Errors in Laboratory Medicine
Pierangelo Bonini,
1,2*
Mario Plebani,
3
Ferruccio Ceriotti,
2
and Francesca Rubboli
2
Background: The problem of medical errors has re-
cently received a great deal of attention, which will
probably increase. In this minireview, we focus on this
issue in the fields of laboratory medicine and blood
transfusion.
Methods: We conducted several MEDLINE queries and
searched the literature by hand. Searches were limited
to the last 8 years to identify results that were not biased
by obsolete technology. In addition, data on the fre-
quency and type of preanalytical errors in our institu-
tion were collected.
Results: Our search revealed large heterogeneity in
study designs and quality on this topic as well as
relatively few available data and the lack of a shared
definition of “laboratory error” (also referred to as
“blunder”, “mistake”, “problem”, or “defect”). Despite
these limitations, there was considerable concordance
on the distribution of errors throughout the laboratory
working process: most occurred in the pre- or postana-
lytical phases, whereas a minority (13–32% according to
the studies) occurred in the analytical portion. The
reported frequency of errors was related to how they
were identified: when a careful process analysis was
performed, substantially more errors were discovered
than when studies relied on complaints or report of near
accidents.
Conclusions: The large heterogeneity of literature on
laboratory errors together with the prevalence of evi-
dence that most errors occur in the preanalytical phase
suggest the implementation of a more rigorous method-
ology for error detection and classification and the
adoption of proper technologies for error reduction.
Clinical audits should be used as a tool to detect errors
caused by organizational problems outside the
laboratory.
© 2002 American Association for Clinical Chemistry
Measuring and improving laboratory-related patient out-
comes require methods that relate the total quality of
laboratory information to more effective patient manage-
ment, including diagnosis, treatment of disease, clinical
monitoring, and disease prevention. The improvement in
analytical quality, documented through proficiency test-
ing, should guarantee that the actual performances of
clinical laboratories are suitable for improving a patient’s
health. Furthermore, increased attention to patients’
needs is demonstrated by efforts to improve the quality of
the entire service provided, e.g., reduction of the turn-
around time (TAT). However, improvement of laboratory
performance does not automatically indicate a reduction
in the number of errors, both analytical and organiza-
tional. Even certification or accreditation processes focus
attention more on the general performance of the labora-
tory than on events such as errors that, by their very
nature, are considered exceptional. Moreover, the lack of
a universally accepted definition of error and above all of
“allowable error rate”, reduces the possibility of evaluat-
ing the impact of laboratory error on patient outcomes.
Although there is abundant scientific literature dealing
with increased laboratory quality (mainly analytical), the
literature on errors in laboratory medicine is scarce. One
reason for this, in addition to the insufficient attention
paid to the problem, is the practical difficulty in reporting
and measuring the number of errors.
In fact, there are several limitations in the study
designs of reports dealing with the frequency and types of
mistakes in the clinical laboratory.
The first limitation is that most of the studies focus
simply on analytical errors, which represent only a per-
centage of the errors in the total testing process, which
includes all pre-, intra-, and postanalytical phases. Other
studies are based on methodologies, such as the split-
specimen design, that are insensitive to total testing
process problems that can occur before specimens are
collected and after results are obtained by the analytical
process (1, 2).
1
Universita ` Vita-Salute San Raffaele, Cattedra di Biochimica Clinica, Via
Olgettina 58, 20132 Milan, Italy.
2
Istituto Scientifico H. S. Raffaele, Servizio Integrato di Medicina di
Laboratorio, Via Olgettina 60, 20132 Milan, Italy.
3
Servizio di Medicina di Laboratorio, Azienda Ospedaliera di Padova, Via
Giustiniani 2, 35100 Padova, Italy.
*Author for correspondence. Fax 39-02-2643-2640; e-mail bonini.
pierangelo@hsr.it.
Received July 20, 2001; accepted February 12, 2002.
Clinical Chemistry 48:5
691– 698 (2002)
Minireview
691