Preanalytic Error Tracking in a Laboratory Medicine Department: Results of a 1-Year Experience
Year: 2006
Type: article
Abstract: Remarkable advances in technology, automation, and testing procedures have produced radical changes in laboratory organization, granting major precision and accuracy of test results. Nevertheless, errors occurring within the whole testing process still influence the quality of laboratory performance. There is heterogeneous information on the error rate within the whole laboratory testing process (from 0.1% to 9.3%) (1). Moreover, the frequencies and types of mistakes differ between one facility and another and between one time period and another (2). Process analysis has demonstrated that laboratory errors occur primarily in the preanalytic phase, influencing patient outcomes and costs (1)(2)(3). Compliance with systems of quality management, such as certification and accreditation, requires accurate procedures for identifying the processes that are more susceptible to errors (4). The Laboratory Medicine Department of the University Hospital of Verona is a laboratory service providing stat and routine tests for clinical chemistry, hematology, coagulation, and immunology, serving an area with a population of 270 000 inhabitants. The laboratory serves a hospital with 750 beds and specialized care units. Inpatient phlebotomies are performed by clinical department staff, whereas blood specimens from outpatients are collected on site by laboratory personnel. After implementation of the new ISO accreditation guidelines, since October 2004 the laboratory staff has been trained to identify and systematically record each error encountered within the global testing process, particularly preanalytic mistakes. Venous blood samples for routine clinical chemistry (lithium heparin gel tubes), hematologic (EDTA tubes), coagulation (buffered sodium citrate tubes), electrophoresis, specific protein, and immunology testing (serum gel tubes) were considered unsuitable according to the following accepted criteria (5)(6): inappropriate volume (excess or deficit in the volume required to perform the analysis), wrong or missing patient identification, inappropriate container, visible hemolysis after centrifugation, clotting, and contamination from infusion route. The laboratory staff was provided with a notebook in which numbers of tubes and errors were recorded daily, and the quality referent of our laboratory systematically discussed and reviewed data on a weekly basis. From October 2004 to September 2005, a total of 423 075 routine venous blood specimens (71 922 from outpatients; 17%) were received in the 5 more representative sections of our laboratory (130 806 for clinical chemistry testing, 113 699 for hematologic testing, 61 301 for coagulation testing, 59 403 for electrophoresis and specific protein testing, and 57 866 for immunology testing). According to the above specified criteria, 3154 (0.74%) preanalytic errors were identified and recorded in the 1-year observational period (Table 11 ). Errors were related to samples in which they were relevant. In agreement with earlier data (1)(7), we observed a significant difference in the error rate between inpatients and outpatients (0.82% vs 0.37%; χ2 test, P <0.001). The most common mistakes could be traced to incorrect procedures for sample collection, including hemolysis and clotting. The prevalences of other types of errors were rather different between specimens collected from inpatients and outpatients. In particular, insufficient volume was a prevailing cause of unsuitable specimens for inpatient samples, whereas the prevalence of inappropriate containers was particularly high for outpatient specimens. Current data on laboratory error rates are mixed (1). In a survey on outpatient phlebotomy success, most unsuitable samples resulted from hemolysis (18.1%), insufficient quantity (16.0%), and clotting (13.4%) (5). These data are comparable to those provided by additional investigations, which confirm that problems directly related to specimen collection are the first causes of preanalytic errors, especially hemolyzed, clotted, insufficient, and incorrect samples (1)(7)(8)(9). This investigation represents a logical extension of previous analyses focused on laboratory errors identified after collection of clinically questionable results in stat departments for a shorter period of time (2)(8) or involving limited areas of laboratory testing, such as clinical chemistry (7) and hematology(9). Additionally, we included a systematic approach for error identification, extending the analysis to either inpatient or outpatient specimens for routine laboratory testing. Results were finally related to the total number of specimens received, rather than to the tests ordered. Nevertheless, the substantial difficulty in linking preanalytic quality improvements to patient outcome might be a drawback of this investigation (4)(10). The magnitude of medical errors requires the introduction of comprehensive reporting systems, encompassing mistakes that fall within the whole diagnostic process (11). Although error tracking within laboratory practice appears crucial, a list of preanalytic performance indicators has not been universally defined (4). A suitable approach is to develop a system based on representative preanalytic performance measures and on criteria for specimen acceptability (6). In our experience, implementation of a systematic error-tracking system in daily practice has provided meaningful information on the local preanalytic processes more susceptible to errors, providing an ideal foundation for efficient feedback and enabling evaluation of specific responsibilities. Preanalytic error tracking, based on specimen acceptability, over a 1-year observational period in the 5 most representative sections of a laboratory medicine department.1 Data are shown as absolute and percentage errors for samples in which each preanalytic error was relevant (hemolysis in specimens for clinical chemistry, coagulation, electrophoresis, specific protein, and immunology testing; clotting in specimens for hematologic and coagulation testing). Preanalytic error tracking, based on specimen acceptability, over a 1-year observational period in the 5 most representative sections of a laboratory medicine department.1 Data are shown as absolute and percentage errors for samples in which each preanalytic error was relevant (hemolysis in specimens for clinical chemistry, coagulation, electrophoresis, specific protein, and immunology testing; clotting in specimens for hematologic and coagulation testing).
Source: Clinical Chemistry
Institution: University of Verona
Language: English
Subfield: Medical Laboratory Technology
Field: Health Professions
Domain: Health Sciences
Open Access status: bronze
OpenAlex ID: W2144213167
