The current practices regarding the minimum frequency of QC are often based on regulatory requirements, such as CLIA’s minimum of 2 levels of control per day or CMS’s more recent guidelines for “Equivalent QC”, which might reduce the frequency to 2 levels once per week or even 2 levels once per month. Another approach under development by CLSI recommends the use of risk management for customizing QC in the laboratory. In this approach for “alternate QC,” manufacturers would make a risk assessment of their analytic system, eliminate risks of failure when possible, mitigation those risks, and report the “residual risks” to the laboratory, which should then help the laboratory customize its QC system. This customization could employ a variety of mechanisms for prevention and control, a primary one being the adjustment of the frequency of QC to monitor the remaining risks. These practices have been developed with primary consideration of an analytical system operating in a laboratory environment, subject to the changing conditions, analyst, and environment. As such, they reflect the isolation of the analytical testing system in a laboratory, rather than considering the context of health care delivery as a whole. A different approach would seem appropriate to establish guidelines for QC frequency in the context of patient safety. In this case we need to look at the whole process wherein an erroneous lab test result may occur and can compromise patient safety.
The current practices regarding the minimum frequency of QC are often based on regulatory requirements, such as CLIA’s minimum of 2 levels of control per day or CMS’s more recent guidelines for “Equivalent QC”, which might reduce the frequency to 2 levels once per week or even 2 levels once per month. Another approach under development by CLSI recommends the use of risk management for customizing QC in the laboratory. In this approach for “alternate QC,” manufacturers would make a risk assessment of their analytic system, eliminate risks of failure when possible, mitigation those risks, and report the “residual risks” to the laboratory, which should then help the laboratory customize its QC system. This customization could employ a variety of mechanisms for prevention and control, a primary one being the adjustment of the frequency of QC to monitor the remaining risks.
These practices have been developed with primary consideration of an analytical system operating in a laboratory environment, subject to the changing conditions, analyst, and environment. As such, they reflect the isolation of the analytical testing system in a laboratory, rather than considering the context of health care delivery as a whole. A different approach would seem appropriate to establish guidelines for QC frequency in the context of patient safety. In this case we need to look at the whole process wherein an erroneous lab test result may occur and can compromise patient safety.
If we wish to adopt levels of patient safety that are equivalent to the airline industry we would have to assume, based on the current state of laboratory instrumentation, that there may be many occasions when analytical systems will be seriously in error during the time period between QC determinations. Yuan et al. (2005) looked at microbiology errors and their clinical impact. Their results showed that out of 301 medically serious laboratory errors, only 32 resulted in a negative clinical impact. Thus the “medical system” reduced the potential effect of errors from 301 to 32, or by an order of magnitude. How does the medical system do this? The medical system here involves the laboratory and the physician who will act on the test result and therefore becomes an important intermediary in the patient testing process. It is our good fortune that the effect of an erroneous test result on patient safety is largely mitigated by the physician. When a physician sees an anomalous or even abnormal laboratory result, the usual response is to repeat the test. There will be occasions when the physician doesn’t perform a repeat test, e.g., there may be insufficient time to retest before a clinical decision is required or the abnormal result may fit the clinical picture. But, most of the time, physicians will confirm the abnormal test results. Because of the important role of the physician, we must ensure that the way we operate does not compromise the physician’s ability to mitigate the effects of errors. Thus, an important objective of a QC System should be to avoid circumstances where, upon repeat testing, the laboratory continues to provide erroneous results. If this occurs, the physician will obtain a confirmation of the previous erroneous result and act upon it (or not) to the detriment of the patient.
If we wish to adopt levels of patient safety that are equivalent to the airline industry we would have to assume, based on the current state of laboratory instrumentation, that there may be many occasions when analytical systems will be seriously in error during the time period between QC determinations.
Yuan et al. (2005) looked at microbiology errors and their clinical impact. Their results showed that out of 301 medically serious laboratory errors, only 32 resulted in a negative clinical impact. Thus the “medical system” reduced the potential effect of errors from 301 to 32, or by an order of magnitude. How does the medical system do this?
The medical system here involves the laboratory and the physician who will act on the test result and therefore becomes an important intermediary in the patient testing process. It is our good fortune that the effect of an erroneous test result on patient safety is largely mitigated by the physician. When a physician sees an anomalous or even abnormal laboratory result, the usual response is to repeat the test. There will be occasions when the physician doesn’t perform a repeat test, e.g., there may be insufficient time to retest before a clinical decision is required or the abnormal result may fit the clinical picture. But, most of the time, physicians will confirm the abnormal test results.
Because of the important role of the physician, we must ensure that the way we operate does not compromise the physician’s ability to mitigate the effects of errors. Thus, an important objective of a QC System should be to avoid circumstances where, upon repeat testing, the laboratory continues to provide erroneous results. If this occurs, the physician will obtain a confirmation of the previous erroneous result and act upon it (or not) to the detriment of the patient.
The physician’s “test repeat cycle” can provide an objective basis for determining QC frequency. As a minimum, the QC cycle time would need to be less than the test-repeat cycle time. While there has been little formal study of the phenomenon of repeat testing, van Walraven (2003) has looked at this from a health economics perspective and shown that physicians repeated tests more frequently and rapidly when the test result was abnormal. Also the repeat rate and rapidity decreased as the acuity of the situation decreased. In this context, different test-repeat cycles might be expected and estimated in different clinical settings. For an outpatient or doctor’s office service, one can estimate this cycle time to be as long as two or even three days. For a non-acute hospital setting, the test repeat cycle may be perhaps 4 hours. For an intensive care setting, the cycle time may be as short as 1 hour for chemistry tests and 30 minutes for blood gases. Although there may not be many published studies that document the test repeat cycle time, it should be possible to estimate the repeat cycle time characteristics in most laboratories. With that information in hand, the laboratory could then plan the frequency of QC to ensure that repeat test results will generally be confirmed after new controls have been analyzed. Meeting this QC objective may impose increased QC activity in some settings, particularly those hospital laboratories that operate at the minimum frequency required by the CLIA regulations. For many physician offices, daily QC should be adequate, but there should be concern about reducing the frequency to once per week. Few medical settings would seem to justify such a low frequency as once per month. There may be other important factors that need to be considered along with the test-repeat cycle. For example, reference laboratories certainty must consider the economics of analyzing large numbers of patient specimens and their own financial consequences if their analytic systems were to produce large numbers of defective results. Likewise in the laboratory, the translation of laboratory errors into patient morbidity and mortality may also illustrate the financial consequences evident in the healthcare setting in which the laboratory operates. Nonetheless, the important role of the physician in the mitigation of risk can be identified right now and should influence the design of our QC systems.
The physician’s “test repeat cycle” can provide an objective basis for determining QC frequency. As a minimum, the QC cycle time would need to be less than the test-repeat cycle time. While there has been little formal study of the phenomenon of repeat testing, van Walraven (2003) has looked at this from a health economics perspective and shown that physicians repeated tests more frequently and rapidly when the test result was abnormal. Also the repeat rate and rapidity decreased as the acuity of the situation decreased. In this context, different test-repeat cycles might be expected and estimated in different clinical settings.
Although there may not be many published studies that document the test repeat cycle time, it should be possible to estimate the repeat cycle time characteristics in most laboratories. With that information in hand, the laboratory could then plan the frequency of QC to ensure that repeat test results will generally be confirmed after new controls have been analyzed.
Meeting this QC objective may impose increased QC activity in some settings, particularly those hospital laboratories that operate at the minimum frequency required by the CLIA regulations. For many physician offices, daily QC should be adequate, but there should be concern about reducing the frequency to once per week. Few medical settings would seem to justify such a low frequency as once per month.
There may be other important factors that need to be considered along with the test-repeat cycle. For example, reference laboratories certainty must consider the economics of analyzing large numbers of patient specimens and their own financial consequences if their analytic systems were to produce large numbers of defective results. Likewise in the laboratory, the translation of laboratory errors into patient morbidity and mortality may also illustrate the financial consequences evident in the healthcare setting in which the laboratory operates. Nonetheless, the important role of the physician in the mitigation of risk can be identified right now and should influence the design of our QC systems.
With patient safety in mind, the rationale for the frequency of QC becomes quite simple, in contrast to other approaches being developed and applied by manufacturers for assessing and mitigating the risk of failure of their analytic systems and by laboratories for mitigating the remaining “residual risks” for those changes that may occur under routine laboratory operations. In a sense, this is risk management too, but made simple by focusing on the physician and the patient, not the manufacturer and the instrument. The important point is that this approach would ensure that the physician continues to have the opportunity to reduce the risk that an erroneous laboratory result poses for the patient.
Shan Yuan,1 Michael L. Astion,1* Jeff Schapiro,1,2† and Ajit P. Limaye1,2 Clinical Impact Associated with Corrected Results in Clinical Microbiology Testing. J Clin Microbiol. 2005 May; 43(5): 2188–2193 Carl van Walraven1and Michael Raymond, Population-based Study of Repeat Laboratory Testing. Clinical Chemistry 49:121997–2005 (2003)
Kent Dooley obtained his Ph.D. in organic chemistry from UBC in 1976. He then did a post doctorate fellowship in inherited metabolic diseases at the BC Children’s Hospital. This was followed by a clinical chemistry fellowship at the University of Texas in Houston where he was subsequently appointed to the and to the position of Assistant Technical Director of Clinical Chemistry at Hermann Hospital and Assistant professor of pathology at the University of Texas Medical School - Houston. In 1981 he moved to the IWK Children’s Hospital in Halifax, Nova Scotia to become the Head of Clinical Biochemistry. In 1984 he became a Founding Fellow of the Canadian Academy of Clinical Biochemists. In 1990 he assumed to role of Director of Pathology and Laboratory Medicine at the IWK Grace Health Center. His primary clinical and research interests are in newborn screening, inherited metabolic diseases and pediatric and maternal-fetal clinical chemistry. From 1999 – 2007 he was the Chair of the Nova Scotia Newborn Screening Committee and from 2001 – 2003 President of the Canadian Society of Clinical Chemists. In 2007 he was awarded the Canadian Society of Clinical Chemists award for Outstanding Contribution to Clinical Chemistry. In 2007 he moved to a position at LifeLabs, a private reference laboratory in Victoria, BC. Dr Dooley continues as an Adjunct Associate Professor of Pathology at Dalhousie University.
Understanding Quality -- Jerry Ehrmeyer Approaches to Clinical Laboratory Utilization -- Art Eggert, Ph.D. Tips on managing the quality of immunoassays -- R. Neill Carey, Ph.D. Defect rates, quality and productivity -- Robert Burnett, Ph.D. What's New with CLIA'88, JCAHO & CAP -- Sharon Ehrmeyer, Ph.D. European Approaches to Analytical Goal-Setting -- Per Hyltoft Petersen QC - The Regulations -- Sharon Ehrmeyer, Ph.D. Total Quality Management for Medical Laboratories: A European point of view -- Dr. Pharm. J.C. Libeer, Ph.D., Ph.D. MV - The Regulations -- Sharon Ehrmeyer, Ph.D. Biological variation data for setting quality specifications -- Callum G. Fraser, Ph.D. QC Validation in Veterinary Laboratories -- Kathleen P. Freeman, DVM, MS, Ph.D. Report of the Norwegian EQA Validator Workshop -- Dr. Heidi Steensland Trueness and Uncertainty -- Dr Xavier Fuentes Arderiu Ph.D. Pharm.D. Good Laboratory Practice versus CLIA -- Janine Denis Cook, Ph.D. Biologic Variation and desirable specifications for quality control --Dr. Carmen Ricos CLS: Victims of our own Success?-- Diana Mass M.A., CLS(NCA) Biologic Variation: Principles & Practice --Callum Fraser Ph.D. Pre-,Post- & Analytical Errors -- David Plaut Biological Database Update -- Dr. Carmen Ricos et al Quality Requirements Update 2002 -- Sharon Ehrmeyer, Ph.D. Are "Scientific Statements" the Scientific Truth? by Callum G. Fraser, Ph.D.Biologic Variation and desirable specifications for quality control 2006 --Dr. Carmen Ricos A multisite validation that selected "Westgard Rules" is efficient and cost-effective -- David Plaut et al. The Quality Goal Index - An Approach with Sigma Metrics - David Parry, Ph.D.