Context.—

Opening a new hospital is a once in a lifetime experience and can be very inspiring for those involved in its activation. However, establishing a safe transfusion practice in a greenfield environment comes with unique challenges and opportunities.

Objective.—

To highlight critical activation components such as on-boarding of new personnel, establishing clinical practices, and integrating critical laboratory software.

Design.—

Our staff initially faced challenges in standardizing transfusion medicine clinical practice inside the laboratory. Our efforts were mainly focused on the appropriate use of various transfusion orders, creating comprehensive policies for type and screening, cost effective utilization of blood products, and establishment of the maximum surgical blood order schedule. The transfusion service was launched with 2 information technology programs that separately facilitated steps in the transfusion process, but did not provide centralized access to the entire process. In these circumstances, we partnered with the laboratory information system team to create a series of interfaces that streamlined each system's functionality and implemented the existing infrastructure with upgrades that enable remote location and management of blood products.

Results.—

The transfusion medicine team spent more than a year training and monitoring workflows to avoid individual variations between technologists and to adopt our own standards of practice. Participation in a structured training plan was also necessary between clinical caregivers to know the safe and efficient use of these standards.

Conclusions.—

Although laboratory and clinical staff are knowledgeable in care delivery, it is always a learning experience to establish a new system because of the natural tendency of resorting to previous practices and resistance to new approaches.

Cleveland Clinic, Ohio, is a surgery-intensive organization that is ranked as the No. 2 hospital across the United States.1  Cleveland Clinic Abu Dhabi (CCAD) is the first international extension of an American hospital outside the United States, providing a Cleveland Clinic model of care in the United Arab Emirates (UAE). CCAD opened its doors in March 2015 as a quaternary-tertiary care hospital. Its mission is to provide compassionate, patient-centered care of the highest quality to the UAE population. With “Patients First” as its principal core value, CCAD seeks to provide the best possible care and outcome for every patient. Presently, an average of 2000 patients are seen in outpatient clinics and at least 100 surgical procedures are performed daily.

Opening a new hospital is always an adventure for those involved, and establishing a safe transfusion practice has its own challenges and opportunities. In our journey, the foremost challenge was having newly on-boarded caregivers with different backgrounds, experiences, and with varying levels of clinical practice. Moreover, many had little to no experience with the systems and information technology (IT) implemented at CCAD. Finally, the capabilities of the IT systems were not fully realized at the onset, leading to extensive work around and rework post activation.

In preparation for the hospital opening, from September 2014 until February 2015, policies and procedures were written to define transfusion medicine standards of practice, and at the same time, clinical workflows were formulated by allied health, nursing, and physician teams. At this time, interfacing IT systems began and physicians' orders were created in the electronic medical record. Also, blood bank technologists (BBTs) were hired progressively during the first year (Figure 1). As activation of the Transfusion Medicine Service (TMS) drew closer, we recognized that staffing and streamlining processes and IT systems were our main areas of opportunity. In this article, we describe our experience in setting up the TMS at CCAD and the solutions that were implemented as a result.

Figure 1

Personnel and transfusion activity. First onboarding indicates when the first 3 blood bank technologists arrived to the hospital. Abbreviations: AABB, American Association of Blood Banks; CABG, coronary artery bypass grafting; CAP, College of American Pathologists; ED, emergency department; HAAD, Health Authority of Abu Dhabi; ISO, International Organization for Standardization; JCI, Joint Commission International.

Figure 1

Personnel and transfusion activity. First onboarding indicates when the first 3 blood bank technologists arrived to the hospital. Abbreviations: AABB, American Association of Blood Banks; CABG, coronary artery bypass grafting; CAP, College of American Pathologists; ED, emergency department; HAAD, Health Authority of Abu Dhabi; ISO, International Organization for Standardization; JCI, Joint Commission International.

Close modal

Unlike other pathology subspecialties, the primary activity of the TMS is therapeutic and not diagnostic. As a general statement, TMS does not produce results, but produces actions contributory to the safe treatment of patients. Given this, the practice is generally directed to solving everyday problems and not in issuing results obtained through diagnostic equipment. It is essential that BBTs have certain qualities that allow them to work in an environment that is different from the diagnostic functions of the laboratory. With this premise, the selection of personnel was made with criteria such as qualifications and experience, resilience in the face of rapid change, ability to work in high-stress conditions, ability to focus on critical tasks, ability to adapt to a multicultural work environment, and ability to communicate well both verbally and in writing.

Staff numbers were calculated to deliver an uninterrupted 24/7 TMS. The lead technologist and the medical director were the first onboard to guide the service's establishment. Given the requirement of developing TMS (Tables 1 and 2), it was decided to have a dedicated TMS workforce. In an effort to have a highly knowledgeable and dedicated team, the decision was made not to offer cross-training for BBTs in the core laboratory. The staff hired was very diverse (Philippines, 5; United Kingdom, 3; United States, 3; New Zealand, 1; Canada, 1; and Pakistan, 1), and all caregivers had a median of 10 years' experience in the field. The most critical decision was for the staff to agree on a standardized laboratory practice and avoid individual variations. As a team, they strived to bridge cultural gaps related to communication, evaluation, decision-making, and how to escalate these items throughout the organization's chain of command.2 

Table 1

Policies and Decisions in the Laboratory Process Control in Transfusion Medicine Service

Policies and Decisions in the Laboratory Process Control in Transfusion Medicine Service
Policies and Decisions in the Laboratory Process Control in Transfusion Medicine Service
Table 2

Policies and Decisions in the Clinical Process Control in Transfusion Medicine Service

Policies and Decisions in the Clinical Process Control in Transfusion Medicine Service
Policies and Decisions in the Clinical Process Control in Transfusion Medicine Service

Difficulties With Standardization of Practice Inside the Blood Bank

Transfusion medicine is an essential hospital service that requires accurate adherence to standards of practice. A number of international organizations (eg, American Association of Blood Banks [AABB], International Society for Blood Transfusion, European Blood Alliance, Council of Europe, World Health Organization) have established practice standards that assure safe transfusion practice.35  However, it is essential that a single system of practice be implemented consistently throughout an organization. Practice variations, stemming from following different sets of standards, could lead to unstructured methodology and unreliable results.

While the policies and procedures at CCAD were modeled after AABB standards, the practice was seen to vary from technologist to technologist in the Blood Bank. To further understand and resolve this dynamic, an internal audit was conducted that involved direct observation of performance of technical procedures. The audit confirmed variations in practice (55.5% of BBTs committed deviations from standard operating procedures) and highlighted an urgent awareness of requirement for standardization.

It was quickly realized that in the absence of rigorous training, the technologists defaulted to their previous experiences for task completion. This was not surprising as most technologists were not familiar with AABB standards.

To improve the quality of performance in our practice, the team used process improvement tools that were designed to reduce errors, variations, and waste. The overall aim was to compare the data from the preaudit and postaudit and provide an evidence-based project on improvement and standardization. The Lean Six Sigma approach offered the perfect collaborative methodology to achieve our project goal. Figure 2 examines all the possible reasons why BBTs were deviating from the standard operating procedures established by the TMS. To visually highlight this in greater detail, the team used the Ishikawa diagram (also known as fishbone diagram). Six major factors that contributed to committing deviations, namely, communication, people, method, management, process, and environment, were identified. From these factors, we further subclassified them into more detailed reasons, using the Ishikawa diagram.

Figure 2

Ishikawa diagram. Reasons why blood bank technologists (BBTs) deviate from standard operating procedures (SOP) in transfusion medicine.

Figure 2

Ishikawa diagram. Reasons why blood bank technologists (BBTs) deviate from standard operating procedures (SOP) in transfusion medicine.

Close modal

A Pareto chart was created to prioritize and focus on those areas with the most number of deviations: a second check, patient labeling, confirmation of lots and expiry dates, and patient history check (Figure 3).

Figure 3

Pareto chart of standard operating procedures' deviations. Columns indicate number of deviations observed and line cumulative percentage.

Figure 3

Pareto chart of standard operating procedures' deviations. Columns indicate number of deviations observed and line cumulative percentage.

Close modal

The hospital laboratory underwent a licensing inspection by the Health Authority of Abu Dhabi (HAAD) before hospital opening.6  To comply with HAAD standards, the Blood Bank team drafted and approved a number of policies and procedures that defined the services' scope of practice. Given the time constraints and pressures for hospital activation, not all documents, although approved, were fully vetted and finalized. Consequently, it was difficult to identify which documents were 100% valid for clinical practice. In addition, some of the documents did not originate from AABB practice standards. This problem was also compounded by the fact that management of these documents was manual and version control was not consistently tracked across the service. To address the issue of identifying the current version of the document, an IT document management program was implemented to control recent versions and updates. All essential procedures were revised soon after to appropriately reflect AABB standards. To ensure understanding and compliance amongst the team, competency assessment tools and quizzes for transfusion medicine policies and procedures were created. Four months later, a follow-up audit found BBTs were aware and up-to-date with all relevant standard operating procedures and the practice had been standardized (Table 3) according to AABB requirements.

Table 3

From the Brainstormed Solutions, We Identified the Root Causes of Main Deviation and Created an Action Plan

From the Brainstormed Solutions, We Identified the Root Causes of Main Deviation and Created an Action Plan
From the Brainstormed Solutions, We Identified the Root Causes of Main Deviation and Created an Action Plan

There are a number of policies and decisions that initially must be established both within the management of our practice in laboratory and also in the administration of blood and blood components (Tables 1 and 2). From the beginning of our journey, the decision to follow the AABB standards was unanimous; however, the regulations did not always define rules around compliance of a certain requirement.

As previously mentioned, CCAD was the first international expansion of an entire US-based hospital to transfer its culture outside of the United States. This endeavor became an ever-changing goal, as a significant number of caregivers lacked previous Cleveland Clinic “main campus” or even US-based experience.

The first wave of physicians recruited for CCAD consisted of 176 physicians from the following geographic areas: North America, 105 (59%); Europe, 49 (28%); Arabic countries, 11 (6%); Australia/New Zeeland, 1 (1%); Africa, 1 (1%); and Asia 9 (5%).

Given that diversity, it was imperative for our team to establish and ensure homogeneity of the clinical practice of different users. Among these challenges, the following can be highlighted:

  1. No clear understanding of different orders for blood ordering (crossmatch and hold, prepare red blood cells, transfuse red blood cells) existed at the time of activation. When the order is released in the electronic medical record, it interfaces with the Laboratory Information System (LIS) and becomes visible to the Blood Bank. Thus, a Crossmatch and Hold Order for elective surgery is only visible to the Blood Bank once released. Often, the ordering physician would sign the order and schedule a release on the day of admission (along with another series of medical orders). This would result in TMS receiving pertinent clinical information on the day of surgery, making inventory management and timely preparation of blood difficult.

  2. Another important aspect that needed attention was the policy for Type and Screening (TYSC) in patients scheduled for surgery. At CCAD, patients are required to undergo 2 independent samplings for ABO typing before assignment of the ABO group. Additionally, irregular antibody determination can be completed up to 30 days before surgery, but only in cases of nontransfusion or recent obstetric history (in the last 90 days). According to this approved workflow, samples for TYSC in scheduled surgeries should be obtained between 5 and 30 days before surgery. In the first 6 months of hospital operations, 54% of TYSC orders received in the Blood Bank were delayed. On the day of surgery, 7% were received; on the evening before scheduled surgery, 17%; 2 to 4 days before surgery, 16%; and in 14% of the cases, the specimen was received too early (more than 30 days before surgery). Meetings were held with the Surgical Committee and IT managers to bridge training gaps and collaborate in areas of further process improvement. At the present time, almost all scheduled patients for surgery have a TYSC resolved 5 days before surgery.

  3. There were obstacles faced in establishing an efficient blood inventory approach (expiration rate, discarded rate, effective utilization of blood products). Blood is a perishable product and as such, responsible stewardship will not only provide direct value to patients, but also minimize expenses for the organization. Managing the blood inventory is a balance between scarcity and waste. The challenge is to keep enough stock to ensure 100% blood supply while maintaining the minimum expiration losses. Existing literature prescribes complex inventory models and algorithms for good inventory management.7  However, a recent survey of hospitals with excellent management of stocks showed that the quality of Blood Bank staff, who must be skilled, regularly trained, and experienced, are key factors in reducing the number of product expirations.8  Also, simple management processes and electronic crossmatch tools provide dependable resources for inventory management.9 

At CCAD, a simple in-house algorithm was created to track and monitor products each quarter until the hospital became fully operational. The average weekly consumption figure for the quarter was divided into 5 business days and is the amount used each day. To include a margin of safety, we multiplied the daily figure by 4 to obtain the number of O-positive units that constitute the security inventory (Table 4). The other groups are calculated according to the ABO percentage of the population, further described in the literature10  (Table 5).

Table 4

Calculation of O-Positive Stock Based on Average Weekly Consumption Units in the Second Quarter of 2016

Calculation of O-Positive Stock Based on Average Weekly Consumption Units in the Second Quarter of 2016
Calculation of O-Positive Stock Based on Average Weekly Consumption Units in the Second Quarter of 2016
Table 5

Different Level of ABO/Rh Stock Based on Calculations of Average Weekly Consumption

Different Level of ABO/Rh Stock Based on Calculations of Average Weekly Consumption
Different Level of ABO/Rh Stock Based on Calculations of Average Weekly Consumption

Following this methodology 6 months post opening, data analysis showed the expiration of red blood cell (RBC) index was in accordance to the most demanding benchmark (2015: 7%; 2016: 0.6%; 2017: 0.2%).

  • 4.

    There was no clear understanding of the use of the Maximum Surgical Blood Order Schedule (MSBOS) program. Transfusion practice varies from institution to institution.11  In coronary surgery, for example, RBC, plasma, and platelet transfusion rates are highly variable, but this variability has also been shown to occur in non-cardiovascular surgeries.12  MSBOS programs are based on establishing the number of RBCs preassigned to each type of intervention, based on hospital usage data or benchmarking with other institutions. This prevents blocking units based on individual surgeon variations for the same procedure. It helps to maintain inventory, reduces expirations, and avoids unnecessary disposal of units (which are sent in a transporter cooler to operating rooms).

The reason for blocking an excessive amount of blood for a patient and sending it in a transport cooler to an operating room is to have the blood on hand in case of an emergency. This practice is known as stockpiling and compromises the availability of units for other patients unless the inventory is expanded (which eventually leads to high expiration rate). On the other hand, reentry of unused and returned units is highly sensitive to temperature changes upon receipt in the Blood Bank and as a result, is a frequent cause of unit wastage.

A classic way to measure unit blockage is by using the unit crossmatched to transfused (C:T) ratio or a more modern version of electronic crossmatch, the issue to transfusion (I:T) ratio. An index greater than 2 to 2.5 indicates stockpiling and should be monitored. Three College of American Pathologists Q-Probes studies of 12,288,404 red cell units in 1639 hospitals revealed that the C:T ratio varied from 1.2 to 2.5 (median, 1.9) and the wastage ratio from 3.0% to 0%. Both parameters were influenced by number of beds, existence of teaching program, use of MSBOS program, or on-site full-time medical director of transfusion services.13  In Table 6 are shown the figures of C:T ratio and percentage wastage at our institution.

Table 6

Cleveland Clinic Abu Dhabi Crossmatched to Transfused (C:T) Ratios and Rates of Units Wastagea

Cleveland Clinic Abu Dhabi Crossmatched to Transfused (C:T) Ratios and Rates of Units Wastagea
Cleveland Clinic Abu Dhabi Crossmatched to Transfused (C:T) Ratios and Rates of Units Wastagea

While ratios were monitored, the following actions were also conducted to achieve a benchmark: creating a crystal report of discarded units and causes; communication plan with wards and operating rooms; training reinforcement on the criteria of return of units and on visual inspection; reorganization of the processing areas inside the laboratory to prevent breakages; communication with the supplier to reduce the breakage of frozen plasma at source and evidenced in the process of thawing; and finally, installation of peripheral refrigerators in the operating room. The implementation of the remote conservation system (BloodTrack, Haemonetics Corporation, Braintree, Massachusetts) helped reduce “out of time” units that had been moved to operating rooms in a transporter cooler.

Safety in transfusion is highly reliant on the expertise of different stakeholders who enable the process. While IT helps to facilitate many of the processes,14  in our opinion, the integration of different IT systems is essential to ensure information is readily accessible to all caregivers involved in the transfusion continuum.

At CCAD, Epic System (Verona, Wisconsin) electronic health record (EHR) and Sunquest Information System (Tucson, Arizona) independently controlled steps within the transfusion process, but neither offered comprehensive control over the entire transfusion process. Under these circumstances, we sought to improve patient safety by introducing 2 new software solutions into our environment and bridging these systems to provide comprehensive transfusion process control. Doing this assured the highest levels of safety for patients receiving blood. This untraditional approach was innovative in our environment and this level of integration does not exist in many hospitals around the globe.15  Below we describe the benefits realized in the different steps of “The Wheel of Transfusion Safety” (Figure 4).

  1. Avoid unnecessary transfusion by justification of blood transfusion. Blood needs to be prescribed according to CCAD guidelines based on the Rational Use of Blood.16  The 2 following elements enable this decision-making: blood needs to be prescribed as based on recent laboratory results, and blood orders need to be justified according to CCAD guidelines. The user gets alerted when an order is placed without a complete blood count or coagulation test resulted in last 24 hours and when it is not justified according to our internal guidelines. Therefore, the clinical justification for blood needs to match the exact test result and patient condition, so a Patient Blood Management Alert will flag when hemoglobin concentration is above the established transfusion guidelines. For example, if the justification is “Intensive care unit (ICU) patient with Hg less than 70 g/L,” but real hemoglobin concentration is 85 g/L, a Patient Blood Management Alert will flag (Figure 5). The alert can be overridden if the patient's clinical diagnosis warrants the transfusion irrespective of the antecedent hemoglobin concentration. Monthly reports monitor alert activity and reason for override.

  2. Identification and sampling. Once the order has been placed in our EHR program, the test and specimen details flow to a Collection Manager system (Sunquest Information System). A positive identification of the patient is done against the patient's wristband with this system. Wristbands are then scanned and labels printed for the tubes at bedside. The labels reflect identifiers of patient, date and time of collection, identification of person who collected the specimen and unique accession number for the specimen. The benefit of having a fully integrated system is to have transparency and monitor specimen status in real time.

  3. Blood compatibility test. ABO type, screening for harmful antibodies, and compatibility test results are automatically transferred from analyzers to LIS and integrated in the patient file. Any discrepancy with the previous patient record will be flagged as an alert for action. Once the compatible units have been allocated to a defined patient, the information is transferred from Sunquest to Epic. Using Epic, nurses can verify the patient records for the identifiers of every unit allocated to this particular patient (donor unit number, type of component, ABO type of unit, and expiration date), and also the unit status will be shown as “OK to transfuse.” Having nursing involved as a partner in this electronic workflow directly enhances the overall care, since it drives accountability of all stakeholders and removes the risk of human communication error.

  4. Issue of blood to peripheral fridges. BloodTrack program permits the electronic allocation of units in remote fridges located in surgical areas. Information flows from Sunquest to BloodTrack to control the location of the unit in real time and to follow the movement of the unit; for example, if the unit has been removed from the fridge, when it was removed, and by whom. This IT interface allows TMS to allocate units directly in the point-of-care site, a significant feature in a large hospital like CCAD.

  5. Administration of blood. This is the last step in the transfusion wheel. After positive identification of the patient and matching with the compatibility label of the unit, the nurse will proceed with IT verification. Using Epic workflow for administration of blood, nurses scan the identifiers of every unit allocated to this particular patient. Any mismatch in the scanning of the following will alert the caregiver with a hard stop: ABO type of unit; donor identification number; product code; and expiration date. Arriving at this stage of efficiency has truly been the result of a dedicated team of multidisciplinary members. Today and as a result of the lessons learned on this TMS journey, the team at CCAD can assure the right blood units are being given to the correct patient (Figure 6).

Figure 4

The Wheel for Safety in Transfusion: different steps need to be interfaced to assure the right blood gets to the right patient. Abbreviation: IT, information technology.

Figure 4

The Wheel for Safety in Transfusion: different steps need to be interfaced to assure the right blood gets to the right patient. Abbreviation: IT, information technology.

Close modal
Figure 5

Patient Blood Management Alert. CCAD, Cleveland Clinic Abu Dhabi; HGB, hemoglobin; RBC, red blood cells.

Figure 5

Patient Blood Management Alert. CCAD, Cleveland Clinic Abu Dhabi; HGB, hemoglobin; RBC, red blood cells.

Close modal
Figure 6

Workflow for administration of blood.

Figure 6

Workflow for administration of blood.

Close modal

While CCAD may be considered an unparalleled extension of Cleveland Clinic model of care, it is not an exact replica of the main campus. We adapted Cleveland Clinic first-class methodology in a new environment with BBTs from diverse cultural backgrounds. This had implications for how caregivers varied in their perception of authority, in following directions, and in how social interaction and relationships evolved in the spirit of safe patient care. The CCAD team and patients benefit from this diversity, as it results in enrichment of work ethic, knowledge sharing, and overall ideas and perspectives of how to best improve clinical care.

The LIS posed another challenge. The LIS is an essential pillar in the Blood Bank and the reliability of the system is essential. The challenge faced was that only 2 of the 14 BBTs were familiar with the IT system implemented. Looking back, training and education on LIS was an opportunity that could have been embraced more. Similar to this was EHR training for the clinicians and nurses. We had to reinforce the end-user training with educational support from clinical educators to assure staff comprehension met CCAD standards for accessing transfusion services.

Another significant challenge to patient safety is the integrity of patient demographic information. Considering the UAE's Bedouin-rich history, date of birth is not widely accessible or accurate in many cases, as the births in older generations happened at home and dates were not well recorded. Additionally, the conversion of dates on a Gregorian calendar from an Islamic calendar was not always accurate. Finally, multiple documents for the same person would have different dates of birth or different spellings of their name in English; for example, for Fatima Mohammad Al Ameri, the middle name could have been spelled as Mohamad (single m), which would be considered a change in the demographics and potentially, the wrong patient. Ultimately, these unique factors have implications for patient safety and quality of care.

We had to build a system of practice and formally document it in advance of the hospital opening. This created the risk that although many of the documents would satisfy regulatory agencies during commissioning, they did not accurately reflect our practice during hospital activation. Furthermore, we did not have an effective electronic document management system. The resulting chaos from multiple versions of the documents circulating frustrated the staff in their daily practice and posed risks to patient safety. The Lean Six Sigma project, explained above, developed from this realization and it took a considerable amount of effort to harmonize the documents and align our practice. The acquisition of an electronic document control system (PolicyTech, Navex, Lake Oswego, Oregon) allowed us to create, track, and manage the documents along with staff acknowledgement after reading.

Opening a new hospital is certainly a once in a lifetime experience and can be very inspiring for those involved in its activation. Our experience in establishing a safe transfusion practice came with its own challenges and opportunities, which we hope others can learn from and apply to future projects.

1
Comarow
A
,
Harder
B.
2017-18 best hospitals honor roll and overview
.
US News & World Report
.
August 8, 2017
. ,
2018
.
2
Erin
M.
Navigating the cultural minefield
.
Harvard Business Review
.
2014
;
92
:
119
124
.
3
National Standard for Blood and Blood Products Safety and Quality
.
National Blood Authority. Australia Government
. ,
2018
.
4
Directive 2002/98/EC of the European Parliament and the Council of 27 January 2003
.
Standards of quality and safety for the collection, testing, processing, storage and distribution of human blood and blood components, and amending Directive 2001/83/EC
.
OJ L 33
;
February
8,
2003
.
5
Standards for Blood Banks and Transfusion Services
.
30th ed
.
Bethesda, MD
:
American Association of Blood Banks;
2016
.
6
Health Policy and Regulation HAAD (Version 1.0)
.
HAAD Clinical Laboratory Standards
.
March
8,
2018
.
7
Brodheim
E
,
Derman
C
,
Prastacos
G.
On the evaluation of a class of inventory policies for perishable products such as blood
.
Manag Sci
.
1975
;
21
(
11
):
1320
1325
.
8
Stanger
S
,
Yates
N
,
Wilding
R
,
Cotton
S.
Blood inventory management: hospital best practice
.
Transfus Med Rev
.
2012
;
26
(
2
):
153
163
.
9
Perera
G
,
Hyam
C
,
Taylor
C
,
Chapman
JF
.
Hospital blood inventory practice: the factors affecting stock level and wastage
.
Transfus Med
.
2009
;
19
(
2
):
99
104
.
10
Issit
PD
.
Applied Blood Group Serology. 3rd ed
.
Miami, FL
:
Montgomery Scientific Publications;
1985
.
11
Bennett-Guerrero
E
,
Zhao
Y
,
O'Brien
S
, et al.
Variations in use of blood transfusion in coronary artery bypass graft surgery
.
JAMA
.
2010
;
304
(
14
):
1568
1575
.
12
Quian
F
,
Osler
T
,
Eaton
M
, et al.
Variation of blood transfusion in patients undergoing major noncardiac surgery
.
Ann Surg
.
2013
;
257
(
2
):
266
278
.
13
Novis
A
,
Renner
S
,
Friedberg
R
, et al.
Quality indicators of blood utilization
.
Arch Pathol Lab Med
.
2002
;
126
(
2
):
150
156
.
14
Hibbs
S
,
Nielsen
N
,
Brunskil
S
, et al.
The impact of electronic decision support on transfusion practice: a systematic review
.
Transfus Med Rev
.
2015
;
29
(
1
):
14
23
.
15
Yazer
M
,
van der Watering
L
,
Lozano
M
, et al.
Development of RBC transfusion indications and the collection of patient-specific pre-transfusion information; summary
.
Vox Sang
.
2017
;
112
(
5
):
487
494
.
16
Carson
JL
,
Guyat
G
,
Heddle
NM
.
Clinical practice guidelines from the AABB: red cell transfusion thresholds and storage
.
JAMA
.
2016
;
316
(
19
):
2025
2035
.

Author notes

The authors have no relevant financial interest in the products or companies described in this article.