Errors in clinical laboratories or errors in laboratory ...

1 Clin Chem Lab Med 2006;44(6):750 759 2006 by Walter de Gruyter Berlin New York. DOI in press - uncorrected proofReviewErrors in clinical laboratories or Errors in laboratorymedicine?Mario Plebani*Department of laboratory Medicine, UniversityHospital of Padova and Center of BiomedicalResearch, Castelfranco Veneto, ItalyAbstractLaboratory testing is a highly complex process and,although laboratory services are relatively safe, theyare not as safe as they could or should be. Clinicallaboratories have long focused their attention onquality control methods and quality assessment pro-grams dealing with analytical aspects of testing.

2 How-ever, a growing body of evidence accumulated inrecent decades demonstrates that quality in clinicallaboratories cannot be assured by merely focusing onpurely analytical aspects. The more recent surveys onerrors in laboratory medicine conclude that in thedelivery of laboratory testing, mistakes occur morefrequently before (pre-analytical) and after (post-ana-lytical) the test has been performed. Most Errors aredue to pre-analytical factors (46 of totalerrors), while a high error rate ( 47% of totalerrors) has also been found in the post-analyticalphase. Errors due to analytical problems have beensignificantly reduced over time, but there is evidencethat, particularly for immunoassays, interference mayhave a serious impact on patients.

3 A description ofthe most frequent and risky pre-, intra- and post-ana-lytical Errors and advice on practical steps for meas-uring and reducing the risk of Errors is therefore givenin the present paper. Many mistakes in the Total Test-ing Process are called laboratory Errors , althoughthese may be due to poor communication, action tak-en by others involved in the testing process ( , phy-sicians, nurses and phlebotomists), or poorly design-ed processes, all of which are beyond the laboratory scontrol. Likewise, there is evidence that laboratoryinformation is only partially utilized.

4 A recent docu-ment from the International Organization for Stan-dardization (ISO) recommends a new, broaderdefinition of the term laboratory error and a clas-sification of Errors according to different criteria. In amodern approach to total quality, centered onpatients needs and satisfaction, the risk of Errors andmistakes in pre- and post-examination steps must beminimized to guarantee the total quality of laboratoryservices.*Corresponding author: Mario Plebani, Department ofLaboratory Medicine, Azienda Ospedaliera-Universita`diPadova, Via Giustiniani 2, 35128 Padova, ItalyPhone:q39-049-8212792, Fax:q39-049-663240,E-mail: intra-analytical; laboratory medi-cine; mistakes; post-analytical procedures; pre-analyt-ical; total testing Institute of Medicine (IOM) report,To Err IsHuman: Building a Safer Health System(1), galva-nized a dramatically expanded level of debate andconcern about patient injuries in healthcare.

5 Patientsafety, a topic that had been little understood, andeven less discussed in healthcare systems, became afrequent focus for journalists, healthcare leaders, andconcerned citizens (2). The IOM report has far-reach-ing implications for all disciplines, including patholo-gy activities and laboratory medicine (3). Laboratoryservices have a great influence on clinical decision-making: 60 70% of the most important decisions onadmission, discharge, and medication are based onlaboratory test results (4). With this high degree ofinfluence, the quality of laboratory testing and report-ing is of utmost importance.

6 The overwhelmingdependence of clinical decision-making and patient-management processes on laboratory reporting (5)must induce laboratory medicine to set higher qualitystandards. Unlike many other medical processes,activities in laboratory medicine are precisely definedand are therefore more controllable than a procedureor treatment in an emergency department or othermedical settings. laboratory medicine enjoys anotherunique advantage in that it pioneered statistical qual-ity control (QC) activities and it is leagues ahead ofother clinical disciplines in introducing qualityimprovement initiatives.

7 However, the real number ofmistakes made in laboratory testing is not fully rec-ognized, because no widespread process is in placeto either determine how often mistakes occur or tosystematically eliminate sources of Errors . Moreover,total testing is complex, consisting of a series of inter-related processes, each involving a series of processsteps, every one of which can result in an error. Lab-oratory activities have been traditionally classified aspre-, intra- and post-analytical. In the past, laboratoryprofessionals focused their attention on intra-analyti-cal Errors and on mistakes resulting in adverseevents, but overlooked the near misses that, appar-ently, cause no harm.

8 Other well-recognized limita-tions of our knowledge regarding Errors and mistakesin laboratory medicine are derived from the scarcescientific literature dealing with this topic, from prac-tical difficulties in study design, as well as in meas-Brought to you by | Taipei Medical University (Taipei Medical University)Authenticated | Date | 5/29/12 4:21 PMPlebani: Errors in clinical laboratories or Errors in laboratory medicine? 751 Article in press - uncorrected proofTable 1 Frequency of Errors in clinical laboratorieswmodified from reference (3) error identified every:a) 330 1000 eventsb) 900 2074 patientsc) 214 8316 laboratory resultsuring and reporting the number of Errors and, aboveall, in defining what a laboratory error actually the perspectiveIn the past, the perspective under which most studieson Errors in laboratory medicine were performed waslimited to what happened inside the clinical labora-tory.

9 Thus, only analytical and some pre-analyticalerrors (those affecting the so-called within labora-tory pre-analytical steps ) were detected. Thisapproach, however, is not consistent with the conceptof patient-centered care, which is one of the six spe-cific aims for improvement of the healthcare systemsuggested in the IOM reportCrossing the QualityChasm(6). The promotion of patient-centered careshould be translated into the need to investigate anypossible defect that occurs in the total testing processand that can eventually have a negative impact on thepatient. From the patient s viewpoint, in fact, anydirect or indirect negative consequence related to alaboratory test must be considered, irrespective ofwhether the source lies in the pre-, intra- or post-ana-lytical step.

10 Moreover, from the patient s viewpoint itis irrelevant whether any error is caused by a labor-atory professional ( , calibration or testing error) orby a non- laboratory operator ( , inappropriate testrequest, error in patient identification, etc). Therefore,the unique framework for considering where mistakescan occur in laboratory testing services is the totaltesting process. A mistake can occur in each of the 11steps in this process or in any of the places where ahandoff can occur, starting from test request and end-ing with the physician s reaction to laboratory infor-mation.