Accreditation and Quality Assurance最新文献

It is important for a testing laboratory to proactively identify risks so as to prevent or reduce the occurrence of failures in the laboratory activities and to ascertain the validity and reliability of test results. The ISO/IEC 17025:2017 standard requires the laboratory to identify specific risks such as those related to impartiality and laboratory activities. The standard also requires the laboratory to carry out actions to address the identified risks. Although the standard does not require the laboratory to follow a formal method of risk management or a documented risk management process, the laboratory should try its best to develop a documented system for efficient risk management. This article aims to provide practical insights and guidance for testing laboratories seeking to develop comprehensive risk management practices to fulfill the requirements of the ISO/IEC 17025:2017 standard. By embracing a robust risk management framework, the testing laboratories can protect the integrity of their results, ensure the safety of the staff, enhance the quality of their services, and promote compliance to the ISO/IEC 17025:2017 standard, ultimately fostering trust and confidence among stakeholders.

Proficiency testing (PT) programmes in mycotoxin analysis predominantly rely on the z-score for participant performance evaluation, which does not incorporate the laboratory's reported measurement uncertainty. In this study, 2,538 results from 15 mycotoxin PT rounds conducted between 2020 and 2025 were analysed using a binary classification framework comprising 12 probability categories (P1 to P12), based on the combined evaluation of z-scores, ζ-scores, and measurement uncertainty adequacy. The analysis revealed that a substantial proportion of results classified as satisfactory under z-score-only evaluation were reclassified as unsatisfactory when the ζ-score was applied, showing that z-score-only assessment may overlook results for which the observed deviation is inconsistent with the participant-reported measurement uncertainty. Measurement uncertainty underestimation was markedly more prevalent than overestimation, and the mathematically constrained P6 category and the mathematically excluded P8 category were not observed in the dataset. These findings support consideration of ζ-score evaluation alongside structured measurement uncertainty assessment in mycotoxin PT programmes, particularly in regulatory contexts where performance classifications may inform accreditation and regulatory oversight.

Artificial intelligence (AI) is increasingly being explored for laboratory testing, calibration, data management, and decision support. For laboratories operating within ISO/IEC 17025-oriented quality systems, however, AI adoption raises specific questions about validation, metrological traceability, measurement uncertainty, data integrity, impartiality, and human accountability. This study is presented as a structured systematic review with thematic synthesis and ISO/IEC 17025 clause mapping. Thirty peer-reviewed studies were included in the qualitative evidence synthesis, while metrological, accreditation, and AI governance documents were used as contextual sources for standards-oriented interpretation. The review shows that most available evidence comes from laboratory-adjacent settings such as clinical diagnostics, environmental sensor calibration, laboratory information systems, and AI governance, rather than from laboratories explicitly accredited to ISO/IEC 17025. The strongest technical evidence relates to analytical pattern recognition and sensor calibration, whereas direct evidence for accreditation governance, uncertainty treatment, and routine operational validation remains limited. The principal gap is the lack of standardized validation and governance frameworks for deploying AI in testing and calibration environments. The proposed framework therefore positions AI as a controlled support tool rather than an autonomous decision-maker and links each AI use case to documented validation, GUM-based uncertainty evaluation where applicable, traceability evidence, change control, audit trails, and authorized signatory review. The paper contributes a clause-by-clause mapping and an operational lifecycle for AI adoption in ISO/IEC 17025-oriented laboratory systems.

Participation in proficiency testing is one of the important measures to monitor the performance of a laboratory. It serves as an external measure of a laboratory’s ability to produce accurate and reliable results and verifies the competence of personnel, methodologies and equipment. By evaluating the results obtained in the proficiency testing program, how well a food microbiology laboratory can detect and/or quantify microorganisms such as Escherichia coli, coliforms can be assessed. Our laboratory has been participating in proficiency testing programs from an ISO/IEC 17043 accredited proficiency testing service provider for various parameters since 2016. This article discusses how we perform single-point assessment and long-term evaluation of our laboratory’s quantitative proficiency testing results. Assessment of a single proficiency testing result is usually done by observing the score (such as z score) which is provided by the proficiency testing service provider in the report. This single-point assessment is usually done by most of the laboratories that participate in the proficiency testing program. However, not many laboratories perform the long-term evaluation of the results accumulated over the years. Long-term evaluation is important because it indicates the consistency of a laboratory’s performance in its measurements over time. In our laboratory, long-term evaluation is done by calculating the two variables, i.e., long-term analytical coefficient of variation (LCV_a) and long-term total bias. This article also discusses the findings of single-point assessment and long-term evaluation of the results obtained from the PT programs our laboratory participated.

Laboratory errors across the total testing process pose significant risks to patient safety, with downstream implications for diagnostic accuracy, particularly in high-workload clinical biochemistry laboratories. The revised International Organization for Standardization (ISO) 15189:2022 standard explicitly mandates risk-based thinking as core operational requirement; however, structured frameworks translating this mandate into practice within NABL-accredited Indian laboratories remain limited. This study aimed to develop and apply a structured risk assessment, Layered Risk Management Framework (LRMF) in a National Accreditation Board for Testing and Calibration Laboratories (NABL)-accredited tertiary care clinical biochemistry laboratory, operationalizing ISO 15189:2022 risk management principles across the total testing process. Following Institutional Ethics Committee approval, a multidisciplinary risk management group conducted comprehensive process mapping of the laboratory workflow from specimen receipt to report release. Risk identification was performed through focused group discussions and retrospective review of quality records. Each risk was quantified using a Risk Priority Number (RPN) derived from likelihood, severity, and detectability scores, classified into five actionable tiers (X1–X5), and causally mapped using the 6 M framework. Of the 103 identified risks, the pre-examination phase contributed the greatest burden (38 %), with human factors dominating causal attribution across all phases (72 %). Twenty-one high-priority risks were escalated for structured corrective and preventive action, including one critical Tier X5 error and two serious Tier X4 errors, all arising from pre-examination human failures involving specimen labeling and patient identification. Responsible personnel, timelines, and residual risk expectations were formally assigned for every prioritized risk. The LRMF demonstrated practical utility as a systematic, reproducible, and accreditation-aligned tool for embedding proactive risk governance within clinical biochemistry laboratory, with applicability across comparable NABL-accredited settings.

External quality assessment (EQA) is an essential component of clinical microbiology, ensuring diagnostic reliability and underpinning antimicrobial resistance (AMR) surveillance. However, longitudinal data evaluating inter-tier laboratory performance and process stability in large urban healthcare networks remain limited. This study aimed to longitudinally evaluate the diagnostic proficiency of clinical microbiology laboratories in Ho Chi Minh City from 2022 to 2025, identifying pathogen-specific technical challenges and analyzing performance trajectories across different laboratory capacity tiers. A retrospective analysis was conducted using cumulative EQA data from 244 participating laboratories. Proficiency was evaluated across 3365 analytical challenges, encompassing Gram staining, bacterial identification (ID), quantitative urine culture, and antimicrobial susceptibility testing (AST). Laboratories were stratified into four structural capacity groups (A, B, C1, and C2). Performance trends and inter-tier disparities were evaluated using Cochran-Armitage trend tests, Pearson Chi-square analysis, and p-control charts. The regional laboratory network demonstrated high diagnostic proficiency, with a four-year pooled unacceptable rate of just 2.4%. A p-chart analysis confirmed that the system remained entirely within statistical control limits (UCL = 4.1%, LCL = 0.7%). Diagnostic errors were highly pathogen-specific, notably concentrated in fastidious or morphologically ambiguous organisms such as Listeria monocytogenes (26.3% Gram stain error in 2023) and Klebsiella pneumoniae (12.2% ID error in 2025). Conversely, flawless accuracy (0.0% error) was achieved in identifying critical multidrug-resistant organisms (MRSA, VRE, and ESBL-producers). Crucially, the proportion of all laboratories achieving perfect 100% target scores increased significantly over time (p < 0.001). By the end of the study, the EQA program had effectively bridged initial capacity gaps, culminating in no significant statistical difference among tiers in achieving perfect diagnostic scores (Group A: 57.5% vs. Group C1 53.0%; p = 0.595). Sustained participation in the EQA program successfully drives continuous quality improvement and maintains robust systemic stability. The program exhibits a powerful equalizing effect, enabling resource-limited laboratories to bridge initial capacity gaps and attain diagnostic excellence comparable to centralized, high-capacity institutions.

Quality control laboratories in the pharmaceutical sector are increasingly required to demonstrate both technical competence and regulatory compliance. WHO Good Practices for Pharmaceutical Quality Control Laboratories (GPPQCL) and ISO/IEC 17025:2017 represent two prominent frameworks addressing these needs. While ISO/IEC 17025:2017 emphasizes accreditation, impartiality, and measurement traceability, GPPQCL emphasizes regulatory compliance, public health protection, and alignment with GMP principles. This review provides a narrative comparative analysis, highlighting similarities, differences, and integration strategies. Key areas include management system requirements, technical requirements, risk management, data integrity, decision rules, accreditation, and regulatory implications. A clause mapping table is presented to facilitate harmonization, and practical recommendations are offered for laboratories pursuing dual implementation. The analysis demonstrates that harmonized adoption enhances technical credibility, regulatory compliance, and public health assurance, enabling laboratories to meet both international and national standards.

This study presents a practical, evidence-based approach to improving local or outdated standard methods by making effective use of validated leftover materials from proficiency testing (PT) schemes. Rather than limiting their use to routine internal quality control (IQC), these well-characterized materials were employed under real laboratory conditions to reveal subtle methodological gaps within existing standard procedures. A case study—histamine in tuna—demonstrates how empirical PT data can expose systematic biases and support the revision of a local standard method. The findings provide supporting evidence that PT leftovers, when properly evaluated for stability and assigned values, can serve as fit-for-purpose evaluation materials for real-world method verification. Overall, this work highlights the potential of PT leftovers as powerful tools for advancing continuous improvement and evidence-based updates in local standardization systems.

Vendor-generated instrument performance verification (IPV) records for inductively coupled plasma mass spectrometry (ICP-MS) are often retained in laboratory archives as pass/fail-oriented documents, but their retrospective record-review value depends on numeric extractability, applicable specification limits, record provenance, and duplicate/overlap control. This study presents an evidence-graded retrospective record-level screening framework for heterogeneous ICP-MS IPV records. The framework combines evidence grading, documentation coding, duplicate/overlap reconciliation, record-specific limit matching, normalized distance-to-limit (NDL), and a six-field Metric Completeness Index (MCI). It was applied as a single-laboratory case demonstration to 24 primary records after duplicate/overlap reconciliation from two PerkinElmer platforms: NexION 300 (16 records, 2012–2025) and NexION 1000 (8 records, 2020–2025). The final inventory contained 21 Grade 1 records, one Grade 4 record, and two Grade 5 preventive-maintenance IPV (PM-IPV) records; only Grade 1 and Grade 2 records were eligible for NDL. Median MCI was 100% for both platforms. NDL screening generated an audit-oriented verification-priority worklist: among 185 eligible metric-level NDL rows, 53 rows were near-limit (0% ≤ NDL < 20%) and 14 rows were out-of-limit (NDL < 0%). These rows are pending source-PDF and limit-match confirmation and are not confirmed nonconformities; the worklist is a screening output, not a conformity decision. The framework logic is vendor-neutral, but implementation requires platform-specific metric dictionaries, specification-limit registers, configuration-matching rules, and duplicate/overlap controls; external validation is required before outputs are used as general ICP-MS performance benchmarks. It supports audit-oriented interpretation of archived IPV records but does not replace analytical method validation or uncertainty-aware conformity assessment.

Metrology—the science of measurement—underpins innovation, trade, and evidence-based decision-making, yet its foundational role in discussions of technological progress is often overlooked. This article explores the findings of NPL’s most recent foresighting exercise, which examines how global change and emerging technologies are reshaping the role of metrology in the future of innovation. Four key drivers: wellbeing, enterprise, sustainability, and security, frame the conclusions, highlighting challenges from decentralised healthcare and quantum technologies to climate monitoring and cybersecurity. They also emphasise the increasing complexity of systems—especially biological, technological, and societal—that metrology must help to understand and manage. An argument is made for a cultural shift towards a “metrology mindset” and the evolving role of National Metrology Institutes (NMIs) in delivering this vision as agile, digital, and globally collaborative centres of trust in data is outlined. By embedding metrological principles into every stage of research and innovation, NMIs can ensure that science continues to improve lives, drive prosperity, and safeguard the planet. The document aims to start a renewed dialogue among metrologists and stakeholders about this vision for the future by describing some important future challenges for NMIs.