Probability Of False Rejection Research Articles

The diagnostic accuracy of quality control rules

Internal quality control in clinical chemistry laboratories are based on analyzing samples of stable control materials among the patient samples. The control results are interpreted by using quality control rules that usually are designed to detect systematic errors. The best rules have a high probability of error detection (Ped), i.e. to detect the maximal allowable (critical) systematic error and a low probability of false rejection (Pfr, false alarm). In this work we show that quality control rules can be represented by points on a ROC curve which appears when Ped is plotted against Pfr and only the control limit is varied. Further, we introduce a new method for choosing the optimal control limit, analogous to choosing the optimal operating point on the ROC curve of a diagnostic test. This decision needs knowledge of the pretest probability of a critical systematic error, the benefit of detecting it when it occurs and the cost of false alarm. The ROC curve analysis showed that if rules based on N = 2 are used, mean rules outperform Westgard rules because the ROC curve of the mean rules was lying above the ROC curves of the Westgard rules. A mean rule also had a lower maximum expected increase in the number of unacceptable patient results reported during the presence of an out-of-control error condition (Max E(NUF)) than comparable Westgard rules.

The importance of quality control validation and relationships with total error quality goals and bias in the interpretation of laboratory results.

The objective of a quality system is to provide accurate and reliable results for clinical decision-making. One part of this is Quality Control (QC) validation. QC validation is not routinely applied in veterinary laboratories. This leads to the inappropriate usage of random QC rules without knowing the Probability of error detection (Ped ) and Probability of false rejection (Pfr ) of a method. In this paper, we will discuss why QC validation is important, when it should be undertaken, why QC validation is done, and why it is not commonly done. We will present the role of total analytical error (TEa) in the QC validation process and the challenges when a consensus TEa has not been published. Finally, we will also discuss the possibilities of 'gray zone' determinations and mention the effects of bias on the quality of results. Reasons for the low prevalence of performing QC validation may include (a) lack of familiarity with the concept, (b) lack of time and resources needed to conduct QC validation, and (c) lack of TEa goal for some measurands. If no TEa is available, the user may elect to use a 'reverse approach' to QC validation. This uses the CV and bias generated from the evaluation of QC measurements, specifying Ped , Pfr , and N (number of QC measurements/run). This identifies the lowest total error that can be controlled under these defined conditions, thus enabling the laboratory to have an estimate of the 'gray zone' associated with results generated with a specific assay.

Patient-based pre-classified real-time quality control (PCRTQC)

BackgroundPatient-based real-time quality control (PBRTQC) has gained increasing attention in clinical laboratory management. Although its valuable characteristics complement traditional quality control measures, its performance and practical application have faced scrutiny. In this study, patient-based pre-classified real-time quality control (PCRTQC), an extended approach was devised to enhance real-time quality control protocols. MethodsPCRTQC distinguishes itself by incorporating an additional patient pre-classification step utilising the OPTICS algorithm, thus addressing interference from diverse patient types. The complete set of patient test results obtained from a clinical chemistry analyser at The First Hospital of China Medical University in 2021 was utilised. Constant error (CE) and proportional error (PE) were introduced as analytical errors. Four analytes were selected to evaluate the PCRTQC, measuring probability for false rejection (Pfr) and the average number of patient samples until error detection (ANPed). Relevant error detection charts were generated. ResultsThe PCRTQC outperformed regression-adjusted real-time quality control (RARTQC) based on the ANPed by approximately 50% for both the CE and PE, compared to the RARTQC, particularly for the total allowable error threshold. ConclusionThe pre-classification step effectively reduced inter-individual variation and improved data preprocessing, filtering, and modelling. The PCRTQC is a robust framework for real-time quality control research.

Repeat patient testing-quality control with canine samples shows promise as an alternative to commercial quality control material for a network of four Sysmex XT-2000iV hematology analyzers.

Repeat patient testing-quality control (RPT-QC) uses retained patient samples as an alternative to commercial quality control material (QCM). We elected to calculate and validate RPT-QC limits for red blood cell count (RBC), hemoglobin (HBG), hematocrit (HCT), and white blood cell count (WBC). (1) To validate RPT-QC across a network of four harmonized Sysmex XT-2000iV hematology analyzers and determine the total error that can be controlled with RPT-QC. (2) To generate quality control (QC) limits using the standard deviation (SD) of the duplicate measurement differences and determine a suitable simple QC rule with a probability of error detection >0.85 and probability of false rejection <0.05. (3) Monitor RPT-QC using sigma metrics as a performance indicator and (4) to challenge RPT-QC to ensure acceptable sensitivity. Fresh adult canine EDTA samples with results within reference intervals were selected and run again on days 2, 3, and 4. QC limits were generated from the SD of the duplicate measurement differences. The QC limits were challenged using interventions designed to promote unstable system performance. The total error detectable by RPT-QC was determined using EZRULES 3 software. In all, 20-40 data points were needed for RPT-QC calculations and validated using 20 additional data points. The calculated limits differed among the network of analyzers. The total error that could be controlled was the same or better than that of the manufacturer's commercially available quality control material using the same analyzer for all measurands except hematocrit, which required a higher total error goal than that proposed by ASVCP guidelines to achieve an acceptable probability of error detection. The challenges designed to mimic unstable system performance were successfully detected as out-of-control QC. The challenges for RPT-QC resulted in acceptable detection of potential unstable system performance. This initial study demonstrates that RPT-QC limits differ among the network of Sysmex XT-2000iV analyzers, indicating a requirement to customize for the individual analyzer and laboratory conditions. RPT-QC could achieve ASVCP total allowable error goals for RBC, HGB, and WBC, but not for HCT. Sigma metrics were consistently >5.5 for RBC, HGB, and WBC, but not for HCT.

Setting up an own laboratory performance-based internal quality control plan - a model for complete blood count

Abstract Quality Control (QC) in Romania is regulated by the Order of the Minister of Health no. 1608/2022 that modifies the previous Order 1301/2007. The new version of the Order introduces a more scientific approach by requesting the laboratories to assess test performance and then elaborate an appropriate internal QC plan. The aim of this study was to demonstrate how to design a QC plan for complete blood count (CBC) in an Emergency Laboratory with continuous activity, in order to comply with the new Order 1608/2022. QC data obtained over a three-month period (April-June 2022) from the Sysmex XN-1000 instrument of the Emergency Laboratory of the County Emergency Clinical Hospital of Târgu Mureș were included. In order to establish an appropriate QC plan, two models were applied and the following parameters were calculated: the number of daily QC runs (N), the probability of false rejection (Pfr), the QC frequency (run size), and the required QC rules. White blood cells achieved high performance, while Hematocrit performance was poor. Different levels of performance were achieved for Platelets. We emphasize that, when all parameters are measured on the same instrument, QC frequency and Pfr should be adjusted in order to develop a QC plan that “fits” all the parameters of the CBC as a whole. In our Emergency Laboratory, the calculated QC plan for CBC is N=2, Pfr=0.03, multi-rule 1:3s/2:2s/R:4s, and a run size of 95 samples which is approximately the same as the number of CBCs performed during one 12-hour shift.

Total observed error, total allowable error, and QC rules for canine serum and urine cortisol achievable with the Immulite 2000 Xpi cortisol immunoassay.

Determining a simple quality control (QC) rule for daily performance monitoring depends on the desired total allowable error (TEa) for the measurand. When no consensus TEa exists, the classical approach of QC rule validation cannot be used. Using the results of previous canine serum and urine cortisol validation studies on the Immulite 2000 Xpi, we applied a reverse engineering approach to QC rule determination, arbitrarily imposing sigma = 5, and determining the resulting TEa for the QC material (QCM; TEaQCM) and the resulting probability of error detection (Ped) for each QC rule. For the simple QC rule 12.5S with Ped = 0.96 and probability of false rejection (Pfr) = 0.03, the associated TEaQCM were 20% and 35% for serum and 28% and 24% for urine QCM1 and QCM2. If these levels of TEaQCM are acceptable for interpretation of patient sample results, then users can internally validate the 12.5S QC rule, provided that their QCM CVs and biases are similar to ours. Otherwise, more stringent QC rules can be validated by using a lower sigma to lower the TEaQCM. With spiked samples (relevant cortisol concentrations in the veterinary patient matrix) at 38.6 and 552 nmol/L of cortisol, TEaQCM at sigma = 5 were much higher (54% and 40% for serum; 90.3% and 42.8% for urine). Spiked samples generate TEa that is probably too high to be suitable for daily QC monitoring; however, it is crucial to verify spiked sample observed total error (TEo; 26% and 18% for serum, 60% and 30% for urine) < TEaQCM, and to use spiked sample TEo for patient result interpretation. In the absence of consensus TEa for cortisol in dogs, we suggest the use of a 12.5S rule, provided that users accept the associated level of TEaQCM also as clinical TEa for results interpretation.

Performance evaluation of internal quality control rules, EWMA, CUSUM, and the novel machine learning model

Abstract Objectives The present study set out to build a machine learning model to incorporate conventional quality control (QC) rules, exponentially weighted moving average (EWMA), and cumulative sum (CUSUM) with random forest (RF) algorithm to achieve better performance and to evaluate the performances the models using computer simulation to aid laboratory professionals in QC procedure planning. Methods Conventional QC rules, EWMA, CUSUM, and RF models were implemented on the simulation data using an in-house algorithm. The models’ performances were evaluated on 170,000 simulated QC results using outcome metrics, including the probability of error detection (Ped), probability of false rejection (Pfr), average run length (ARL), and power graph. Results The highest Pfr (0.0404) belonged to the 1–2s rule. The 1–3s rule could not detect errors with a 0.9 Ped up to 4 SD of systematic error. The random forest model had the highest Ped for systematic errors lower than 1 SD. However, ARLs of the model require the combined utility of the RF model with conventional QC rules having lower ARLs or more than one QC measurement is required. Conclusions The RF model presented in this study showed acceptable Ped for most degrees of systematic error. The outcome metrics established in this study will help laboratory professionals planning internal QC.

Prediction of Clinically Significant Cancer Using Radiomics Features of Pre-Biopsy of Multiparametric MRI in Men Suspected of Prostate Cancer.

Simple SummaryRadiomics is the field of computer-based medical image analysis that incorporates various radiological imaging features, such as texture and shape parameters, from scans to derive algorithms. These mathematical algorithms have the potential to predict the biological characteristics of disease. In this study, we obtained quantitative imaging texture features of pre-biopsy multiparametric MRI of men suspected of prostate cancer and extracted from the T2WI and ADC images focusing on gray-level co-occurrence matrices (GLCM). These were correlated with the Gleason score of the histopathology of radical prostatectomy specimen, including the prediction of clinically significant prostate cancer. The knowledge gained through this prospective protocol-based study should facilitate establishing that GLCM texture features alone can be used as a biomarker for predicting the presence of clinically significant PCa.Background: Texture features based on the spatial relationship of pixels, known as the gray-level co-occurrence matrix (GLCM), may play an important role in providing the accurate classification of suspected prostate cancer. The purpose of this study was to use quantitative imaging parameters of pre-biopsy multiparametric magnetic resonance imaging (mpMRI) for the prediction of clinically significant prostate cancer. Methods: This was a prospective study, recruiting 200 men suspected of having prostate cancer. Participants were imaged using a protocol-based 3T MRI in the pre-biopsy setting. Radiomics parameters were extracted from the T2WI and ADC texture features of the gray-level co-occurrence matrix were delineated from the region of interest. Radical prostatectomy histopathology was used as a reference standard. A Kruskal–Wallis test was applied first to identify the significant radiomic features between the three groups of Gleason scores (i.e., G1, G2 and G3). Subsequently, the Holm–Bonferroni method was applied to correct and control the probability of false rejections. We compared the probability of correctly predicting significant prostate cancer between the explanatory GLCM radiomic features, PIRADS and PSAD, using the area under the receiver operation characteristic curves. Results: We identified the significant difference in radiomic features between the three groups of Gleason scores. In total, 12 features out of 22 radiomics features correlated with the Gleason groups. Our model demonstrated excellent discriminative ability (C-statistic = 0.901, 95%CI 0.859–0.943). When comparing the probability of correctly predicting significant prostate cancer between explanatory GLCM radiomic features (Sum Variance T2WI, Sum Entropy T2WI, Difference Variance T2WI, Entropy ADC and Difference Variance ADC), PSAD and PIRADS via area under the ROC curve, radiomic features were 35.0% and 34.4% more successful than PIRADS and PSAD, respectively, in correctly predicting significant prostate cancer in our patients (p < 0.001). The Sum Entropy T2WI score had the greatest impact followed by the Sum Variance T2WI. Conclusion: Quantitative GLCM texture analyses of pre-biopsy MRI has the potential to be used as a non-invasive imaging technique to predict clinically significant cancer in men suspected of having prostate cancer.

Comparison of normalized and non-normalized block caving comminution models

In block cave mining, rock fragmentation is a key parameter that influences the production level design and mine planning. Fragmentation occurs mainly by natural breakage during the caving process and in the draw column. The breakage that occurs within the column is known as secondary fragmentation. Secondary fragmentation has been successfully described using the block caving comminution model, which replicates the fragmentation mechanics between particles under drawn and vertical loads in a draw column. This model is based on a kinetic and population balance approach, in which non-normalized and normalized assumptions can be used depending on material and comminution system behaviour. In this paper, the non-normalized and normalized approaches are applied and compared to laboratory data to determine which assumption should be used for secondary breakage in block caving. Approaches are compared graphically, with the mean square error and the Fisher test with a false-rejection probability of 0.05. Based on a statistical analysis, the results show that the normalized model can be applied to all the rock types tested.

Application of the sigma metrics to evaluate the analytical performance of cystatin C and design a quality control strategy.

Sigma metrics are commonly used to evaluate laboratory management. In this study, we aimed to evaluate the analytical performance of cystatin C using sigma metrics and to develop an individualized quality control scheme for cystatin C concentrations. Bias was calculated based on the samples used for the external quality assessment. The coefficient of variation was calculated using six months of internal quality control measurements at two levels, and desirable specification derived from biological variation was used as the quality goal. The sigma value for cystatin C was calculated using the above data. The internal quality control scheme and improvement measures were formulated according to the Westgard sigma standards for batch size and quality goal index. The sigma values for cystatin C, for quality control levels 1 and 2, were 3.04 and 4.95, respectively. The 13s/22s/R4s/41s/6x multirules (n = 6, R = 1), with a batch size of 45 patient samples, were selected as the internal quality control schemes for cystatin C. With different concentrations of cystatin C, the power function graph showed a probability for error detection of 94% and 100% and a probability for false rejection of 4% and 2%, respectively. According to the quality goal index of cystatin C, its precision needs to be improved. With a 'desirable' biological variation of 6.50%, the Westgard rule 13s/22s/R4s/41s/6x (n = 6, R = 1, batch size of 45) with high efficacy for determining the detection error is recommended for individualized quality control schemes of cystatin C.

Repeat patient testing-based quality control shows promise for use in veterinary biochemistry testing.

Repeat patient testing-based quality control (RPT-QC) is a form of statistical QC and an alternative to commercial quality control materials (QCM). This study investigated the suitability of canine heparinized plasma for use in RPT-QC and assessed the predicted performance of RPT-QC for the detection of analytical error in chemistry testing. The stability of canine plasma for RPT-QC was investigated via storage at two temperatures for three or six time points. Storage data were analyzed using repeated measures ANOVA and by comparing results for stored specimens to baseline data using predetermined criteria. To generate RPT-QC limit-setting and -validation data, leftover plasma was prospectively measured. Once control limits were established, these were challenged by measuring specimens for which the repeat aliquot had been manipulated to mimic analytical error. Finally, the predicted performance of RPT-QC and QCM-QC with four control rules was investigated using Westgard's EZ Rules 3. Refrigerated storage of canine plasma for 7days allowed mild changes facilitating RPT-QC. RPT-QC limits for 12 of 17 common measurands were validated. Validated limits successfully flagged differences from manipulated specimen pairs as "error." The predicted performance of RPT-QC for analytical error detection (represented by smallest achievable allowable total error, given a probability of error detection≥85% and a probability of false rejection≤5%) for four common control rules is comparable to that of QCM-QC. This study provides evidence that RPT-QC using canine heparinized plasma refrigerated for 7days can be used with simple control rules and low numbers of control materials, suggesting RPT-QC is applicable to both reference and in-clinic laboratory settings.

Research progress on internal statistical quality control procedures for quantitative tests in clinical laboratories

The laboratory medicine is aimed to support clinical decisions and patient health by providing accurate results. The internal statistical quality control (SQC) can help laboratories detecting the instability of the analytical system and preventing laboratories from reporting the patient results with medically important errors, so it is essential to ensure the quality of testing results and patient safety. The traditional methods of designing SQC strategy are based on the probability of error detection (Ped) and the probability of false rejection (Pfr). With the introduction of risk management concepts, the design of SQC strategy began to be based on the patient risk parameter [MaxE(Nuf)] proposed by Parvin, which means the maximum increase in the number of unacceptable patient results reported compared to the in-control condition during the existence of an undetected out-of control error condition. MaxE(Nuf) is related to the SQC frequency and patient risk, which is very essential for optimizing the SQC frequency and designing a risk-based SQC strategy. Key words: Quality control; Risk management; Diagnostic tests, routine; Total quality management

Evaluating the analytical quality control of urinary albumin measurements using sigma metrics

BackgroundThere is no worldwide recognized reference system and standard for urinary albumin measurement until now, so the analytical quality from different laboratories has always varied. In this study, we aimed to evaluate the analytical performance of a urinary albumin assay system using Sigma-metric, and thereby choose a suitable control rule to guarantee the analytical quality of the assays. MethodTwo levels of diluted reference material (ERM-DA47OK/IFCC) were used to calculate the biases, the coefficient of variation (CV) were calculated from six months of internal quality control measurements at two levels, and the external quality assessment standard of China for urinary albumin (30%) was used as the total allowable error(TEa). ResultsThe Sigma values for quality control levels 1 and 2 were 4.28 and 6.14, leading to recommended Westgard rules of 13s/22s/R4s/41s (N = 2, R = 2) and 13s(N = 2, R = 1), respectively. Westgard rule 13s/22s/R4s/41s(N = 2, R = 2) was selected for the quality control of the urinary albumin measurements, and with it, the power function graph showed a high efficacy for determining the detection errors with a probability of false rejection of 1.004% and a probability of error detection of 98.80%. ConclusionWith a TEa of 30% recommended by the external quality assessment standard of China, Westgard rule 13s/22s/R4s/41s(N = 2, R = 2) with a high efficacy for determining the detection error is recommended for the quality control of urinary albumin measurements.

Authentication of Keys Distributed by the Diffie-Hellman Method for Mobile Devices based on Authentication Codes and Magnetometric Data

The wide distribution of mobile wireless devices (smartphones, tablets, etc.) warrants cryptographic protection of information transmitted by these devices, which requires supplying these devices with keys and providing their authentication. Recently, research on key authentication methods within scenarios of pairing mobile devices, has been increasingly relevant. In these conditions, mobile devices are located close to each other, up to the physical contact, and an additional key sharing authenticated channel protected from interception is established for purposes of key authentication.&#x0D; The analysis of additional channel versions: visual, acoustic, vibration, tactile, and magnetometric shows advantages of the latter one in terms of speed and reliability of authentication, as well as easer se of this channel compared to other methods. A magnetometer channel forms if there are magnetometers in mobile devices that measure the Earth’s magnetic field. Random fluctuations of the magnetic field at the measurement point allow to generate in a pair of mobile devices random sequences that coincide with a high probability and which can be used to authenticate session keys.&#x0D; The “Magparing” protocol of mobile devices keys authentication, which are distributed by Diffie-Hellman method is studied. his protocol is based on the magnetometer data measuring. We demonstrate that the protocol is succeptible to “man-in-the-middle” attack. This fact is confirmed also by software AVISPA simulation.&#x0D; A new key authentication method based on the use of authenticating code (A-code) is proposed. Authenticators for Diffie-Hellman values are generated based on A-code blocks and random sequences obtained by the reading of magnetometer’s values. An approach for A-code parameters optimization, that conforms to the requirements for authentication key length and both to the probabilities of false key removal and false key deception, is investigated. We present an example of code parameters for key authentication with the length of 256 bits, providing the probability of a false key agreement of not greater than 10-9 and a probability of false rejection of a true key equal to –10–6, which shows viability of the proposed authentication method.

Estimation of the random error of binary tests using adaptive polynomials

We investigate how to estimate error probabilities for binary measurements and tests when a gold standard is not available. Recent studies have shown that, without a gold standard, the widely used false-acceptance probability () and false-rejection probability () are difficult to estimate. We show by mathematical analysis that this problem is inherent to the unavailability of a gold standard. Instead, the proposed method determines the random components of the error probabilities: the inconsistent acceptance and rejection probabilities and . The method is efficient and robust as the number of model parameters is not predetermined but depends on the goodness of fit.

New methods of testing nonlinear hypothesis using iterative NLLS estimator

This research paper discusses the method of testing nonlinear hypothesis using iterative Nonlinear Least Squares (NLLS) estimator. Takeshi Amemiya [1] explained this method. However in the present research paper, a modified Wald test statistic due to Engle, Robert [6] is proposed to test the nonlinear hypothesis using iterative NLLS estimator. An alternative method for testing nonlinear hypothesis using iterative NLLS estimator based on nonlinear hypothesis using iterative NLLS estimator based on nonlinear studentized residuals has been proposed. In this research article an innovative method of testing nonlinear hypothesis using iterative restricted NLLS estimator is derived. Pesaran and Deaton [10] explained the methods of testing nonlinear hypothesis. This paper uses asymptotic properties of nonlinear least squares estimator proposed by Jenrich [8]. The main purpose of this paper is to provide very innovative methods of testing nonlinear hypothesis using iterative NLLS estimator, iterative NLLS estimator based on nonlinear studentized residuals and iterative restricted NLLS estimator. Eakambaram et al. [12] discussed least absolute deviation estimations versus nonlinear regression model with heteroscedastic errors and also they studied the problem of heteroscedasticity with reference to nonlinear regression models with suitable illustration. William Grene [13] examined the interaction effect in nonlinear models disused by Ai and Norton [14] and suggested ways to examine the effects that do not involve statistical testing. Peter [15] provided guidelines for identifying composite hypothesis and addressing the probability of false rejection for multiple hypotheses.

Comparison of 3 options for choosing control limits in biochemistry testing

The purpose of statistical quality control (QC) is to provide peace of mind with regard to the production of results that are suitable for analytically sound clinical interpretation and making reliable decisions about patient diagnosis, monitoring, and prognosis. In this study, we compared 3 options for choosing control limits for biochemistry testing. They focus on the probability of error detection (Ped) and probability of false rejection (Pfr) achievable for a veterinary biochemical analyzer using the following 3 combinations: the quality control material (QCM) manufacturer's acceptable ranges; a standard 12s rule customized for the instrument's observed performance; and candidate rules selected for the instrument's observed performance using a computerized program (EZrules). For assessing customized QC, we used mean, SD, CV, bias, total error, and sigma metrics calculated from 3months of control measurements on a laboratory biochemical analyzer, for 24 commonly used analytes, on 2 QCM levels. Given the desirable combination of high Ped (>90%) and low Pfr (≤5%), the candidate rules selected by the computerized program-related EZrules provided the best performance combinations. The present work shows acceptable QC performance basing the QC on customization of the acceptable ranges of results from the achievable performance of an individual instrument. The QC performance is maximized by the application of candidate rules based on customized ranges obtained from a computerized QC tool, providing the ability to achieve the highest Ped and acceptably low Pfr values.

Investigation and analysis of internal quality control on clinical chemistry, clinical immunology and clinical hematology of mutual recognition laboratories in 142 medical institutions in Beijing

Objective To investigate the internal quality control (IQC) on clinical chemistry, clinical immunology and clinical hematology in mutual recognition laboratories in medical institutions in Beijing. Methods By means of questionnaire survey and on-site investigation, fresh frozen serum and whole blood samples with assigned values by reference method were measured to investigate the status of IQC on clinical chemistry, clinical immunology and clinical hematology in 142 mutual recognition laboratories in medical institutions of Beijing, and results were analyzed. Results 142 copies of questionnaireson clinical chemistry, clinical immunology and clinical hematology were send out and 120, 97, and 101 laboratories returned the questionnaires respectively .The information feedback rate was 84.5%, 68.3% and 71.1% respectively.All the questionnaires were effective. Questionnaires survey results showed that more than 50% laboratories set up quality control goals and the most of the goals were probability for error detection (Ped) 95%, probability for false rejection(Pfr)5%; About 70% laboratories usecd the same quality control plan for different tests ; The most frequently used quality control rules are 12s/13s/22s. On-site investigation showed that , take the results of clinical chemistry for example, based on the desirable biological variation and WS/T 403-2012 , most of the tests can't meet the quality control goalsunder the existing quality control condition. Conclusion Clinical laboratories should consider their actual situations, assess their own quality levels that they can reach , set reasonable quality standards for themselves , and make appropriate individualized quality control plan. (Chin J Lab Med, 2016, 39: 922-929) Key words: Clinical chemistry tests; Immunologic tests; Hematologic tests; Quality control

The statistical evaluation of binary tests without gold standard: Robustness of latent variable approaches

Binary tests classify items into two categories such as reject/accept or positive/negative. Such tests are usually evaluated in terms of their misclassification probabilities FAP (false acceptance probability) and FRP (false rejection probability). A common complication arises when there is no gold standard or reference standard. Various methods based on latent variable modelling have been proposed for this situation. We present the results of a simulation study in which these methods are tried in a range of scenarios, to study how robust they are to departures from the assumptions on which they are based. The study convincingly shows that in general, the ambition of estimating FAP and FRP without gold standard is unattainable, since all methods easily derail when assumptions are not precisely met. The study also shows that the random components of the FAP and FRP can be reliably estimated by a straightforward modification of one of the tested methods.

Monitoring fuel quality: a case study for quinizarin marker content of unleaded petrol marketed in Greece

Fuel adulteration and cross-contamination lead to low-quality fuel products, which may cause increased environmental pollution, loss of taxes and engine problems. An establishment of a quality monitoring mechanism based on laboratory measurements may reveal problematic areas of the fuel supply chain. For the purposes of this work, 97 unleaded petrol samples were measured in order to quantify mass concentration of quinizarin, a substance used in Greece to easily mark the presence of 95 Research Octane Number unleaded petrol in other types of automotive fuels. The samples were obtained from petroleum retail stations selling different brands of fuels and located in different geographic regions of Greece. Statistical analysis of the results revealed quinizarin mass concentrations below the 3 mg L−1 legislation specification limit and significant differences between brands and geographic regions, which may attributed to the structure of the fuel supply chain in Greece in combination with quinizarin properties and way of handling. Moreover, certain approaches were used for the calculation of decision limits for assessing compliance or non-compliance. These approaches take measurement reproducibility or estimated in-house uncertainty into account, in order to minimize the probability of false rejection.