Original article 105
Evaluating laboratory key performance using quality indicators in Alexandria University Hospital Clinical Chemistry Laboratories Mostafa M. Rizka, Adel Zakib, Nermine Hossama and Yasmin Aboul-Elaa a Clinical Pathology Department, Faculty of Medicine and bBiostatistics Department, Medical Research Institute, Alexandria University, Egypt
Correspondence to Nermine Hossam, MD, Clinical Pathology Department, Faculty of Medicine, Alexandria University, Alexandria, 21311 Egypt Tel: + 20 122 337 4277; fax: + 002035411281; e-mail: [email protected]
Received 30 November 2013 Accepted 20 July 2014 Journal of the Egyptian Public Health Association 2014, 89:105–113
Background The performance of clinical laboratories plays a fundamental role in the quality and effectiveness of healthcare. Objectives To evaluate the laboratory performance in Alexandria University Hospital Clinical Laboratories using key quality indicators and to compare the performance before and after an improvement plan based on ISO 15189 standards. Materials and methods The study was carried out on inpatient samples for a period of 7 months that was divided into three phases: phase I included data collection for evaluation of the existing process before improvement (March–May 2012); an intermediate phase, which included corrective, preventive action, quality initiative and steps for improvement (June 2012); and phase II, which included data collection for evaluation of the process after improvement (July 2012–September 2012). Results In terms of the preanalytical indicators, incomplete request forms in phase I showed that the total number of received requests were 31 944, with a percentage of defected request of 33.66%; whereas in phase II, there was a significant reduction in all defected request items (Po0.001) with a percentage of defected requests of 9.64%. As for the analytical indicators, the proficiency testing accuracy score in phase I showed poor performance of 10 analytes in which total error (TE) exceeded total error allowable (TEa), with a corresponding sigma value of less than 3, which indicates test problems and an unreliable method. The remaining analytes showed an acceptable performance in which TE did not exceed the TEa, with a sigma value of more than 6. Following an intervention of 3 months, the performance showed marked improvement. Error tracking in phase I showed a TE of (5.11%), whereas in phase II it was reduced to 2.48% (Po0.001). For the postanalytical indicators, our results in phase I showed that the percentage of nonreported critical results was 26.07%. In phase II, there was a significant improvement (Po0.001). The percentage of nonreported results was 11.37%, the reasons were either inability to contact the authorized doctor (8.24%), wrong patient identification (1.0%), lack of reporting by lab doctor (1.11%), and finally, lack of reporting by the lab technician (1.03%). Conclusion and recommendations Standardization and monitoring of each step in the total testing process is very important and is associated with the most efficient and well-organized laboratories. Keywords: external quality assessment, key performance indicators, quality indicators, turnaround time, total testing process J Egypt Public Health Assoc 89:105–113 & 2014 Egyptian Public Health Association 0013-2446
Introduction Quality indicator (QI) is a quality tool enabling laboratories seeking improvement to quantify the laboratory’s performance by selecting a certain comparative criterion; its aim is to assess the performance to initiate corrective measures to ensure continual improvement [1]. When identifying the QIs, the acronym SMART is often applied. SMART denotes selected goals that should be specific, measurable, achievable, realistic, and timely [2]. 0013-2446 & 2014 Egyptian Public Health Association
Assessment of the quality of laboratory services using QIs requires a systematic, transparent, and consistent approach for collection and analysis of data. QIs were classified as preanalytical, analytical, and postanalytical according to the phase of the laboratory process they measure. Classically, the preanalytical phase includes all processes from the time a laboratory request is made by a physician until the sample is ready for testing; this phase is complex DOI: 10.1097/01.EPX.0000453262.85383.70
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
106
Journal of the Egyptian Public Health Association
and involves many processes, most of them external to the laboratory, and is prone to the maximum variation and the highest proportion of errors [3]. ISO 15189: 2007 standard specifies the information that must be provided on the request form, irrespective of whether the test requisition is made on paper or electronically as follows: unique identification of the patient; name or other unique identifier of the physician or other legally authorized individual; and type of primary sample, tests requested, and date and time of primary sample collection and relevant clinical information about the patient necessary for interpretation purposes; at a minimum, this should include sex and date of birth [4].
tion must understand and address the variables involved in the process. Documentation and communication processes must be regularly monitored and implemented under ongoing systems for quality monitoring [11].
Acceptance of improper specimens for analysis (insufficient sample volume, inappropriate collection container, hemolyzed, lipemic, icteric, or clotted specimens, inadequate ratio volume sample/anticoagulant, and unidentified sample) may lead to erroneous results that could affect patient care, but only by monitoring the rejected specimens on a regular basis and identifying factors associated with the rejection; can the lab avoid errors and promote continuous quality improvement of the laboratory service offered [5].
Accuracy, precision, timeliness, and authenticity are the four pillars of efficient laboratory services. However, timeliness, which is expressed as the turnaround time (TAT), is often used by the clinicians as the benchmark for laboratory performance; laboratorians may disagree with such a priority, arguing that unless analytical quality can be achieved, none of the other characteristics matter [15]. Delays in TAT elicit immediate complaints from users while adequate TAT goes unremarked. Unsatisfactory TAT is a major source of complaints to the laboratory in terms of poor service and requires much time and effort from laboratory staff in complaint resolution and service improvement [16].
Quality management of the analytical phase of the testing process is the most standardized and regulated part and has received the most attention [6]. External Quality Assessment programs are designed to provide a regular, objective, and independent assessment of a participant’s ability to provide an acceptable standard of service by comparison with peers [7]. Each laboratory is provided with a numerical indicator of its competence, a performance index or score, together with information on the performance of the group as a whole, enabling its proficiency relative to the group to be compared. The participating laboratories use the information on their performance to make appropriate changes and improvements [8]. The total testing process (TTP) consists of a series of interrelated processes, each involving a series of process steps, every one of which can result in an error. TTP consists of series of activities, starting with the clinical question in the clinician’s mind, leading to test selection, sample collection, transport to the laboratory, analysis, reporting back to the clinician, and final interpretation and decision making by the clinician. Identifying and learning from errors is stated to be a key factor for improving the quality and safety in healthcare; to guarantee excellent performance and service, the process of identifying and treating error risks must be integrated into the TTP. A suitable system for grading laboratory errors on the basis of their seriousness should help identify priorities for quality improvement and enable a focus on corrective/preventive actions [9]. Improving quality in the postanalytical phase represents as such a valuable goal for reducing errors and improving patient safety. A key issue in postanalytical quality is represented by the effectiveness of laboratory data communication, particularly communication of critical test results [10]. For the critical value reporting process to be effective, the organiza-
The content, format, and physical presentation of the information significantly affect the interpretation and use of laboratory data by clinicians [12,13]. Mistakes in the content and completeness of laboratory reports as well as misunderstanding by the treating physician as to the significance of the information in the report, among other factors, can delay the treatment of a serious disease and alter outcomes [14].
Reliability cannot be achieved in a clinical laboratory through the control of accuracy in the analytical phase of the testing process alone, but is also determined by preanalytical and postanalytical factors [17]. Thus, the aim of this study was to promote accuracy in the Alexandria University Hospital Laboratory (AUHL) including the analytical phase as well as assurance of reliability of preanalytical and postanalytical activities.
Materials and methods Phase I
The laboratory is an ISO 15189–2007-certified laboratory affiliated to a public university hospital serving a population of 234 403. It provides care for inpatients, outpatients, and emergency cases. Samples were collected by nurses, transported to the laboratory sample reception desk for classification, and distributed to the appropriate units where testing was conducted. In 2012, 262 100 requests and 2 668 984 tests were performed. A study was carried out on all inpatient tests presented to the Clinical Chemistry laboratory for a period of 7 months that was divided into three phases: phase I included data collection for evaluation of the existing process (March–May 2012), the intermediate phase included corrective, preventive action, quality initiative, and steps for improvement (June 2012), and phase II included data collection for evaluation of the process after improvement (July–September 2012). QIs were classified as preanalytical, analytical, and postanalytical according to the laboratory process that they measured. Preanalytical QIs included incomplete request forms, and rejected samples. Analytical QIs
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 107
included accuracy of proficiency testing. Postanalytical QIs included critical result report and laboratory TAT [18]. An incomplete request form QI was used to reveal any omission of demographic items (age, requesting doctor, suspected diagnosis, etc.) per total requests. The next QI referred to unsuitable specimens. This indicator was monitored by defining a preanalytical sample error as a rejected specimen per total number of received samples: any sample that was not suitable for one or more tests in the total order because the specimen did not fulfill the acceptability criteria, or the sample was not received [19]. The analytical QI included accuracy of proficiency testing, calculated from the external quality program report. This report was based on Six-Sigma results and reported every 3 months to each functional unit. X ðTEa biasÞ Sigma metrics : ¼ ; SD of the analytes measured where TEa is the total allowable error. TEa values of various parameters were taken from the Clinical Laboratories Improvement Act guidelines [20]. Bias was computed from the external quality assurance records using the following formula: bias = mean of all laboratories using the same instrument and method – our results. SD was determined from the calculated laboratory mean and calculated SD obtained from the internal QC data. Critical result reporting: for immediate notification of a physician (or other clinical personnel responsible for patient care) when examination results for critical properties fall within established ‘alert’ or ‘critical’ intervals, a hard copy of the report would be sent to the appropriate healthcare provider as soon as possible. It was calculated as number of nonreported critical results/total number of critical results. Laboratory TAT was measured from the time of test registration to the time of results verification; data were obtained from lab information system (LIS). All the key performance indicators were expressed as percentages, except for TAT (expressed in days) and QPIs (as average score). Intervention
The major link in the laboratory service and the hospital is through the nursing staff, which act as phlebotomists according to the AUH policy. Multiple methods were used to determine the educational needs and training for the nurses as well as the phlebotomist; these include: evaluating knowledge and skills in job description, asking nurses supervisors, testing nurses, and phlebotomists, and analyzing past performance appraisal. Educational lectures and video films were presented to nurses and technicians together with workshops with a special focus on quality awareness, quality participation, ISO 15189 standards, AUHL mission and vision. Sessions were divided for the preanalytical team to include phlebotomy skills and preanalytical standards and errors.
Analytical team sessions included quality concepts, westguard rules, inventory management, and proficiency testing evaluation. The for postanalytical team including medical secretaries and nurses as well as technicians, received a special session on the critical results reporting and importance of diagnosis registration and its benefits for clinical correlation, learning the standardized abbreviation for common diseases, and development of a data base for all resident doctors including name, ward, and mobile number to facilitate communication with the responsible physician, which lead to effective critical value reporting and diminished failure of communication between the lab and the clinical wards in the hospital. For laboratory TAT, the steps of improvement adopted were as follows: (1) Development of a time table for sample receiving from wards to avoid overcrowding in the accession area and delaying of sample registration and transportation. (2) Increase the number of computers used in the accession area of the lab with increasing number of personnel for data entry. (3) Implementation of a regular schedule for transportation of the sample from the accession area to the central analytical lab associated with a work list. (4) Availability of the resident doctor in the accession area. (5) Formal instruction to all wards for minimal sample requirement to avoid problems of short samples. (6) Delegation of passage of the sample on a barcode reader in the central lab to a certain technician to decrease TAT. (7) Regular check for incomplete results and results above linearity for further dilution.
Statistical analysis
It was carried out using the PSWA Statistics software (version 18; SPSS, Chicago, Illinois, USA). Categorical variables were described using frequencies and percentages. The w2-test was used to test associations. A statistical significance level of 95% (Po0.05) was considered for all statistical analyses. All tests used in this study were two-sided.
Results Preanalytical phase
In terms of the incomplete request form and rejected samples, results showed that there was a significant statistical improvement between phase I and phase II (Po0.05) (Tables 1 and 2). For the incomplete request form, phase I showed that the total number of received requests was 31 944, absent admission number was missing in 590 (1.85%), absent requested doctor identification in 2028 (6.35%), and absent diagnosis in 558 (1.75%). In phase II, the total
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
108
Journal of the Egyptian Public Health Association
Table 1. Frequency of absent information that must be provided on the request form presented as a percentage of total received requests before (phase I) and after (phase II) improvement Number of defects (% of received request) Defected item Incomplete patient information Absent admission number Absent patient preparation Absent physician ID Absent diagnosis Absent date and time Absent type of sample
Phase I (N = 31 944) 326 590 949 2028 558 3126 3176
Phase II (N = 28 286)
(1.02) (1.85) (2.97) (6.35) (1.75) (9.79) (9.94)
68 80 475 965 184 503 452
(0.24) (0.28) (1.68) (3.41) (0.65) (1.78) (1.60)
P value 0.001 0.001 0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 2. Frequency of different causes of sample rejection presented as a percentage of total received sample before (phase I) and after (phase II) improvement Received sample [n (%)] Cause of rejection Hemolysis Clotted Sample/anticoagulant ratio Miss-identification Lipemic Total
Phase I (N = 50 440) 1584 494 179 41 16 2314
Phase II (N = 45 180)
(3.14) (0.98) (0.35) (0.08) (0.03) (4.59)
841 311 95 32 6 1285
(1.86) (0.69) (0.21) (0.07) (0.01) (2.84)
P value 0.001 0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
number of received requests was 28 286, admission number was missed in 80 (0.28%), and absent doctor identification in 965 (3.41%). For the rejected sample, in phase I, the total number of received samples was 50 440, with a total number of rejected samples of 2314 (4.59%), whereas in phase II, the total number of received samples was 45 180, and the total number of rejected samples was 1285 (2.84%). Analytical phase
The proficiency testing accuracy score measured by the sigma metrics in phase I showed sigma value more than 6 for triglycerides and creatine kinase (CK). Sigma value in the range 3–5.9 was observed for three parameters, namely, albumin, potassium (K), and total protein. A sigma value of less than 3 was observed for total bilirubin, calcium, chloride (Cl), cholesterol, creatinine, glucose, sodium, urea, uric acid (UA), alanine aminotransferase (ALT), alkaline phosphatase (ALP), aspartate aminotransferase (AST), and lactate dehydrogenase. In phase II, sigma value of more than 6 was observed for UA and CK and a value in the range 3–5.9 was observed for 13 parameters, namely, albumin, calcium, cholesterol, creatinine, glucose, K, total protein, Na, triglyceride, urea, ALT, ALP, AST, and lactate dehydrogenase. A sigma value of less than 3 was observed for total bilirubin and Cl (Table 3). Postanalytical phase
Study of critical results reporting and laboratory TAT monitoring showed a significant improvement between phase I and phase II. As for critical results reporting, in phase I, the total number of critical results to be communicated was 3180, the total number of nonreported
results was 829 (26.07%), whereas in phase II, the total number of critical results to be communicated was 2805 and the total number of nonreported results was 319 (11.37%) (with Po0.05) (Tables 4–6). For laboratory TAT, in phase I, the distribution of TAT was on average of 172.7 min as shown in (Tables 7 and 8). For specimens reported within 120 min, the average TAT was 95.58 min (30.44%). For specimens reported between 120 and 180 min, the average TATwas 148 min (39.21%), and for specimens reported more than 180, the average TAT was 273.8 min (30.35%). In phase II, the distribution of TAT was an average of 151.29 min. For specimens reported within 120 min, the average TAT was 93.11 min (72.44%), for specimens reported between 120 and 180 min, the average TAT was 139.4 min (24.78%), and for specimens reported more than 180, the average TAT was 221.36 min (2.78%).
Discussion The major aim of the quality system is to reduce or, ideally, eliminate all defects within the entire testing process. Lippi et al. [13] emphasized the development of a strategy to enhance quality throughout the TTP. This strategy starts with a systematic analysis of specimen workflow, continues with education and thorough monitoring by error tracking systems, and ends with the elimination or redesign of deficient procedures. To our knowledge, this is the first study to assess the performance of the laboratory services in a University Hospital in Alexandria. A similar study carried out by Chhillar et al. [21] reported that the major preanalytical errors of concern noticed were that the treating physician’s names were missing in 13.1% of the forms
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 109
Table 3. Total errors allowable, average bias, SD, and sigma values for the different parameters before (phase I) and after (phase II) improvement
ALP, alkaline phosphatase; ALT, alanine aminotransferase; CK, creatine kinase; LDH, lactate dehydrogenase; TEa, total errors allowable; UA, uric acid.
and their signatures were missing in almost 13.4% of the lab forms. Clinical information was illegible or difficult to decipher in 89.25% of the forms. The diagnosis was not reported in 61.2% of the lab request forms and the information pertaining on the type of specimen was missing in 61.6% of the forms [21].
As for rejected samples, phase II showed a significant reduction (Po0.01). This is in agreement with Lippii et al. [18], who reported that the lack of a standardized protocol for sample collection, including patient preparation, specimen acquisition, handling, and storage, accounts for up to 93% of the errors currently encountered
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
110 Journal of the Egyptian Public Health Association
Table 4. Frequency of different type of errors before (phase I) and after (phase II) improvement Number of error (% of total received) Type of error
Phase I (N = 27 612)
Missed result Missed sample (order) Missed sample (received) Incomplete results Total
296 199 160 755 1410
Phase II (N = 24 507)
(1.07) (0.72) (0.58) (2.73) (5.11)
141 85 57 325 608
P value
(0.58) (0.35) (0.23) (1.33) (2.48)
0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 5. Frequency of different causes of inaccurate report per total number of released reports presented before (phase I) and after (phase II) improvement Total released [n (%)] Defected item
Phase I (N = 27 612)
Patient information Patient code Location Comment Signature of the authorized person Total
420 630 210 320 50 1630
Phase II (N = 24 507)
(1.52) (2.28) (0.76) (1.16) (0.18) (5.90)
80 95 110 125 35 445
P value
(0.33) (0.39) (0.45) (0.51) (0.14) (1.82)
0.001 0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 6. Frequency of causes of nonreported critical results before (phase I) and after (phase II) improvement Number of defects (% of total received) Causes
Phase I (N = 3180)
Wrong patient identification Lack of reporting by lab technician Lack of reporting by lab physician Inability to contact the authorized physician Total
109 80 94 546 829
Phase II (N = 2805)
(3.43) (2.52) (2.96) (17.17) (26.07)
28 29 31 231 319
P value
(1.00) (1.03) (1.11) (8.24) (11.37)
0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 7. The mean and proportions of overall turnaround time before (phase I) and after (phase II) improvement Phase I Group
Phase II
Received sample [n (%)] Mean TAT (min) Received sample [n (%)] Mean TAT (min) P value
Specimens reported within 120 min Specimens reported between 120 and 180 min Specimen reported after 180 min Total
673 867 671 2211
(30.44) (39.21) (30.35) (100.00)
95.58 148.76 273.80 172.71
1590 544 61 2195
(72.44) (24.78) (2.78) (100.00)
93.11 139.4 221.36 151.29
0.001 0.001 0.001 0.001
Statistically significant at Pr0.05. TAT, turnaround time.
Table 8. The mean turnaround time of each phase before (phase I) and after (phase II) improvement Phase I
Phases Mean TAT (min)
Phase II
Registration– collection
Collection– receive
Receive– complete
Registration– collection
Collection– receive
Receive– complete
P value
15.28
16.61
140.82
11.36
12.25
127.68
0.001
Statistically significant at Pr0.05. TAT, turnaround time.
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 111
Figure 1.
Manpower
Method
Porter busy or not available
Receiving all wards in the same time
Inadequate number of technician
Samples sent to central lab without work list Problem in sample receiving in central lab
Absence of resident doctor in accession Training required on computer system
Prolonged turnaround time Absence automatic dilution of result exceeding linearity
Short sample
Inadequate computers
No automatic verification of normal results
Material
Machine
Fishbone diagram as quality tools for root cause analysis of prolonged turnaround time.
within the entire diagnostic process. Also, difficulty in controlling this phase of testing is attributed to the lack of complete control or supervision of the laboratory of this phase (such as phlebotomy performed by nonlaboratory personnel) [22,23]. For proficiency testing, validation of quality control of our lab was performed from the data of internal QC and External Quality Assessment to establish the SD and bias, respectively, for each analyte. We obtained sigma of more than 6 for TG, CK-total; this implies that the analytical method in use is appropriate and needs no changes. The QC strategies that should be implemented in such cases need not be massive. The parameters that showed a wide variation in the sigma values should be evaluated. The methodology should be reassessed; there is also a need to strictly follow Westgard multirules as well as increase the number of QC runs to abolish this discrepancy. Total bilirubin, Cl, creatinine, glucose, sodium, urea, UA, ALT, ALP, and AST, showed the worst performance in our laboratory; focusing special attention to them is mandatory for improving performance. Most of the other parameters showed sigma metrics more than 3 indicating acceptable laboratory performance.
In a similar study carried out by Singh et al. [24], sigma value of more than 6 was observed for triglycerides, CPK total, and amylase for both the levels of QC. Creatinine and HDL showed a sigma value of more than 6 for the normal levels of quality control and 4.6 and 2.9 for the L3 of quality control, respectively. The sigma values for L2 and L3 were 5.9 and 7, respectively, for SGOT. They yielded sigma metrics in the range 3.1–5.9 for five parameters, namely, sugar, SGPT, ALP, bilirubin, and total protein. Cholesterol showed a sigma value of 3.3 for L2 and less than 3 (2.79) for L3. A sigma value of less than 3 for both the levels of QC was observed for urea, sodium, and K [24]. For critical results, the most common cause was inability to contact the authorized doctor, followed by wrong patient identification, lack of reporting by lab doctor, and finally, lack of reporting by lab technicians. After steps of improvement, there was a significant reduction (Po0.05) in all causes. Tate and Gardner [25] found that fewer than 10% of critical values were reported in their institution. They identified weaknesses in the reporting process as follows (a) critical values were not always reported by the clinical laboratory; (b) when critical
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
112 Journal of the Egyptian Public Health Association
values were reported, it was often to someone not directly involved in the patient’s care; (c) documentation of reporting of critical values the laboratory was incomplete; (d) clinicians’ awareness of critical values was not always documented; (e) clinicians’ decisions on corrective actions were not documented adequately; and (e) the time interval between the availability of critical test results and the institution of corrective measures was sometimes unacceptably long. Improvement in the reporting of critical values requires that the process is well defined and widely supported by users of laboratory services [25,26]. Laboratory TAT: analysis of laboratory TAT and possible causes of delay were determined and steps for improvement and TAT reduction were adopted (Fig. 1). The possible causes for prolonged TAT were stated as follows. TAT from registration–collection, receiving of all hospital wards at the same time without a definite time table for the distribution of samples during working hours leads to overcrowding of samples in the accession area. Inadequate numbers of computer systems and need for technician training lead to delayed registration and possible mistakes. The absence of the resident doctor in the accession area to manage our preanalytical phase leads several errors with the subsequent prolonged TAT. TAT from collection–receive, absence of the porter was the main cause of delay in this phase; receiving of samples in the central lab and samples’ scanning on the second barcode reader were another source of delay because sample receiving in central lab was not delegated to a defined technician. TAT from receiving–complete, any delay in this phase was most probably because of preanalytical causes rather than analytical factors. Examples include the following: (a) short samples which require extra steps of manual centrifugation, aliquoting, and manual loading on the analyzer; manual loading results in the risk of incomplete results with subsequent prolonged TAT, (b) Samples with results exceeding linearity and needing further dilution, and (c) samples not passing on the analytical barcode (second barcode) reader in the central lab causing inability of the results to be sent to LIS. A similar study was carried out by Chung et al. [27], and the results were as follows: for the distribution of TAT of the 13 594 specimens, 98.0% of the specimens were reported within 60 min, with an average TAT of 43.2 min, and the times taken to complete each of the three phases (preanalytical, analytical, and postanalytical) were 29.7, 13.9, and 0.2 min, respectively. The results of analysis that were reported after 60 min accounted for 2.0%, Test results reported between 60 and 90 min accounted for 89.5% of the specimens reported after 60 min, and results from analyses reported after 90 min accounted for the remaining 10.5%. TATs of the specimens reported between 60 and 90 min were 65.3 min and the time taken to complete each phase was 42.9, 22.4, and 0.02 min and for specimens after 90 min, the average TAT result was 96.9 min, and the time taken to complete each of the three phases were 35.5, 61.4, and 0.02.
In contrast to our study, the classification of TAT into three phases was on the basis of the following time points, that is, preanalytical phase (barcode printing– scanning on an autoanalyzer), analytical phase (scanning– result to LIS), and postanalytical phase (result to LIS– report to HIS). Second, blood specimens were transported by a conveyor belt. Next, centrifugation was performed by two staff members. Finally, the analytical phase was fixed according to the test items of each instrument because the variable ‘waiting time for analysis (specimen loading–barcode scanning)’ was excluded [27]. Study limitations
This study was carried out on inpatient samples and only three clinical wards were involved; they cannot be considered entirely representative of the current situation in laboratories. However, this selection was considered sufficient to perform an in-depth evaluation and to monitor the effective workflow and related procedures. We did not investigate an essential step of the preanalytical phase and the appropriateness of test requests; this aspect is a major concern for patient safety. We did not consider the effect of errors on patient care and safety; that is, the percentage of errors that have resulted in unnecessary laboratory test repetition, further inappropriate investigations, and episodes of negative clinical outcomes. In addition, we did not show the frequency of errors attributable to processes external to the laboratory, particularly in the preanalytical phase, and those attributable to procedures performed when the sample reaches the laboratory. However, the strength of the study lies in its exploration of the initial steps of the testing cycle and its focus on the role of the clinical laboratory in understanding and improving the procedures performed in the preanalytical phase.
Conclusion and recommendations The present study showed a high prevalence of errors in all steps of the analysis cycle (preanalytical and postanalytical steps). Most errors occur in the preanalytical phase of TTP. A significant improvement in the performance of the Alexandria University Hospital Clinical Chemistry Laboratories was evident after corrective action had been taken.
Acknowledgements The authors certify that they have no affiliations with or involvement in any organization or entity with financial or non financial interest.
Conflicts of interest There are no conflicts of interest.
References 1 Agency for Healthcare Research and Quality. National quality measures clearinghouse: desirable measure attributes. Available at: http://www.qualitymeasures.ahrq.gov/resources/measure_use.aspx#attributes [Accessed 25 February 2013]. 2 Bovend’Eerdt TJH. Writing SMART rehabilitation goals and achieving goal attainment scaling: a practical guide. Clin Rehabil 2009; 23:4352–4361.
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 113
3 Plebani M1, Chiozza ML. Audit in laboratory medicine. Eur J Clin Chem Clin Biochem 1996; 34:655–657.
15 www.cap.org. Laboratory accreditation program hematology checklist. Northfield, IL: College of American Pathologists.
4 International Organization for Standardization. ISO 15189: medical laboratories – particular requirements for quality and competence. Geneva: International Organization for Standardization.
16 Plebani M. Errors in clinical laboratories or errors in laboratory medicine? Clin Chem Lab Med 2006; 44:750–759. 17 Sciacovelli L, Plebani M. The IFCC Working Group on laboratory errors and patient safety. Clin Chim Acta 2009; 404:79–85. 18 Lippi G1, Guidi GC, Mattiuzzi C, Plebani M. Preanalytical variability: the dark side of the moon in laboratory testing. Clin Chem Lab Med 2006; 44: 358–365.
5 Lippi G1, Banfi G, Buttarello M, Ceriotti F, Daves M, Dolci A, et al. Recommendations for detection and management of unsuitable samples in clinical laboratories. Clin Chem Lab Med 2007; 45:728–736. 6 Bonini P1, Plebani M, Ceriotti F, Rubboli F. Errors in laboratory medicine. Clin Chem 2002; 48:691–698. 7 Richardson H, Gun-Munro J, Petersons A, Webb S. External quality assessment in the quality management of POCT. J Near Patient Test Technol 2005; 4:164–166. 8 ISO/DIS 8402. Quality management and quality assurance vocabulary. The International Organization for Standardization: ISO/DIS. 9 Westgard JO1, Burnett RW, Bowers GN. Quality management science in clinical chemistry: a dynamic framework for continuous improvement of quality. Clin Chem 1990; 36:1712–1716. 10 Carraro P1, Plebani M. Errors in a stat laboratory: types and frequencies 10 years later. Clin Chem 2007; 53:1338–1342. 11 Howanitz PJ1, Steindel SJ, Heard NV. Laboratory critical values policies and procedures: a college of American Pathologists Q-Probes Study in 623 institutions. Arch Pathol Lab Med 2002; 126:663–669. 12 Plebani M, Wu A. Foreward. Clin Chim Acta 2004; 346:1–2. 13 Lippi G1, Giavarina D, Montagnana M, Luca Salvagno G, Cappelletti P, Plebani M, Guidi GC. SIBioC (Italian Society of Clinical Biochemistry and Clinical Molecular Biology); SIMEL (Italian Society of Laboratory Medicine); CISMEL (Italian Committee for Standardization of Laboratory and Hematological Methods); Inter-associative Study Group on the Extra-Analytical Variability of Laboratory Testing. National survey on critical values reporting in a cohort of Italian laboratories. Clin Chem Lab Med 2007; 45:1411–1413. 14 Harry M, Schroeder R. Six SIGMA: the breakthrough management strategy revolutionizing the world’s top corporations. A CURRENCY Book, published by doubleday, a division of Random House Inc.
19 Westgard JO, Westgard SA. The quality of laboratory testing today: an assessment of sigma metrics for analytic quality using performance data from proficiency testing surveys and the CLIA criteria for acceptable performance. Am J Clin Pathol 2006; 125:343–354. 20 Westgard JO, Klee GG. Quality management. In: Burtis CA, Ashwood ER, Bruns DE, editors. Tietz textbook of clinical chemistry. Philadelphia: Saunders. 21 Chhillar N, Khurana S, Agarwal R, Singh NK. Effect of pre-analytical errors on quality of laboratory medicine at a neuropsychiatry institute in north India. Indian J Clin Biochem 2011; 26:46–49. 22 Titus K. Clinicians talk shop – here’s what they want. CAP Today 2003; 17:1–82. 23 Stankovic AK. The laboratory is a key partner in assuring patient safety. Clin Lab Med 2004; 24:1023–1035. 24 Singh B, Goswami B, Gupta VK, Chawla R, Mallika V. Application of sigma metrics for the assessment of quality assurance in clinical biochemistry laboratory in India: a pilot study. Indian J Clin Biochem 2011; 26:131–135. 25 Tate KE1, Gardner RM. Computers, quality, and the clinical laboratory: a look at critical value reporting. Proc Annu Symp Comput Appl Med Care 1993; 114:193–197. 26 Lippi G, Plebani M. Informatics aids to reduce failure rates in notification of abnormal outpatient test results. Arch Intern Med 2009; 169:1815 author reply 1816–1817. 27 Chung HJ1, Lee W, Chun S, Park HI, Min WK. Analysis of turnaround time by subdividing three phases for outpatient chemistry specimens. Ann Clin Lab Sci 2009; 39:144–149.
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Evaluating laboratory key performance using quality indicators in Alexandria University Hospital Clinical Chemistry Laboratories Mostafa M. Rizka, Adel Zakib, Nermine Hossama and Yasmin Aboul-Elaa a Clinical Pathology Department, Faculty of Medicine and bBiostatistics Department, Medical Research Institute, Alexandria University, Egypt
Correspondence to Nermine Hossam, MD, Clinical Pathology Department, Faculty of Medicine, Alexandria University, Alexandria, 21311 Egypt Tel: + 20 122 337 4277; fax: + 002035411281; e-mail: [email protected]
Received 30 November 2013 Accepted 20 July 2014 Journal of the Egyptian Public Health Association 2014, 89:105–113
Background The performance of clinical laboratories plays a fundamental role in the quality and effectiveness of healthcare. Objectives To evaluate the laboratory performance in Alexandria University Hospital Clinical Laboratories using key quality indicators and to compare the performance before and after an improvement plan based on ISO 15189 standards. Materials and methods The study was carried out on inpatient samples for a period of 7 months that was divided into three phases: phase I included data collection for evaluation of the existing process before improvement (March–May 2012); an intermediate phase, which included corrective, preventive action, quality initiative and steps for improvement (June 2012); and phase II, which included data collection for evaluation of the process after improvement (July 2012–September 2012). Results In terms of the preanalytical indicators, incomplete request forms in phase I showed that the total number of received requests were 31 944, with a percentage of defected request of 33.66%; whereas in phase II, there was a significant reduction in all defected request items (Po0.001) with a percentage of defected requests of 9.64%. As for the analytical indicators, the proficiency testing accuracy score in phase I showed poor performance of 10 analytes in which total error (TE) exceeded total error allowable (TEa), with a corresponding sigma value of less than 3, which indicates test problems and an unreliable method. The remaining analytes showed an acceptable performance in which TE did not exceed the TEa, with a sigma value of more than 6. Following an intervention of 3 months, the performance showed marked improvement. Error tracking in phase I showed a TE of (5.11%), whereas in phase II it was reduced to 2.48% (Po0.001). For the postanalytical indicators, our results in phase I showed that the percentage of nonreported critical results was 26.07%. In phase II, there was a significant improvement (Po0.001). The percentage of nonreported results was 11.37%, the reasons were either inability to contact the authorized doctor (8.24%), wrong patient identification (1.0%), lack of reporting by lab doctor (1.11%), and finally, lack of reporting by the lab technician (1.03%). Conclusion and recommendations Standardization and monitoring of each step in the total testing process is very important and is associated with the most efficient and well-organized laboratories. Keywords: external quality assessment, key performance indicators, quality indicators, turnaround time, total testing process J Egypt Public Health Assoc 89:105–113 & 2014 Egyptian Public Health Association 0013-2446
Introduction Quality indicator (QI) is a quality tool enabling laboratories seeking improvement to quantify the laboratory’s performance by selecting a certain comparative criterion; its aim is to assess the performance to initiate corrective measures to ensure continual improvement [1]. When identifying the QIs, the acronym SMART is often applied. SMART denotes selected goals that should be specific, measurable, achievable, realistic, and timely [2]. 0013-2446 & 2014 Egyptian Public Health Association
Assessment of the quality of laboratory services using QIs requires a systematic, transparent, and consistent approach for collection and analysis of data. QIs were classified as preanalytical, analytical, and postanalytical according to the phase of the laboratory process they measure. Classically, the preanalytical phase includes all processes from the time a laboratory request is made by a physician until the sample is ready for testing; this phase is complex DOI: 10.1097/01.EPX.0000453262.85383.70
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
106
Journal of the Egyptian Public Health Association
and involves many processes, most of them external to the laboratory, and is prone to the maximum variation and the highest proportion of errors [3]. ISO 15189: 2007 standard specifies the information that must be provided on the request form, irrespective of whether the test requisition is made on paper or electronically as follows: unique identification of the patient; name or other unique identifier of the physician or other legally authorized individual; and type of primary sample, tests requested, and date and time of primary sample collection and relevant clinical information about the patient necessary for interpretation purposes; at a minimum, this should include sex and date of birth [4].
tion must understand and address the variables involved in the process. Documentation and communication processes must be regularly monitored and implemented under ongoing systems for quality monitoring [11].
Acceptance of improper specimens for analysis (insufficient sample volume, inappropriate collection container, hemolyzed, lipemic, icteric, or clotted specimens, inadequate ratio volume sample/anticoagulant, and unidentified sample) may lead to erroneous results that could affect patient care, but only by monitoring the rejected specimens on a regular basis and identifying factors associated with the rejection; can the lab avoid errors and promote continuous quality improvement of the laboratory service offered [5].
Accuracy, precision, timeliness, and authenticity are the four pillars of efficient laboratory services. However, timeliness, which is expressed as the turnaround time (TAT), is often used by the clinicians as the benchmark for laboratory performance; laboratorians may disagree with such a priority, arguing that unless analytical quality can be achieved, none of the other characteristics matter [15]. Delays in TAT elicit immediate complaints from users while adequate TAT goes unremarked. Unsatisfactory TAT is a major source of complaints to the laboratory in terms of poor service and requires much time and effort from laboratory staff in complaint resolution and service improvement [16].
Quality management of the analytical phase of the testing process is the most standardized and regulated part and has received the most attention [6]. External Quality Assessment programs are designed to provide a regular, objective, and independent assessment of a participant’s ability to provide an acceptable standard of service by comparison with peers [7]. Each laboratory is provided with a numerical indicator of its competence, a performance index or score, together with information on the performance of the group as a whole, enabling its proficiency relative to the group to be compared. The participating laboratories use the information on their performance to make appropriate changes and improvements [8]. The total testing process (TTP) consists of a series of interrelated processes, each involving a series of process steps, every one of which can result in an error. TTP consists of series of activities, starting with the clinical question in the clinician’s mind, leading to test selection, sample collection, transport to the laboratory, analysis, reporting back to the clinician, and final interpretation and decision making by the clinician. Identifying and learning from errors is stated to be a key factor for improving the quality and safety in healthcare; to guarantee excellent performance and service, the process of identifying and treating error risks must be integrated into the TTP. A suitable system for grading laboratory errors on the basis of their seriousness should help identify priorities for quality improvement and enable a focus on corrective/preventive actions [9]. Improving quality in the postanalytical phase represents as such a valuable goal for reducing errors and improving patient safety. A key issue in postanalytical quality is represented by the effectiveness of laboratory data communication, particularly communication of critical test results [10]. For the critical value reporting process to be effective, the organiza-
The content, format, and physical presentation of the information significantly affect the interpretation and use of laboratory data by clinicians [12,13]. Mistakes in the content and completeness of laboratory reports as well as misunderstanding by the treating physician as to the significance of the information in the report, among other factors, can delay the treatment of a serious disease and alter outcomes [14].
Reliability cannot be achieved in a clinical laboratory through the control of accuracy in the analytical phase of the testing process alone, but is also determined by preanalytical and postanalytical factors [17]. Thus, the aim of this study was to promote accuracy in the Alexandria University Hospital Laboratory (AUHL) including the analytical phase as well as assurance of reliability of preanalytical and postanalytical activities.
Materials and methods Phase I
The laboratory is an ISO 15189–2007-certified laboratory affiliated to a public university hospital serving a population of 234 403. It provides care for inpatients, outpatients, and emergency cases. Samples were collected by nurses, transported to the laboratory sample reception desk for classification, and distributed to the appropriate units where testing was conducted. In 2012, 262 100 requests and 2 668 984 tests were performed. A study was carried out on all inpatient tests presented to the Clinical Chemistry laboratory for a period of 7 months that was divided into three phases: phase I included data collection for evaluation of the existing process (March–May 2012), the intermediate phase included corrective, preventive action, quality initiative, and steps for improvement (June 2012), and phase II included data collection for evaluation of the process after improvement (July–September 2012). QIs were classified as preanalytical, analytical, and postanalytical according to the laboratory process that they measured. Preanalytical QIs included incomplete request forms, and rejected samples. Analytical QIs
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 107
included accuracy of proficiency testing. Postanalytical QIs included critical result report and laboratory TAT [18]. An incomplete request form QI was used to reveal any omission of demographic items (age, requesting doctor, suspected diagnosis, etc.) per total requests. The next QI referred to unsuitable specimens. This indicator was monitored by defining a preanalytical sample error as a rejected specimen per total number of received samples: any sample that was not suitable for one or more tests in the total order because the specimen did not fulfill the acceptability criteria, or the sample was not received [19]. The analytical QI included accuracy of proficiency testing, calculated from the external quality program report. This report was based on Six-Sigma results and reported every 3 months to each functional unit. X ðTEa biasÞ Sigma metrics : ¼ ; SD of the analytes measured where TEa is the total allowable error. TEa values of various parameters were taken from the Clinical Laboratories Improvement Act guidelines [20]. Bias was computed from the external quality assurance records using the following formula: bias = mean of all laboratories using the same instrument and method – our results. SD was determined from the calculated laboratory mean and calculated SD obtained from the internal QC data. Critical result reporting: for immediate notification of a physician (or other clinical personnel responsible for patient care) when examination results for critical properties fall within established ‘alert’ or ‘critical’ intervals, a hard copy of the report would be sent to the appropriate healthcare provider as soon as possible. It was calculated as number of nonreported critical results/total number of critical results. Laboratory TAT was measured from the time of test registration to the time of results verification; data were obtained from lab information system (LIS). All the key performance indicators were expressed as percentages, except for TAT (expressed in days) and QPIs (as average score). Intervention
The major link in the laboratory service and the hospital is through the nursing staff, which act as phlebotomists according to the AUH policy. Multiple methods were used to determine the educational needs and training for the nurses as well as the phlebotomist; these include: evaluating knowledge and skills in job description, asking nurses supervisors, testing nurses, and phlebotomists, and analyzing past performance appraisal. Educational lectures and video films were presented to nurses and technicians together with workshops with a special focus on quality awareness, quality participation, ISO 15189 standards, AUHL mission and vision. Sessions were divided for the preanalytical team to include phlebotomy skills and preanalytical standards and errors.
Analytical team sessions included quality concepts, westguard rules, inventory management, and proficiency testing evaluation. The for postanalytical team including medical secretaries and nurses as well as technicians, received a special session on the critical results reporting and importance of diagnosis registration and its benefits for clinical correlation, learning the standardized abbreviation for common diseases, and development of a data base for all resident doctors including name, ward, and mobile number to facilitate communication with the responsible physician, which lead to effective critical value reporting and diminished failure of communication between the lab and the clinical wards in the hospital. For laboratory TAT, the steps of improvement adopted were as follows: (1) Development of a time table for sample receiving from wards to avoid overcrowding in the accession area and delaying of sample registration and transportation. (2) Increase the number of computers used in the accession area of the lab with increasing number of personnel for data entry. (3) Implementation of a regular schedule for transportation of the sample from the accession area to the central analytical lab associated with a work list. (4) Availability of the resident doctor in the accession area. (5) Formal instruction to all wards for minimal sample requirement to avoid problems of short samples. (6) Delegation of passage of the sample on a barcode reader in the central lab to a certain technician to decrease TAT. (7) Regular check for incomplete results and results above linearity for further dilution.
Statistical analysis
It was carried out using the PSWA Statistics software (version 18; SPSS, Chicago, Illinois, USA). Categorical variables were described using frequencies and percentages. The w2-test was used to test associations. A statistical significance level of 95% (Po0.05) was considered for all statistical analyses. All tests used in this study were two-sided.
Results Preanalytical phase
In terms of the incomplete request form and rejected samples, results showed that there was a significant statistical improvement between phase I and phase II (Po0.05) (Tables 1 and 2). For the incomplete request form, phase I showed that the total number of received requests was 31 944, absent admission number was missing in 590 (1.85%), absent requested doctor identification in 2028 (6.35%), and absent diagnosis in 558 (1.75%). In phase II, the total
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
108
Journal of the Egyptian Public Health Association
Table 1. Frequency of absent information that must be provided on the request form presented as a percentage of total received requests before (phase I) and after (phase II) improvement Number of defects (% of received request) Defected item Incomplete patient information Absent admission number Absent patient preparation Absent physician ID Absent diagnosis Absent date and time Absent type of sample
Phase I (N = 31 944) 326 590 949 2028 558 3126 3176
Phase II (N = 28 286)
(1.02) (1.85) (2.97) (6.35) (1.75) (9.79) (9.94)
68 80 475 965 184 503 452
(0.24) (0.28) (1.68) (3.41) (0.65) (1.78) (1.60)
P value 0.001 0.001 0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 2. Frequency of different causes of sample rejection presented as a percentage of total received sample before (phase I) and after (phase II) improvement Received sample [n (%)] Cause of rejection Hemolysis Clotted Sample/anticoagulant ratio Miss-identification Lipemic Total
Phase I (N = 50 440) 1584 494 179 41 16 2314
Phase II (N = 45 180)
(3.14) (0.98) (0.35) (0.08) (0.03) (4.59)
841 311 95 32 6 1285
(1.86) (0.69) (0.21) (0.07) (0.01) (2.84)
P value 0.001 0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
number of received requests was 28 286, admission number was missed in 80 (0.28%), and absent doctor identification in 965 (3.41%). For the rejected sample, in phase I, the total number of received samples was 50 440, with a total number of rejected samples of 2314 (4.59%), whereas in phase II, the total number of received samples was 45 180, and the total number of rejected samples was 1285 (2.84%). Analytical phase
The proficiency testing accuracy score measured by the sigma metrics in phase I showed sigma value more than 6 for triglycerides and creatine kinase (CK). Sigma value in the range 3–5.9 was observed for three parameters, namely, albumin, potassium (K), and total protein. A sigma value of less than 3 was observed for total bilirubin, calcium, chloride (Cl), cholesterol, creatinine, glucose, sodium, urea, uric acid (UA), alanine aminotransferase (ALT), alkaline phosphatase (ALP), aspartate aminotransferase (AST), and lactate dehydrogenase. In phase II, sigma value of more than 6 was observed for UA and CK and a value in the range 3–5.9 was observed for 13 parameters, namely, albumin, calcium, cholesterol, creatinine, glucose, K, total protein, Na, triglyceride, urea, ALT, ALP, AST, and lactate dehydrogenase. A sigma value of less than 3 was observed for total bilirubin and Cl (Table 3). Postanalytical phase
Study of critical results reporting and laboratory TAT monitoring showed a significant improvement between phase I and phase II. As for critical results reporting, in phase I, the total number of critical results to be communicated was 3180, the total number of nonreported
results was 829 (26.07%), whereas in phase II, the total number of critical results to be communicated was 2805 and the total number of nonreported results was 319 (11.37%) (with Po0.05) (Tables 4–6). For laboratory TAT, in phase I, the distribution of TAT was on average of 172.7 min as shown in (Tables 7 and 8). For specimens reported within 120 min, the average TAT was 95.58 min (30.44%). For specimens reported between 120 and 180 min, the average TATwas 148 min (39.21%), and for specimens reported more than 180, the average TAT was 273.8 min (30.35%). In phase II, the distribution of TAT was an average of 151.29 min. For specimens reported within 120 min, the average TAT was 93.11 min (72.44%), for specimens reported between 120 and 180 min, the average TAT was 139.4 min (24.78%), and for specimens reported more than 180, the average TAT was 221.36 min (2.78%).
Discussion The major aim of the quality system is to reduce or, ideally, eliminate all defects within the entire testing process. Lippi et al. [13] emphasized the development of a strategy to enhance quality throughout the TTP. This strategy starts with a systematic analysis of specimen workflow, continues with education and thorough monitoring by error tracking systems, and ends with the elimination or redesign of deficient procedures. To our knowledge, this is the first study to assess the performance of the laboratory services in a University Hospital in Alexandria. A similar study carried out by Chhillar et al. [21] reported that the major preanalytical errors of concern noticed were that the treating physician’s names were missing in 13.1% of the forms
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 109
Table 3. Total errors allowable, average bias, SD, and sigma values for the different parameters before (phase I) and after (phase II) improvement
ALP, alkaline phosphatase; ALT, alanine aminotransferase; CK, creatine kinase; LDH, lactate dehydrogenase; TEa, total errors allowable; UA, uric acid.
and their signatures were missing in almost 13.4% of the lab forms. Clinical information was illegible or difficult to decipher in 89.25% of the forms. The diagnosis was not reported in 61.2% of the lab request forms and the information pertaining on the type of specimen was missing in 61.6% of the forms [21].
As for rejected samples, phase II showed a significant reduction (Po0.01). This is in agreement with Lippii et al. [18], who reported that the lack of a standardized protocol for sample collection, including patient preparation, specimen acquisition, handling, and storage, accounts for up to 93% of the errors currently encountered
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
110 Journal of the Egyptian Public Health Association
Table 4. Frequency of different type of errors before (phase I) and after (phase II) improvement Number of error (% of total received) Type of error
Phase I (N = 27 612)
Missed result Missed sample (order) Missed sample (received) Incomplete results Total
296 199 160 755 1410
Phase II (N = 24 507)
(1.07) (0.72) (0.58) (2.73) (5.11)
141 85 57 325 608
P value
(0.58) (0.35) (0.23) (1.33) (2.48)
0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 5. Frequency of different causes of inaccurate report per total number of released reports presented before (phase I) and after (phase II) improvement Total released [n (%)] Defected item
Phase I (N = 27 612)
Patient information Patient code Location Comment Signature of the authorized person Total
420 630 210 320 50 1630
Phase II (N = 24 507)
(1.52) (2.28) (0.76) (1.16) (0.18) (5.90)
80 95 110 125 35 445
P value
(0.33) (0.39) (0.45) (0.51) (0.14) (1.82)
0.001 0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 6. Frequency of causes of nonreported critical results before (phase I) and after (phase II) improvement Number of defects (% of total received) Causes
Phase I (N = 3180)
Wrong patient identification Lack of reporting by lab technician Lack of reporting by lab physician Inability to contact the authorized physician Total
109 80 94 546 829
Phase II (N = 2805)
(3.43) (2.52) (2.96) (17.17) (26.07)
28 29 31 231 319
P value
(1.00) (1.03) (1.11) (8.24) (11.37)
0.001 0.001 0.001 0.001 0.001
Statistically significant at Pr0.05.
Table 7. The mean and proportions of overall turnaround time before (phase I) and after (phase II) improvement Phase I Group
Phase II
Received sample [n (%)] Mean TAT (min) Received sample [n (%)] Mean TAT (min) P value
Specimens reported within 120 min Specimens reported between 120 and 180 min Specimen reported after 180 min Total
673 867 671 2211
(30.44) (39.21) (30.35) (100.00)
95.58 148.76 273.80 172.71
1590 544 61 2195
(72.44) (24.78) (2.78) (100.00)
93.11 139.4 221.36 151.29
0.001 0.001 0.001 0.001
Statistically significant at Pr0.05. TAT, turnaround time.
Table 8. The mean turnaround time of each phase before (phase I) and after (phase II) improvement Phase I
Phases Mean TAT (min)
Phase II
Registration– collection
Collection– receive
Receive– complete
Registration– collection
Collection– receive
Receive– complete
P value
15.28
16.61
140.82
11.36
12.25
127.68
0.001
Statistically significant at Pr0.05. TAT, turnaround time.
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 111
Figure 1.
Manpower
Method
Porter busy or not available
Receiving all wards in the same time
Inadequate number of technician
Samples sent to central lab without work list Problem in sample receiving in central lab
Absence of resident doctor in accession Training required on computer system
Prolonged turnaround time Absence automatic dilution of result exceeding linearity
Short sample
Inadequate computers
No automatic verification of normal results
Material
Machine
Fishbone diagram as quality tools for root cause analysis of prolonged turnaround time.
within the entire diagnostic process. Also, difficulty in controlling this phase of testing is attributed to the lack of complete control or supervision of the laboratory of this phase (such as phlebotomy performed by nonlaboratory personnel) [22,23]. For proficiency testing, validation of quality control of our lab was performed from the data of internal QC and External Quality Assessment to establish the SD and bias, respectively, for each analyte. We obtained sigma of more than 6 for TG, CK-total; this implies that the analytical method in use is appropriate and needs no changes. The QC strategies that should be implemented in such cases need not be massive. The parameters that showed a wide variation in the sigma values should be evaluated. The methodology should be reassessed; there is also a need to strictly follow Westgard multirules as well as increase the number of QC runs to abolish this discrepancy. Total bilirubin, Cl, creatinine, glucose, sodium, urea, UA, ALT, ALP, and AST, showed the worst performance in our laboratory; focusing special attention to them is mandatory for improving performance. Most of the other parameters showed sigma metrics more than 3 indicating acceptable laboratory performance.
In a similar study carried out by Singh et al. [24], sigma value of more than 6 was observed for triglycerides, CPK total, and amylase for both the levels of QC. Creatinine and HDL showed a sigma value of more than 6 for the normal levels of quality control and 4.6 and 2.9 for the L3 of quality control, respectively. The sigma values for L2 and L3 were 5.9 and 7, respectively, for SGOT. They yielded sigma metrics in the range 3.1–5.9 for five parameters, namely, sugar, SGPT, ALP, bilirubin, and total protein. Cholesterol showed a sigma value of 3.3 for L2 and less than 3 (2.79) for L3. A sigma value of less than 3 for both the levels of QC was observed for urea, sodium, and K [24]. For critical results, the most common cause was inability to contact the authorized doctor, followed by wrong patient identification, lack of reporting by lab doctor, and finally, lack of reporting by lab technicians. After steps of improvement, there was a significant reduction (Po0.05) in all causes. Tate and Gardner [25] found that fewer than 10% of critical values were reported in their institution. They identified weaknesses in the reporting process as follows (a) critical values were not always reported by the clinical laboratory; (b) when critical
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
112 Journal of the Egyptian Public Health Association
values were reported, it was often to someone not directly involved in the patient’s care; (c) documentation of reporting of critical values the laboratory was incomplete; (d) clinicians’ awareness of critical values was not always documented; (e) clinicians’ decisions on corrective actions were not documented adequately; and (e) the time interval between the availability of critical test results and the institution of corrective measures was sometimes unacceptably long. Improvement in the reporting of critical values requires that the process is well defined and widely supported by users of laboratory services [25,26]. Laboratory TAT: analysis of laboratory TAT and possible causes of delay were determined and steps for improvement and TAT reduction were adopted (Fig. 1). The possible causes for prolonged TAT were stated as follows. TAT from registration–collection, receiving of all hospital wards at the same time without a definite time table for the distribution of samples during working hours leads to overcrowding of samples in the accession area. Inadequate numbers of computer systems and need for technician training lead to delayed registration and possible mistakes. The absence of the resident doctor in the accession area to manage our preanalytical phase leads several errors with the subsequent prolonged TAT. TAT from collection–receive, absence of the porter was the main cause of delay in this phase; receiving of samples in the central lab and samples’ scanning on the second barcode reader were another source of delay because sample receiving in central lab was not delegated to a defined technician. TAT from receiving–complete, any delay in this phase was most probably because of preanalytical causes rather than analytical factors. Examples include the following: (a) short samples which require extra steps of manual centrifugation, aliquoting, and manual loading on the analyzer; manual loading results in the risk of incomplete results with subsequent prolonged TAT, (b) Samples with results exceeding linearity and needing further dilution, and (c) samples not passing on the analytical barcode (second barcode) reader in the central lab causing inability of the results to be sent to LIS. A similar study was carried out by Chung et al. [27], and the results were as follows: for the distribution of TAT of the 13 594 specimens, 98.0% of the specimens were reported within 60 min, with an average TAT of 43.2 min, and the times taken to complete each of the three phases (preanalytical, analytical, and postanalytical) were 29.7, 13.9, and 0.2 min, respectively. The results of analysis that were reported after 60 min accounted for 2.0%, Test results reported between 60 and 90 min accounted for 89.5% of the specimens reported after 60 min, and results from analyses reported after 90 min accounted for the remaining 10.5%. TATs of the specimens reported between 60 and 90 min were 65.3 min and the time taken to complete each phase was 42.9, 22.4, and 0.02 min and for specimens after 90 min, the average TAT result was 96.9 min, and the time taken to complete each of the three phases were 35.5, 61.4, and 0.02.
In contrast to our study, the classification of TAT into three phases was on the basis of the following time points, that is, preanalytical phase (barcode printing– scanning on an autoanalyzer), analytical phase (scanning– result to LIS), and postanalytical phase (result to LIS– report to HIS). Second, blood specimens were transported by a conveyor belt. Next, centrifugation was performed by two staff members. Finally, the analytical phase was fixed according to the test items of each instrument because the variable ‘waiting time for analysis (specimen loading–barcode scanning)’ was excluded [27]. Study limitations
This study was carried out on inpatient samples and only three clinical wards were involved; they cannot be considered entirely representative of the current situation in laboratories. However, this selection was considered sufficient to perform an in-depth evaluation and to monitor the effective workflow and related procedures. We did not investigate an essential step of the preanalytical phase and the appropriateness of test requests; this aspect is a major concern for patient safety. We did not consider the effect of errors on patient care and safety; that is, the percentage of errors that have resulted in unnecessary laboratory test repetition, further inappropriate investigations, and episodes of negative clinical outcomes. In addition, we did not show the frequency of errors attributable to processes external to the laboratory, particularly in the preanalytical phase, and those attributable to procedures performed when the sample reaches the laboratory. However, the strength of the study lies in its exploration of the initial steps of the testing cycle and its focus on the role of the clinical laboratory in understanding and improving the procedures performed in the preanalytical phase.
Conclusion and recommendations The present study showed a high prevalence of errors in all steps of the analysis cycle (preanalytical and postanalytical steps). Most errors occur in the preanalytical phase of TTP. A significant improvement in the performance of the Alexandria University Hospital Clinical Chemistry Laboratories was evident after corrective action had been taken.
Acknowledgements The authors certify that they have no affiliations with or involvement in any organization or entity with financial or non financial interest.
Conflicts of interest There are no conflicts of interest.
References 1 Agency for Healthcare Research and Quality. National quality measures clearinghouse: desirable measure attributes. Available at: http://www.qualitymeasures.ahrq.gov/resources/measure_use.aspx#attributes [Accessed 25 February 2013]. 2 Bovend’Eerdt TJH. Writing SMART rehabilitation goals and achieving goal attainment scaling: a practical guide. Clin Rehabil 2009; 23:4352–4361.
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
Performance of university in Alexandria University clinical chemistry laboratories Rizk et al. 113
3 Plebani M1, Chiozza ML. Audit in laboratory medicine. Eur J Clin Chem Clin Biochem 1996; 34:655–657.
15 www.cap.org. Laboratory accreditation program hematology checklist. Northfield, IL: College of American Pathologists.
4 International Organization for Standardization. ISO 15189: medical laboratories – particular requirements for quality and competence. Geneva: International Organization for Standardization.
16 Plebani M. Errors in clinical laboratories or errors in laboratory medicine? Clin Chem Lab Med 2006; 44:750–759. 17 Sciacovelli L, Plebani M. The IFCC Working Group on laboratory errors and patient safety. Clin Chim Acta 2009; 404:79–85. 18 Lippi G1, Guidi GC, Mattiuzzi C, Plebani M. Preanalytical variability: the dark side of the moon in laboratory testing. Clin Chem Lab Med 2006; 44: 358–365.
5 Lippi G1, Banfi G, Buttarello M, Ceriotti F, Daves M, Dolci A, et al. Recommendations for detection and management of unsuitable samples in clinical laboratories. Clin Chem Lab Med 2007; 45:728–736. 6 Bonini P1, Plebani M, Ceriotti F, Rubboli F. Errors in laboratory medicine. Clin Chem 2002; 48:691–698. 7 Richardson H, Gun-Munro J, Petersons A, Webb S. External quality assessment in the quality management of POCT. J Near Patient Test Technol 2005; 4:164–166. 8 ISO/DIS 8402. Quality management and quality assurance vocabulary. The International Organization for Standardization: ISO/DIS. 9 Westgard JO1, Burnett RW, Bowers GN. Quality management science in clinical chemistry: a dynamic framework for continuous improvement of quality. Clin Chem 1990; 36:1712–1716. 10 Carraro P1, Plebani M. Errors in a stat laboratory: types and frequencies 10 years later. Clin Chem 2007; 53:1338–1342. 11 Howanitz PJ1, Steindel SJ, Heard NV. Laboratory critical values policies and procedures: a college of American Pathologists Q-Probes Study in 623 institutions. Arch Pathol Lab Med 2002; 126:663–669. 12 Plebani M, Wu A. Foreward. Clin Chim Acta 2004; 346:1–2. 13 Lippi G1, Giavarina D, Montagnana M, Luca Salvagno G, Cappelletti P, Plebani M, Guidi GC. SIBioC (Italian Society of Clinical Biochemistry and Clinical Molecular Biology); SIMEL (Italian Society of Laboratory Medicine); CISMEL (Italian Committee for Standardization of Laboratory and Hematological Methods); Inter-associative Study Group on the Extra-Analytical Variability of Laboratory Testing. National survey on critical values reporting in a cohort of Italian laboratories. Clin Chem Lab Med 2007; 45:1411–1413. 14 Harry M, Schroeder R. Six SIGMA: the breakthrough management strategy revolutionizing the world’s top corporations. A CURRENCY Book, published by doubleday, a division of Random House Inc.
19 Westgard JO, Westgard SA. The quality of laboratory testing today: an assessment of sigma metrics for analytic quality using performance data from proficiency testing surveys and the CLIA criteria for acceptable performance. Am J Clin Pathol 2006; 125:343–354. 20 Westgard JO, Klee GG. Quality management. In: Burtis CA, Ashwood ER, Bruns DE, editors. Tietz textbook of clinical chemistry. Philadelphia: Saunders. 21 Chhillar N, Khurana S, Agarwal R, Singh NK. Effect of pre-analytical errors on quality of laboratory medicine at a neuropsychiatry institute in north India. Indian J Clin Biochem 2011; 26:46–49. 22 Titus K. Clinicians talk shop – here’s what they want. CAP Today 2003; 17:1–82. 23 Stankovic AK. The laboratory is a key partner in assuring patient safety. Clin Lab Med 2004; 24:1023–1035. 24 Singh B, Goswami B, Gupta VK, Chawla R, Mallika V. Application of sigma metrics for the assessment of quality assurance in clinical biochemistry laboratory in India: a pilot study. Indian J Clin Biochem 2011; 26:131–135. 25 Tate KE1, Gardner RM. Computers, quality, and the clinical laboratory: a look at critical value reporting. Proc Annu Symp Comput Appl Med Care 1993; 114:193–197. 26 Lippi G, Plebani M. Informatics aids to reduce failure rates in notification of abnormal outpatient test results. Arch Intern Med 2009; 169:1815 author reply 1816–1817. 27 Chung HJ1, Lee W, Chun S, Park HI, Min WK. Analysis of turnaround time by subdividing three phases for outpatient chemistry specimens. Ann Clin Lab Sci 2009; 39:144–149.
Copyright © Journal of the Egyptian Public Health Association. Unauthorized reproduction of this article is prohibited.
