Lot-to-lot variation and verification.

Lot-to-lot verification is an integral component for monitoring the long-term stability of a measurement procedure. The practice is challenged by the resource requirements as well as uncertainty surrounding experimental design and statistical analysis that is optimal for individual laboratories, although guidance is becoming increasingly available. Collaborative verification efforts as well as application of patient-based monitoring are likely to further improve identification of any differences in performance in a relatively timely manner. Appropriate follow up actions of failed lot-to-lot verification is required and must balance potential disruptions to clinical services provided by the laboratory. Manufacturers need to increase transparency surrounding release criteria and work closer with laboratory professionals to ensure acceptable reagent lots are released to end users. A tripartite collaboration between regulatory bodies, manufacturers, and laboratory medicine professional bodies is key to developing a balanced system where regulatory, manufacturing, and clinical requirements of laboratory testing are met, to minimize differences between reagent lots and ensure patient safety. Clinical Chemistry and Laboratory Medicine has served as a fertile platform for advancing the discussion and practice of lot-to-lot verification in the past 60 years and will continue to be an advocate of this important topic for many more years to come.

Moving sum of number of positive patient result as a quality control tool.

BACKGROUND: Recently, the total prostate-specific antigen (PSA) assay used in a laboratory had a positive bias of 0.03 µg/L, which went undetected. Consequently, a number of post-prostatectomy patients with previously undetectable PSA concentrations (defined as <0.03 µg/L in that laboratory) were being reported as having detectable PSA, which suggested poorer prognosis according to clinical guidelines. METHODS: Through numerical simulations, we explored (1) how a small bias may evade the detection of routine quality control (QC) procedures with specific reference to the concentration of the QC material, (2) whether the use of 'average of normals' approach may detect such a small bias, and (3) describe the use of moving sum of number of patient results with detectable PSA as an adjunct QC procedure. RESULTS: The lowest QC level (0.86 µg/L) available from a commercial kit had poor probability (<10%) of a bias of 0.03 µg/L regardless of QC rule (i.e. 1:2S, 2:2S, 1:3S, 4:1S) used. The average number of patient results affected before error detection (ANPed) was high when using the average of normals approach due to the relatively wide control limits. By contrast, the ANPed was significantly lower for the moving sum of number of patient results with a detectable PSA approach. CONCLUSIONS: Laboratory practitioners should ensure their QC strategy can detect small but critical bias, and may require supplementation of ultra-low QC levels that are not covered by commercial kits with in-house preparations. The use of moving sum of number of patient results with a detectable result is a helpful adjunct QC tool.

Detecting long-term drift in reagent lots.

BACKGROUND: Between-reagent lot verification is a routine laboratory exercise in which a set of samples is tested in parallel with an existing reagent lot and a candidate reagent lot (before the candidate lot is committed to test patient samples). The exercise aims to verify and maintain consistency in the analytical performance of a test. We examined the limitations of a routine between-reagent lot verification procedure in detecting long-term analytical drift and looked for a more sensitive alternative. METHOD: Via numerical simulations, we examined the statistical power of the current regression-based (weighted Deming regression) procedure for between-reagent lot verification in detecting proportional bias and constant bias. An alternative procedure applying the Student t-test to separately examine cumulative regression slopes and intercepts across multiple reagent lots was proposed and evaluated by numerical simulations. RESULTS: The regression-based procedure had poor statistical power in detecting proportional bias and constant bias when small numbers of samples were used in each between-reagent lot verification exercise. Furthermore, the method failed to detect long-term drifts in analytical performance. The proposed approach based on the Student t-test can detect long-term (cumulative) drifts in regression slopes and intercepts. This method detected a mild downward drift in the serum sodium assay in our hospital that was missed by routine between-reagent lot verification. CONCLUSIONS: The proposed method objectively and systematically detects long-term proportional and constant bias separately. However, the statistical power of this procedure remains unsatisfactory when used with small sample sizes. Sharing of information between laboratories may provide sufficient statistical power to detect clinically important analytical shifts.

Toward Sustainable Environmental Quality: Priority Research Questions for Asia.

Environmental and human health challenges are pronounced in Asia, an exceptionally diverse and complex region where influences of global megatrends are extensive and numerous stresses to environmental quality exist. Identifying priorities necessary to engage grand challenges can be facilitated through horizon scanning exercises, and to this end we identified and examined 23 priority research questions needed to advance toward more sustainable environmental quality in Asia, as part of the Global Horizon Scanning Project. Advances in environmental toxicology, environmental chemistry, biological monitoring, and risk-assessment methodologies are necessary to address the adverse impacts of environmental stressors on ecosystem services and biodiversity, with Asia being home to numerous biodiversity hotspots. Intersections of the food-energy-water nexus are profound in Asia; innovative and aggressive technologies are necessary to provide clean water, ensure food safety, and stimulate energy efficiency, while improving ecological integrity and addressing legacy and emerging threats to public health and the environment, particularly with increased aquaculture production. Asia is the largest chemical-producing continent globally. Accordingly, sustainable and green chemistry and engineering present decided opportunities to stimulate innovation and realize a number of the United Nations Sustainable Development Goals. Engaging the priority research questions identified herein will require transdisciplinary coordination through existing and nontraditional partnerships within and among countries and sectors. Answering these questions will not be easy but is necessary to achieve more sustainable environmental quality in Asia. Environ Toxicol Chem 2020;39:1485-1505. © 2020 The Authors. Environmental Toxicology and Chemistry published by Wiley Periodicals LLC on behalf of SETAC.

Moving standard deviation and moving sum of outliers as quality tools for monitoring analytical precision.

INTRODUCTION: An increase in analytical imprecision (expressed as CVa) can introduce additional variability (i.e. noise) to the patient results, which poses a challenge to the optimal management of patients. Relatively little work has been done to address the need for continuous monitoring of analytical imprecision. METHODS: Through numerical simulations, we describe the use of moving standard deviation (movSD) and a recently described moving sum of outlier (movSO) patient results as means for detecting increased analytical imprecision, and compare their performances against internal quality control (QC) and the average of normal (AoN) approaches. RESULTS: The power of detecting an increase in CVa is suboptimal under routine internal QC procedures. The AoN technique almost always had the highest average number of patient results affected before error detection (ANPed), indicating that it had generally the worst capability for detecting an increased CVa. On the other hand, the movSD and movSO approaches were able to detect an increased CVa at significantly lower ANPed, particularly for measurands that displayed a relatively small ratio of biological variation to CVa. CONCLUSION: The movSD and movSO approaches are effective in detecting an increase in CVa for high-risk measurands with small biological variation. Their performance is relatively poor when the biological variation is large. However, the clinical risks of an increase in analytical imprecision is attenuated for these measurands as an increased analytical imprecision will only add marginally to the total variation and less likely to impact on the clinical care.

Missed detection of significant positive and negative shifts in gentamicin assay: implications for routine laboratory quality practices.

INTRODUCTION: A product recall was issued for the Roche/Hitachi Cobas Gentamicin II assays on 25th May 2016 in Australia, after a 15 - 20% positive analytical shift was discovered. Laboratories were advised to employ the Thermo Fisher Gentamicin assay as an alternative. Following the reintroduction of the revised assay on 12th September 2016, a second reagent recall was made on 20th March 2017 after the discovery of a 20% negative analytical shift due to erroneous instrument adjustment factor. MATERIALS AND METHODS: The practices of an index laboratory were examined to determine how the analytical shifts evaded detection by routine internal quality control (IQC) and external quality assurance (EQA) systems. The ability of the patient result-based approaches, including moving average (MovAvg) and moving sum of outliers (MovSO) approaches in detecting these shifts were examined. RESULTS: Internal quality control data of the index laboratory were acceptable prior to the product recall. The practice of adjusting IQC target following a change in assay method resulted in the missed negative shift when the revised Roche assay was reintroduced. While the EQA data of the Roche subgroup showed clear negative bias relative to other laboratory methods, the results were considered as possible 'matrix effect'. The MovAvg method detected the positive shift before the product recall. The MovSO did not detect the negative shift in the index laboratory but did so in another laboratory 5 days before the second product recall. CONCLUSIONS: There are gaps in current laboratory quality practices that leave room for analytical errors to evade detection.

Verification of out-of-control situations detected by "average of normal" approach.

OBJECTIVES: "Average of normal" (AoN) or "moving average" is increasingly used as an adjunct quality control tool in laboratory practice. Little guidance exists on how to verify if an out-of-control situation in the AoN chart is due to a shift in analytical performance, or underlying patient characteristics. DESIGN AND METHODS: Through simulation based on clinical data, we examined 1) the location of the last apparently stable period in the AoN control chart after an analytical shift, and 2) an approach to verify if the observed shift is related to an analytical shift by repeat testing of archived patient samples from the stable period for 21 common analytes. RESULTS: The number of blocks of results to look back for the stable period increased with the duration of the analytical shift, and was larger when smaller AoN block sizes were used. To verify an analytical shift, 3 archived samples from the analytically stable period should be retested. In particular, the process is deemed to have shifted if a difference of >2 analytical standard deviations (i.e. 1:2s rejection rule) between the original and retested results are observed in any of the 3 samples produced. The probability of Type-1 error (i.e., false rejection) and power (i.e., detecting true analytical shift) of this rule are <0.1 and >0.9, respectively. CONCLUSIONS: The use of appropriately archived patient samples to verify an apparent analytical shift is preferred to quality control materials. Nonetheless, the above findings may also apply to quality control materials, barring matrix effects.

Bayesian approach to guide termination of retrospective retesting after detection of a systematic quality control failure.

BACKGROUND: When a systematic error is detected in the analytical process, ideally, one seeks to retest only patient samples between the onset of the error and the time the error was detected. In practice, the onset of error is often unknown, and patient samples are retrospectively retested back to the last acceptable QC sample. This can be wasteful of reagents and operator time. METHODS: An alternative approach that is based on the expected number of spurious results is described to determine when retrospective retesting should terminate. Assuming each patient sample was independently measured by an analytical process with an underlying Gaussian distribution, a Bayesian model that takes into account the difference between the original and retest result of each patient sample was developed. RESULTS: We are able to significantly reduce the number of samples retested, while ensuring that the average number of spurious results observed under the proposed retesting procedure was similar to or only marginally higher than the baseline number of spurious results when the assay was in control. CONCLUSION: Patient samples measured after the systematic error have high probabilities of being retested under the proposed retesting procedure.