American Association for Clinical Chemistry
Better health through laboratory medicine
Patient Safety Focus: Pre-analytic Labeling Errors

Patient Safety Logo

Pre-analytic Labeling Errors
What is the Error Rate?

By Erin Grimm, MD

Arriving at a benchmark for an acceptably low rate of lab pre-analytic errors isn’t simple, and few published studies exist on the subject. The lack of published studies may be due to inadequate detection methods, reluctance to publish error rates, and differing error definitions. Here’s a look at what has been published.

Transfusion Errors

Mistransfusion of blood remains a serious patient safety issue, with the vast majority of mistransfusions resulting from avoidable errors at various points in the transfusion chain. During the 1990s, mistakes in pretransfusion testing accounted for approximately 14% of ABO-incompatible transfusions in New York (1). Such errors often result in death or major morbidity (2).

Mistransfusions also result from transfusion specimen mislabeling and what is called wrong blood in tube. WBITs are specimens where the tube and requisition are internally consistent and contain all necessary identifiers for patient A, but the blood in the tube is actually from patient B. The vast majority of these samples are identified through comparing the current tube’s blood type with a previous blood type stored in the patient record. However, this analysis would not catch the error if the patient does not have a blood type on record or if patients A and B have the same blood type.

The box below lists three studies that attempted to document WBIT rates. The researchers reported the observed WBIT rate, as well as the corrected rate, which accounts for the random chance that two individuals will have the same blood type. The category “mislabeled” includes any sample received in the lab that failed to meet local criteria for sample acceptance. Overall, the WBIT rates are relatively constant among the three institutions and over time, but the rates of mislabeled transfusion specimens vary from 0.7% to 3.2% (7,8).

Comparison of WBIT Error Rates




Study population

Dzik et al. (7)

Observed rate:
Mean 0.7% (5,161/692,505)
Median 0.6% (1/165)

Observed rate:
0.04% (84/190,406)

Corrected rate:
Median 0.05% (1/1986)

Voluntary study composed of self report by 62 institutions from 10 different nations, including Canada, the U.K., France, Japan, and the U.S.

Murphy et al. (8)

Observed rate:
Mean 3.2% (14,114/445,726)
Median 2.5% (1/40)

Observed rate:
45/132 922 (0.03%)

Corrected rate:
Median 0.08% (1/1,303)
Mean 0.07% (1/1,501)

Voluntary study of hospital in Northern England and Wales. Mislabeled rate based on 110 institutions. WBIT based on 85 institutions.

Lumadue et al.(9)

1.4% (1/71)
N = 40,770

0.035% (1/2,800)

Johns Hopkins Hospital

Corrected rate =

Number of WBIT samples



Number of samples with a historical blood type


where Q equals the chance that two random individuals will have the same ABO group.

General Lab Errors

A good source of data on pre-analytical error rates in U.S. clinical labs is the CAP Q-probes and Q-Tracks programs. One Q-Probes study of 120 institutions reported the average error rate as 0.03% (3). This study also found that institutions that do not perform a clerical check of requisitions against verified lab results report a high percentage of lab errors in the pre-verification stage. While finding errors at this stage is generally superior to finding them in the post-verification stage or after the results are released to the patient record, this may be misleading if an institution does not have a designated step that performs a post-verification check. In other words, you find the errors that you look for, but you may miss the ones that you are not looking for (3). In fact, leaders in the field generally believe that routine error-finding methods grossly underestimate actual error rates (4).

Success Stories

Institutions that implement intensive error prevention interventions usually introduce them after an extensive error detection program. This way, the institution can judge the success of their intervention. When UCLA implemented an error reduction program that included transitioning to 24-hour phlebotomy services, implementing an electronic event report system, and installing an automated specimen processing system, they reported their critical specimen mislabeling rate—defined as any laboratory blood specimen that is unlabeled, mislabeled, or has a specimen/requisition mismatch—to be < 0.1% (5).

Another institution that reported implementation of a robust error-reduction intervention is the University of Iowa Medical Center. They implemented barcode technology for patient identification in their transfusion service throughout the 772-bed hospital and decreased their transfusion sample rejection for any reason from an average of 1.82% to 0.17% (6).

Defining Mislabeled

Another consideration when evaluating the spectrum of published specimen mislabeling rates is defining what constitutes a “mislabeled” specimen. The transfusion literature notes that the requirements for adequate labeling vary. Some institutions require one patient identifier (rare among surveyed institutions), while other use two patient identifiers (the most common requirement), which generally consist of the patient’s name and a unique identifier, such as a hospital record number. There are also institutions that require multiple identifiers on either/both the tube and paperwork for acceptance, such as name, hospital number, date of draw, and phlebotomist’s initials (7,8). Once mislabeling rates are established, there is equal variability in the possible fates of the mislabeled sample, including: discarding the specimen without logging in; discarding it but logging it in; testing the sample but holding results; allowing for addition of missing information; and allowing relabeling.

Considering the published data on error rates and the inescapability of human error, it seems likely that mislabeling rates among transfusion specimens and general lab specimens occur at rates of a few per thousand specimens. Rates of less than a few per thousand either use less rigorous error detection methodologies or involve strong interventions, such as 24-hour phlebotomy teams or automation using bar code–based patient identification and specimen collection.


  1. Linden JV, Wagner K, Voytovich AE, Sheehan J. Transfusion errors in New York State: an analysis of 10 years’ experience. Transfusion 2000;40:1207–1213.
  2. Stainsby D, Russell J, Cohen H, Lileyman, J. Reducing adverse events in blood transfusion. British Journal of Haematology 2005;131:8–12.
  3. Valenstein PN, Raab SS, Walsh MK. Identification errors involving clinical laboratories: a College of American Pathologist q-probes study of patient and specimen identification errors at 120 institutions. Arch Pathol Lab Med 2006;130:1106–1113.
  4. Valenstein PN, Sirota RL. Identification errors in pathology and laboratory medicine. Clin Lab Med 2004;24:979–996.
  5. Wagar EA, Tamashiro L, Bushra Y, Lilborne, L, Bruckner DA. Patient safety in the clinical laboratory: a longitudinal analysis of specimen identification errors. Arch Pathol Lab Med 2006;130:1662–1668.
  6. Porcella A, Walker K. Patient safety with blood products administration using wireless and bar-code technology. AMIA Annu Symp Proc 2005:614–618.
  7. Dzik WH, Murphy MF, Andreu G, Heddle N, Hogman C, Kekomaki R, Murphy S, Shimizu M, Smit-Sibinga CT. An international study of the performance of blood sample collection. Vox Sanguinis 2003;85:40–47.
  8. Murphy MF, Stearn BE, Dzik WH. Current performance of patient sample collection in the UK. Transfusion Medicine 2004;14:113–121.
  9. Lumadue JA, Boyd JS, Ness PM. Adherence to a strict specimen labeling policy decreases the incidence of erroneous blood grouping of blood bank specimens. Transfusion 1997;37:1169–1172.

Eric Grimm, MD, is a resident in pathology and laboratory medicine in the University of Washington Department of Laboratory Medicine.

Patient Safety Focus Editorial Board

Michael Astion, MD, PhD
Department of Laboratory Medicine
University of Washington, Seattle

Peggy A. Ahlin, BS, MT(ASCP)
ARUP Laboratories
Salt Lake City, Utah 
James S. Hernandez, MD, MS
Mayo Clinic College of Medicine
Rochester, Minn.
Devery Howerton, PhD
Centers for Disease Control and Prevention
Atlanta, Ga.

Sponsored by ARUP Laboratories, Inc.