey. Characteristics of the als with expected intermediate knowledge was made up of
respondents are shown in Table 1. There were 289 poten- 19 pathology residents at an average training level of
tial topics listed for an average of approximately nine per PGY-3 (range, 2-5) and nine non–transfusion medicine
respondent. After removing redundant and nonspecific physicians (five hematology-oncology, two medicine, one
responses from the first survey, the second rating survey anesthesiology, one critical care). The mean scores and
consisted of 78 topics. Of 76 BEST members e-mailed, 36 ranges of the three a priori defined basic, intermediate,
responded and of these 27 were physicians (36%). Char- and expert knowledge categories were 42 (30-48), 62 (30-
acteristics of the survey respondents are shown in Table 1. 87), and 82% (61%-96%). Demonstrating accuracy, a one-
As the topics for rating were those already indicated as way ANOVA demonstrated a significant difference in exam
important by BEST members, CVI scores showed a distri- scores among the three groups (p < 0.0001).
bution to the right (median, 0.77). However, there was evi- In regard to Rasch analysis, the mean examinee fit
dence of topic discrimination with only 36 of the initial 74 and item fit were 0.99 and 1.00, respectively. The exam
topics achieving a CVI of more than 0.8 and 16 achieving a reliability was 0.79. One question demonstrated a fit score
CVI of more than 0.9 (Table 2). In addition, although the above 1.5. This question (Question 20) involved the appro-
topics were presented randomly, related topics were priate treatment for a patient with a high international
ranked highly (e.g., management of dyspnea, transfusion- normalized ratio on warfarin. Only 42% of transfusion
associated circulatory overload [TACO], and transfusion- medicine experts chose the correct response compared to
related acute lung injury [TRALI]). The topics were 56% of PGY-1 residents. Not all transfusion medicine

Volume 54, May 2014 TRANSFUSION 1227

HASPEL ET AL.

DISCUSSION
TABLE 2. Topics with CVI of more than 0.8 and more than 0.9*
Transfusion thresholds The first step in curricular design is a
*RBC transfusion thresholds in acute anemia
needs assessment,37 that is, determining
*RBC transfusion thresholds for nonbleeding, hospitalized patients without cardiac
conditions the extent of an educational problem
RBC transfusion thresholds in hospitalized patients with cardiac conditions and what specific areas may need
*PLT transfusion thresholds for invasive or surgical procedures
improvement. Although there is evi-
*PLT transfusion thresholds for bleeding patients
PLT transfusion thresholds for prophylaxis (nonbleeding patient) dence of limited and ineffective physi-
*Plasma transfusion thresholds for invasive or surgical procedures cian training in transfusion medicine,
Plasma transfusion thresholds for prophylaxis (nonbleeding patients)
there are no available validated tools to
Indications for cryoprecipitate
Understanding that there is no target Hb level at which an RBC transfusion is indicated accurately perform a needs assessment.
in all patients In this study, we have created a rigor-
*Understanding the relationship between Hb level, tissue oxygenation, and role of RBC
ously validated exam to gauge transfu-
transfusion
*Principles of bedside assessment of a patient’s need for transfusion sion medicine knowledge.
Transfusion reactions While transfusion medicine exams
*Diagnosis and management of TRALI
have been published, they have gener-
*Diagnosis and management of TACO (fluid overload)
*Management of dyspnea during and/or after transfusion ally been created by small groups of
*Understanding processes for reporting transfusion reactions individuals without use of an accepted
Diagnosis and management of acute hemolytic transfusion reactions
approach for reaching expert con-
Diagnosis and management of transfusion-related anaphylaxis
Diagnosis and management of allergic transfusion reactions (nonsimple urticarial, sensus.7-10 In our study, to ensure
nonanaphylaxis) content validity, we used a modified
Management of fever during and/or after transfusion
Delphi method with a large interna-
Understanding how to obtain informed consent for transfusion
Understanding transfusion-transmitted infectious disease risk tional group of transfusion medicine
Safe administration of transfusions experts, with an average of more than
*Understanding the importance of correct recipient identification
20 years of practice experience, to deter-
*Understanding the importance of proper collection of blood samples for transfusion
Knowing correct procedures for ordering or requesting blood products mine exam topics. The anonymity
Knowing how to appropriately transfuse blood products (e.g., rate, fluid compatibility) associated with this method allows for
Nonserologic testing
the best possible input as there is
*Understanding relationship of PLT counts to bleeding risk
Knowing the dose response to RBC transfusion (1 unit = 1 g/dL Hb increase) no concern for stronger personalities
Understanding causes or interpretation of prolonged PT/PTT driving content.12 Previous exams have
Blood bank testing
also had minimal assessment of accu-
Understanding risks of un-cross-matched blood
Principles of ABO/Rh blood group compatibility for RBCs racy and precision. We used Rasch
Other analysis, a tool used in validation
*Principles of warfarin reversal (e.g., blood products, vitamin K)
of high-stakes medical testing, to dem-
*Management of patients requiring massive transfusion
Indications for irradiated blood onstrate the exam has adequate
Knowing how and when to contact a transfusion medicine specialist performance to distinguish diffe-
Indications for using iron as opposed to blood transfusion
rent transfusion medicine knowledge
PT = prothrombin time; PTT = partial thromboplastin time. levels.15-17 Rasch analysis also enabled
question removal without affecting
exam quality.
specialists may be familiar with nontransfusion manage- One possible limitation of our study is that our
ment of warfarin as they may not routinely order medica- content experts were primarily made up of transfusion
tions for patients. After input from BEST membership the medicine specialists and not the end-users (i.e., the phy-
item was retained, as it was agreed that such knowledge is sicians transfusing the products). As the transfusion medi-
very important for patient management related to trans- cine experts that determined content came from a variety
fusion medicine. of practices and countries, we believe that their expert
Rasch analysis also allows determination of the effect opinions represent a broad enough perspective to ensure
of question removal on exam quality. Based on question appropriate topics for inclusion. As inclusion of topics was
length, topic and fit values, we selected three questions based on relevance for all physicians who transfuse blood
(Questions 5, 11, and 23) for which we believed removal products, at the very least, the core topic list and exam can
would shorten time to complete the exam without affect- be used as a framework to build on in developing
ing quality. Upon reanalysis of this 20-item test, a one-way specialty-specific curricula and assessment tools.
ANOVA still demonstrated a p value of less than 0.0001. In summary, we have created, to our knowledge, the
The average item and examinee fit values were 1.0 and first extensively validated transfusion medicine exam. We
0.99. The reliability was 0.80. believe that the topic list will be helpful in curriculum

1228 TRANSFUSION Volume 54, May 2014

 TRANSFUSION MEDICINE EXAM

TABLE 3. Exam question topics and references

Question Topic(s) Reference
1 RBC transfusion thresholds in acute anemia* Napolitano et al.21
2 Management of dyspnea, diagnosis and management of TACO* Mazzei et al.22
3 Indications for irradiated blood Alyea and Anderson23
4 PLT transfusion thresholds for invasive or surgical procedures* Dzik24
5 RBC transfusion thresholds in hospitalized patients with cardiac conditions Carson et al.25
6 Management of dyspnea, diagnosis and management of TRALI* Kopko and Popovsky26
7 PLT transfusion thresholds for invasive or surgical procedures* Nester and AuBuchon27
8 Diagnosis and management of acute hemolytic transfusion reactions Davenport28
9 RBC transfusion thresholds for nonbleeding, hospitalized patients without cardiac conditions* Menitove29
10 Diagnosis and management of allergic transfusion reactions (nonsimple urticarial, Vamvakas30
nonanaphylaxis)
11 Plasma transfusion thresholds for invasive or surgical procedures* Dzik24
12 Understanding processes for reporting transfusion reactions* Heddle and Webert31
13 Understanding transfusion-transmitted infectious disease risk Klein and Anstee32
14 Plasma transfusion thresholds for invasive/surgical procedures* Dzik24
15 Understanding processes for reporting transfusion reactions* Kopko and Popovsky26
16 RBC transfusion thresholds in acute anemia* Napolitano21
17 Understanding transfusion-transmitted infectious disease risk Ramirez-Arcos and Goldman33
18 Management of patients requiring massive transfusion* Nester and AuBuchon27
19 PLT transfusion thresholds for prophylaxis Klein and Anstee34
20 Principles of warfarin reversal* Keeling et al.35
21 Understanding the importance of correct recipient identification and understanding the Davenport28
importance of proper collection of blood samples for transfusion*
22 Management of dyspnea, diagnosis and management of TRALI* Kopko and Popovsky26
23 Knowing the dose response to RBC transfusion (1 unit = 1 g/dL Hb increase) Circular of Information36
* CVI of more than 0.9.

REFERENCES
TABLE 4. Demographics of pilot
exam respondents* 1. Wallis JP, Dzik S. Is fresh frozen plasma overtransfused in
Respondents 49/49 the United States? Transfusion 2004;44:1674-5.
A priori expected knowledge level 2. Pantanowitz L, Kruskall MS, Uhl L. Cryoprecipitate. Pat-
Basic 9 (18)
Intermediate 28 (57) terns of use. Am J Clin Pathol 2003;119:874-81.
Expert 12 (24) 3. Bennett-Guerrero E, Zhao Y, O’Brien SM, et al. Variation
Self-rated knowledge level in use of blood transfusion in coronary artery bypass graft
Beginner 16 (33)
Intermediate 21 (43) surgery. JAMA 2010;304:1568-75.
Advanced 8 (16) 4. Salem-Schatz SR, Avorn J, Soumerai SB. Influence of
Expert 4 (8) knowledge and attitudes on the quality of physicians’
Country
Canada 31 transfusion practice. Med Care 1993;31:868-78.
United States 69 5. Karp JK, Weston CM, King KE. Transfusion medicine in
* Data are reported as number/response rate %, number (%), or American undergraduate medical education. Transfusion
percent. 2011;51:2470-9.
6. Duguid J, Copplestone JA, CMO’s National Blood Transfu-
sion Committee. Teaching transfusion in UK medical
design and the exam will be useful for individuals trying to schools: a survey by the National Blood Transfusion Com-
assess transfusion medicine knowledge in their work or mittee. Med Educ 2008;42:439.
educational setting. The exam is available by contacting 7. O’Brien KL, Champeaux AL, Sundell ZE, et al. Transfusion
the corresponding author. medicine knowledge in Postgraduate Year 1 residents.
Transfusion 2010;50:1649-53.
8. Gharehbaghian A, Javadzadeh Shahshahani H, Attar M,
ACKNOWLEDGMENT et al. Assessment of physicians’ knowledge in transfusion
medicine, Iran, 2007. Transfus Med 2009;19:132-8.
The authors thank Mark Fung, MD, PhD, for assistance in recruit-
9. Rock G, Berger R, Pinkerton P, et al. A pilot study to
ing individuals to take the pilot exam.
assess physician knowledge in transfusion medicine.
Transfus Med 2002;12:125-8.
10. Arinsburg SA, Skerrett DL, Friedman MT, et al. A survey to
CONFLICT OF INTEREST
assess transfusion medicine education needs for clini-
The authors report no conflicts of interest or funding sources. cians. Transfus Med 2012;22:44-9.

Volume 54, May 2014 TRANSFUSION 1229

HASPEL ET AL.

11. Strauss RG. Transfusion medicine education in medical 25. Carson JL, Terrin ML, Noveck H, et al., FOCUS Investiga-
school: only the first of successive steps to improving tors. Liberal or restrictive transfusion in high-risk patients
patient care. Transfusion 2010;50:1632-5. after hip surgery. N Engl J Med 2011;365:2453-62.
12. Edgren G. Developing a competence-based core curricu- 26. Kopko PM, Popovsky MA. Transfusion related acute lung
lum in biomedical laboratory science: a Delphi study. injury. In: Popovsky MA, editor. Transfusion reactions. 4th
Med Teach 2006;28:409-17. ed. Bethesda (MD): American Association of Blood Banks
13. Polit DF, Beck CT, Owen SV. Is the CVI an acceptable Press; 2012. p. 191-216.
indicator of content validity? Appraisal and recommenda- 27. Nester. T, AuBuchon JP. Hemotherapy decisions and their
tions. Res Nurs Health 2007;30:459-67. outcomes. In: Roback JD, Grossman BJ, Harris T, et al.
14. Owen SV, Rinder HM, Grimes MM, et al. Senior pathology editors. Technical manual. 17th ed. Bethesda (MD):
resident in-service examination (RISE) scores correlate American Association of Blood Banks Press; 2011.
with outcomes of the American Board of Pathology (ABP) p. 571-616.
certifying examinations. Am J Clin Pathol 2011;136:499- 28. Davenport RD. Hemolytic transfusion reactions. In:
506. Popovsky MA, editor. Transfusion reactions. 4th ed.
15. Tavakol M, Dennick R. Psychometric evaluation of a Bethesda (MD): American Association of Blood Banks
knowledge based examination using Rasch analysis: an Press; 2012. p. 1-52.
illustrative guide: AMEE Guide No. 72. Med Teach 2013; 29. Menitove JE. Red cell transfusion therapy in anemia. In:
35:e838-48. Mintz P, editor. Transfusion therapy. 3rd ed. Bethesda
16. Yu CH. A simple guide to the Item Response Theory (IRT) (MD): American Association of Blood Banks Press; 2011.
and Rasch modeling. 2012. [cited 2013 Apr 7]. Available p. 37-54.
from: http://www.creative-wisdom.com/computer/sas/ 30. Vamvakas EC. Allergic and anaphylactic reactions. In:
IRT.pdf Popovsky MA, editor. Transfusion reactions. 4th ed.
17. De Champlain AF. A primer on classical test theory and Bethesda (MD): American Association of Blood Banks
item response theory for assessments in medical educa- Press; 2012. p. 99-148.
tion. Med Educ 2010;44:109-17. 31. Heddle NM, Webert KE. Febrile non-hemolytic transfu-
18. Linacre JM. Misfit diagnosis: infit outfit mean-square sion reactions. In: Popovsky MA, editor. Transfusion
standardized. 2013. [cited 2013 Apr 7]. Available from: reactions. 4th ed. Bethesda (MD): American Association
http://www.winsteps.com/winman/index.htm of Blood Banks Press; 2012. p. 53-98.
?diagnosingmisfit.htm 32. Klein HG, Anstee DJ. Mollison’s blood transfusion in
19. Tavakol M, Dennick R. Making sense of Cronbach’s alpha. clinical medicine. 11th ed. Malden: Blackwell Publishing;
Int J Med Educ 2011;2:53-5. 2005. p. 701-73.
20. Linacre JM. Reliability and separation of measures. 2013. 33. Ramirez-Arcos S, Goldman M. Bacterial contamination.
[cited 2013 Apr 7]. Available from: http://www.winsteps In: Popovsky MA, editor. Transfusion reactions. 4th ed.
.com/winman/index.htm?reliability.htm Bethesda (MD): American Association of Blood Banks
21. Napolitano LM, Kurek S, Luchette FA, et al. Clinical Press; 2012. p. 153-90.
practice guideline: red blood cell transfusion in adult 34. Klein HG, Anstee DJ. Mollison’s blood transfusion in
trauma and critical care. Crit Care Med 2009;37:3124-57. clinical medicine. 11th ed. Malden: Blackwell Publishing;
22. Mazzei. CA, Popovsky MA, Kopko PM. Noninfectious 2005. p. 611-65.
complication of blood transfusion. In: Roback JD, 35. Keeling D, Baglin T, Tait C, et al. Guidelines on oral anti-
Grossman BJ, Harris T, et al. editors. Technical manual. coagulation with warfarin—fourth edition. Br J Haematol
17th ed. Bethesda (MD): American Association of Blood 2011;154:311-24.
Banks Press; 2011. p. 727-62. 36. American Association of Blood Banks, American Red
23. Alyea EP, Anderson KC. Transfusion associated graft Cross, America’s Blood Centers, and the Armed Services
versus host disease. In: Popovsky MA, editor. Transfusion Blood Programs. Circular of information for human blood
reactions. 4th ed. Bethesda (MD): American Association and blood components. 2013. [cited 2013 Apr 7]. Available
of Blood Banks Press; 2012. p. 217-38. from: http://www.aabb.org/resources/bct/Documents/
24. Dzik W. Component therapy before bedside procedures. coi0809r.pdf
In: Mintz P, editor. Transfusion therapy. 3rd ed. Bethesda 37. Kern DE, Thomas PA, Howard DM, et al. Curriculum
(MD): American Association of Blood Banks Press; 2011. development for medical education. Baltimore (MD):
p. 1-36. Johns Hopkins University Press; 1998.

1230 TRANSFUSION Volume 54, May 2014