Quality-Improvement Measures as Effective Ways of Preventing Laboratory Errors

Revolutionizing clinical laboratories: The impact of artificial intelligence in diagnostics and patient care

INTRODUCTION

The integration of artificial intelligence (AI) is fundamentally transforming clinical laboratories, significantly improving diagnostic precision and operational effectiveness in the fields of pathology, microbiology, and biochemistry. This evolution holds great promise for advancing patient care and enhancing disease management strategies.

METHODOLOGY

A comprehensive literature review was performed using established databases, including Google Scholar, Embase, Science Direct, Scopus, and PubMed. Recent studies published between 2020 and 2024 were sourced using focused keywords related to AI's application in clinical laboratory settings. The inclusion criteria prioritized peer-reviewed articles that contributed to innovations in diagnostic methodologies and operational efficiency. A thematic analysis was conducted to collate findings regarding AI's impact across the preanalytical, analytical, and postanalytical phases of laboratory work.

DISCUSSION

AI significantly enhances various laboratory processes, such as histopathology, immunohistochemistry, and microbiological diagnostics. Notable applications include workflow automation, detailed analysis of biomarker data, and real-time processing to facilitate clinical decision-making. However, the benefits of AI come with challenges, including concerns about data integrity, ethical implications, and potential biases in algorithms, requiring careful management as AI becomes more integrated into clinical practice.

CONCLUSION

The future of clinical laboratories is poised for increased automation and the incorporation of AI and IoT technologies. While these advancements offer the potential for improved healthcare outcomes through greater accuracy and efficiency, evolving ethical and legal frameworks are crucial to address issues related to data privacy and accountability of algorithms. Ongoing adaptation and exploration of AI applications are vital to fully harnessing its capabilities in diagnostics.

The Impact of Artificial Intelligence on Microbial Diagnosis

Traditional microbial diagnostic methods face many obstacles such as sample handling, culture difficulties, misidentification, and delays in determining susceptibility. The advent of artificial intelligence (AI) has markedly transformed microbial diagnostics with rapid and precise analyses. Nonetheless, ethical considerations accompany AI adoption, necessitating measures to uphold patient privacy, mitigate biases, and ensure data integrity. This review examines conventional diagnostic hurdles, stressing the significance of standardized procedures in sample processing. It underscores AI's significant impact, particularly through machine learning (ML), in microbial diagnostics. Recent progressions in AI, particularly ML methodologies, are explored, showcasing their influence on microbial categorization, comprehension of microorganism interactions, and augmentation of microscopy capabilities. This review furnishes a comprehensive evaluation of AI's utility in microbial diagnostics, addressing both advantages and challenges. A few case studies including SARS-CoV-2, malaria, and mycobacteria serve to illustrate AI's potential for swift and precise diagnosis. Utilization of convolutional neural networks (CNNs) in digital pathology, automated bacterial classification, and colony counting further underscores AI's versatility. Additionally, AI improves antimicrobial susceptibility assessment and contributes to disease surveillance, outbreak forecasting, and real-time monitoring. Despite a few limitations, integration of AI in diagnostic microbiology presents robust solutions, user-friendly algorithms, and comprehensive training, promising paradigm-shifting advancements in healthcare.

Validation of the efficacy of pooled serum for serum glucose inhouse quality control material in comparison with commercial internal quality control in clinical chemistry laboratory

Background

This study aimed to create an in-house glucose quality control material for humans, addressing the challenge of obtaining high-cost commercially prepared quality control materials in developing countries.

Methods

An in-house quality control material for glucose was prepared using a pooled serum sample and analyzed using a fully automated chemistry analyzer following the ISO 80 guidelines. The mean, standard deviation (SD), and coefficient of variance were calculated from the first 30 days of measurement, and the variability was checked over eight months using SPSS software. The study used Pearson's correlation with a 95% confidence interval and a P-value less than 0.05, which was statistically significant.

Results

The average mean ± SD of human serum glucose was 185.2 ± 8.4 mg/dL, indicating that the precision between each measurement was better. The prepared in-house quality control material was stable for approximately five months without any significant change in the serum glucose concentration (mg/dl) (p-value<0.05).

Conclusions

The study suggests that room-temperature, 2-8 °C, and -20 °C to -30 °C storage of human serum samples for glucose analysis is a viable option, with stable glucose concentrations for up to 30 days. Pooled serum is a cost-effective method for in-house quality control, especially in resource-limited laboratories.

Monkeypox Diagnosis in Clinical Settings: A Comprehensive Review of Best Laboratory Practices

An outbreak of monkeypox (Mpox) was reported in more than 40 countries in early 2022. Accurate diagnosis of Mpox can be challenging, but history, clinical findings, and laboratory diagnosis can establish the diagnosis. The pre-analytic phase of testing includes collecting, storing, and transporting specimens. It is advised to swab the lesion site with virus transport medium (VTM) containing Dacron or polyester flock swabs from two different sites. Blood, urine, and semen samples may also be used. Timely sampling is necessary to obtain a sufficient amount of virus or antibodies. The analytical phase of infectious disease control involves diagnostic tools to determine the presence of the virus. While polymerase chain reaction (PCR) is the gold standard for detecting Mpox, genome sequencing is for identifying new or modified viruses. As a complement to these methods, isothermal amplification methods have been designed. ELISA assays are also available for the determination of antibodies. Electron microscopy is another effective diagnostic method for tissue identification of the virus. Wastewater fingerprinting provides some of the most effective diagnostic methods for virus identification at the community level. The advantages and disadvantages of these methods are further discussed. Post-analytic phase requires proper interpretation of test results and the preparation of accurate patient reports that include relevant medical history, clinical guidelines, and recommendations for follow-up testing or treatment.

Applications of Artificial Intelligence in Microbial Diagnosis

The diagnosis is an important factor in healthcare care, and it is essential to identify microorganisms that cause infections and diseases. The application of artificial intelligence (AI) systems can improve disease management, drug development, antibiotic resistance prediction, and epidemiological monitoring in the field of microbial diagnosis. AI systems can quickly and accurately detect infections, including new and drug-resistant strains, and enable early detection of antibiotic resistance and improved diagnostic techniques. The application of AI in bacterial diagnosis focuses on the speed, precision, and identification of pathogens and the ability to predict antibiotic resistance.

Artificial Intelligence for Kidney Stone Spectra Analysis: Using Artificial Intelligence Algorithms for Quality Assurance in the Clinical Laboratory

Objective

To determine if a set of artificial intelligence (AI) algorithms could be leveraged to interpret Fourier transform infrared spectroscopy (FTIR) spectra and detect potentially erroneous stone composition results reported in the laboratory information system by the clinical laboratory.

Background

Nephrolithiasis (kidney stones) is highly prevalent, causes significant pain, and costs billions of dollars annually to treat and prevent. Currently, FTIR is considered the reference method for clinical kidney stone constituent analysis. This process, however, involves human interpretation of spectra by a qualified technologist and is susceptible to human error.

Methods

This prospective validation study was conducted from October 29, 2020, to October 28, 2021, to test if the addition of AI algorithm overreads to FTIR spectra could improve the detection rate of technologist-misclassified FTIR spectra. The preceding year was used as a control period. Disagreement between the AI overread and technician interpretation was resolved by an independent human interpretation. The rate of verified human misclassifications that resulted in revised reported results was the primary end point.

Results

Spectra of 81,517 kidney stones were reviewed over the course of 1 year. The overall clinical concordance between the technologist and algorithm was 90.0% (73,388/81,517). The report revision rate during the AI implementation period was nearly 8 times higher than that during the control period (relative risk, 7.9; 95% CI, 4.1-15.2).

Conclusion

This study demonstrated that an AI quality assurance check of human spectra interpretation resulted in the identification of a significant increase in erroneously classified spectra by clinical laboratory technologists.

Gauge against the machine: revisiting quality for multi-targeted serology platforms

Diagnosis of infections of public health significance, such as leptospirosis, often present challenges for laboratories. To counter common challenges and ensure quality driven health responses, rigorous validation and verification processes are required. Despite such rigor, however, can one be certain laboratory reports are truly reflective of infection, given the risk of rare, but potentially very significant quality oversights? Here we present a real-world scenario where diagnosis of leptospirosis cases was compromised over a 6-year period despite quality measures suggesting a well performing serological assay. A subsequent investigation revealed this was attributed to the programming of an automated microtitration plate analyser, evading detection by both quality control and external quality assurance processes. The quality oversight provides insight into potential limitations in quality processes in multi-targeted serological platforms.

Data-Driven Anomaly Detection in Laboratory Medicine: Past, Present, and Future

BACKGROUND

Anomaly detection is an integral component of operating a clinical laboratory. It covers both the recognition of laboratory errors and the rapid reporting of clinically impactful results. Procedures for identifying laboratory errors and highlighting critical results can be improved by applying modern data-driven approaches.

CONTENT

This review will prepare the reader to appraise anomaly detection literature, identify common sources of anomalous results in the clinical laboratory, and offer potential solutions for common shortcomings in current laboratory practices.

SUMMARY

Laboratories should implement data-driven approaches to detect technical anomalies and keep them from entering the medical record, while also using the full array of clinical metadata available in the laboratory information system for context-dependent, patient-centered result interpretations.

Hematology and coagulation preanalytics for clinical chemists: Factors intrinsic to the sample and extrinsic to the patient

In hematology and coagulation, diligence in the preanalytical phase of testing is of critical importance to obtaining reliable test results. If the sample used for testing is unsuitable, even outstanding analytical procedures and technology cannot produce a clinically-reliable result. Therefore, the intent of this manuscript is to review preanalytical factors intrinsic to the sample that affect the hematology and coagulation testing. Factors intrinsic to the sample (excluding in vivo anomalies) can be controlled, theoretically, by phlebotomists (including nurses) and laboratorians in the preanalytical phase of testing. Furthermore, the management and prevention of such factors is highlighted. Erroneous control of preanalytical factors can produce laboratory errors.

Practical considerations for laboratories: Implementing a holistic quality management system

A laboratory quality management system (LQMS) is an essential element for the effective operation of research, clinical, testing, or production/manufacturing laboratories. As technology continues to rapidly advance and new challenges arise, laboratories worldwide have responded with innovation and process changes to meet the continued demand. It is critical for laboratories to maintain a robust LQMS that accommodates laboratory activities (e.g., basic and applied research; regulatory, clinical, or proficiency testing), records management, and a path for continuous improvement to ensure that results and data are reliable, accurate, timely, and reproducible. A robust, suitable LQMS provides a framework to address gaps and risks throughout the laboratory path of workflow that could potentially lead to a critical error, thus compromising the integrity and credibility of the institution. While there are many LQMS frameworks (e.g., a model such as a consensus standard, guideline, or regulation) that may apply, ensuring that the appropriate framework is adopted based on the type of work performed and that key implementation steps are taken is important for the long-term success of the LQMS and for the advancement of science. Ultimately, it ensures accurate results, efficient operations, and increased credibility, enabling protection of public health and safety. Herein, we explore LQMS framework options for each identified laboratory category and discuss prerequisite considerations for implementation. An analysis of frameworks' principles and conformity requirements demonstrates the extent to which they address basic components of effective laboratory operations and guides optimal implementation to yield a holistic, sustainable framework that addresses the laboratory's needs and the type of work being performed.

COVID-19 Exposure During Neurology Practice: Results of American Academy of Neurology Survey

BACKGROUND

To determine the exposure risk for coronavirus 2019 (COVID-19) during neurology practice. Neurological manifestations of COVID-19 are increasingly being recognized mandating high level of participation by neurologists.

METHODS

An American Academy of Neurology survey inquiring about various aspects of COVID-19 exposure was sent to a random sample of 800 active American Academy of Neurology members who work in the United States. Use of second tier protection (1 or more including sterile gloves, surgical gown, protective goggles/face shield but not N95 mask) or maximum protection (N95 mask in addition to second tier protection) during clinical encounter with suspected/confirmed COVID-19 patients was inquired.

RESULTS

Of the 81 respondents, 38% indicated exposure to COVID-19 at work, 1% at home, and none outside of work/home. Of the 28 respondents who did experience at least 1 symptom of COVID-19, tiredness (32%) or diarrhea (8%) were reported. One respondent tested positive out of 12 (17%) of respondents who were tested for COVID-19 within the last 2 weeks. One respondent received health care at an emergency department/urgent care or was hospitalized related to COVID-19. When seeing patients, maximum protection personal protective equipment was used either always or most of the times by 16% of respondents in outpatient setting and 56% of respondents in inpatient settings, respectively.

CONCLUSIONS

The data could enhance our knowledge of the factors that contribute to COVID-19 exposure during neurology practice in United States, and inform education and advocacy efforts to neurology providers, trainees, and patients in this unprecedented pandemic.

Does admission order form design really matter? A reduction in urea blood test ordering

Introduction

Laboratory blood testing is one of the most high-volume medical procedures and continues to increase steadily with instances of inappropriate testing resulting in significant financial implications. Studies have suggested that the design of a standard hospital admission order form and laboratory request forms influence physician test ordering behaviour, reducing inappropriate ordering and promoting resource stewardship.

Aim/method

To redesign the standard medicine admission order form-laboratory request section to reduce inappropriate blood urea nitrogen (BUN) testing.

Results

A redesign of the standard admission order form used by general internal medicine physicians and residents in two large teaching hospitals in one health zone in Alberta, Canada led to a significant step reduction in the ordering of the BUN test on hospital admission.

Conclusions

Redesigning the standard medicine admission order form-laboratory request section can have a beneficial effect on the reduction in BUN ordering altering physician ordering patterns and behaviour.

Deep into Laboratory: An Artificial Intelligence Approach to Recommend Laboratory Tests

Laboratory tests are performed to make effective clinical decisions. However, inappropriate laboratory test ordering hampers patient care and increases financial burden for healthcare. An automated laboratory test recommendation system can provide rapid and appropriate test selection, potentially improving the workflow to help physicians spend more time treating patients. The main objective of this study was to develop a deep learning-based automated system to recommend appropriate laboratory tests. A retrospective data collection was performed at the National Health Insurance database between 1 January 2013, and 31 December 2013. We included all prescriptions that had at least one laboratory test. A total of 1,463,837 prescriptions from 530,050 unique patients was included in our study. Of these patients, 296,541 were women (55.95%), the range of age was between 1 and 107 years. The deep learning (DL) model achieved a higher area under the receiver operating characteristics curve (AUROC micro = 0.98, and AUROC macro = 0.94). The findings of this study show that the DL model can accurately and efficiently identify laboratory tests. This model can be integrated into existing workflows to reduce under- and over-utilization problems.

Histopathology laboratory paperwork as a potential risk of COVID-19 transmission among laboratory personnel

BACKGROUND

Healthcare workers have a higher risk of acquiring coronavirus disease 2019 (COVID-19). The process of requesting pathological investigations is usually handled manually through paper-based forms. This study evaluated the potential for paper-based request forms to transmit severe acute respiratory virus coronavirus-2 (SARS-CoV-2) to laboratory staff in order to make recommendations for dealing with hospital paperwork in a post-COVID-19 world.

METHODS

Paper-based forms were tracked from the time of test ordering until the release of the pathology report by calculating the time taken for the forms to reach the laboratory, and the exposure of each staff group to forms received from both high and moderate COVID-19 risk areas.

RESULTS

Four hundred and thirty-two (83%) of 520 forms were received in the laboratory within 24 h. The remaining 88 (17%) forms took ≥24 h to be handled by laboratory personnel. The mean daily exposure time to the paperwork for various laboratory staff was as follows: receptionists, 2.7 min; technicians, 5.5 min; and pathologists, 54.6 min.

CONCLUSION

More than 80% of the forms were handled by laboratory personnel within 24 h, carrying a high potential risk for viral transmission. It is recommended that paper-based request forms should be replaced by electronic requests that could be printed in the laboratory if required. Another option would be to sterilize received paperwork to ensure the safety of laboratory personnel. More studies are needed to detect the stability of SARS-CoV-2 on different surfaces and determine the potential risk of COVID-19 transmission via paper.

Evaluation of evidence-based urinalysis reflex to culture criteria: Impact on reducing antimicrobial usage

OBJECTIVES

To implement evidence-based urinalysis (UA) reflex criteria and to evaluate the impact of the intervention on reducing unnecessary antibiotic usage.

METHODS

A prospective intervention study was conducted on 4130 urine samples that were subjected to UA during March to May 2020. Results were analyzed in order to evaluate the effectiveness of newly implemented evidence-based criteria in predicting positive urine cultures. The intervention involved implementing evidence-based UA reflex criteria to ensure a high predictive value of the UA reflex parameters. Multivariable logistic regression was utilized to evaluate the effectiveness of these UA parameters in predicting positive urine cultures and to assess the impact of the new UA criteria on antibiotic usage.

RESULTS

A total of 4130 patient samples were included in the study; 60.1% (n = 2484) were from female patients and 39.9% (n = 1646) were from male patients. The total number of negative urine reflex samples was 3116, which accounted for 75.4% of the total UA reflex samples. In contrast, 24.6% of the urine reflex samples (n = 1014) returned positive UA results and were reflexed to urine culture. Among the urine samples that were cultured, 9% (n = 91) were negative urine cultures, while 91.0% (n = 923) were positive urine cultures. Chi-square analysis indicated highly statistically significant associations between the combination parameters of (≥5 white blood cells (WBCs) and positive nitrite) and positive urine cultures (Chi-square = 516.428, p < 0.001) and (≥5 WBCs and moderate or large esterase) and positive urine cultures (Chi-square = 503.387, p < 0.001). Additionally, Chi-square analysis indicated a highly statistically significant association between the combination parameters of (≥5 WBCs and ≥1 bacteria) and positive urine cultures (Chi-square = 434.806, p < 0.001). The statistical analysis showed that the implementation of evidence-based UA reflex criteria significantly decreased the number of urine cultures performed and potentially decreased the number of patients inappropriately treated with antibiotics from 45.1% to 9%.

CONCLUSIONS

In conclusion, ≥5 WBCs and positive nitrite yielded the highest positive predictive value of 98.00% and showed a highly significant association with positive urine cultures. It was observed that the new UA reflex criteria are highly effective in predicting positive urine cultures, thus potentially resulting in the reduction of unnecessary antibiotic usage.

Measurement uncertainty as a universal concept: can it be universally applicable in routine laboratory practice?

Measurement uncertainty (MU) of results is one of the basic recommended and accepted statistical methods in laboratory medicine, with which analytical and clinical evaluation of laboratory test quality is assessed. Literature data indicate that the calculation of MU is not a simple process, but that its assessment in daily laboratory practice should be reduced to routine and simple presentation, understandable to both laboratory professionals and physicians. In order to achieve this, it is necessary to understand the purpose of the test for which MU is to be determined. Various suggestions have been given for presentation of MU as a quantitative indicator of the quality of the final measurement result in the medical laboratory. Although MU refers to the final measurement result, this metrological concept reflects the entire laboratory measurement process. The data on estimated MU is used to interpret the measured numerical result, and represents quantitatively the quality of the measurement itself, i.e. how different are the results of multiple measurements of the analyte of interest in the same sample, as well as whether the method of determination itself is subjected to significant random and systematic deviation. Initially, in the metrological concept, the MU is viewed in relation to the true value of the analyte of interest. However, the true value of the analyte measured in the biological fluid matrix of the study population cannot be known. It is therefore considered the closest value obtained by the perfect method, for which the bias and inaccuracy, as measures of systematic and random error, are equal to zero, which is practically impossible to achieve in routine laboratory practice. Although current standards require accredited medical laboratories to estimate MU, none of these guidelines provide clear guidance on how this can be achieved in daily laboratory work. This review examines literary data and documents dealing with MU issues, but also highlights what additional terms and data should be considered when interpreting MU. This paper ultimately draws attention, and once again points out, that a simpler solution is needed for this universal concept to be formally and universally applicable in routine laboratory medicine practice.

Who to Test, When, and for What: Why Diagnostic Stewardship in Infectious Diseases Matters

New rapid molecular diagnostic technologies for infectious diseases provide faster diagnostic test results and, if used correctly, will enable more rapid delivery of care to patients. This perspective piece outlines how this new technology can be used more effectively-with a focus on collaborative team approaches and tools clinicians and laboratorians can use to optimally affect patient care. This article also showcases a patient case, outlining problems with the diagnostic process as it currently stands, and poses potential strategies on how this process may be improved.

Pathology Services in Nigeria: Cross-Sectional Survey Results From Three Cancer Consortia

PURPOSE

Cancer incidence is increasing in sub-Saharan Africa, yet there is little information on the capacity of pathology laboratories in this region. We aimed to assess the current state of pathology services in Nigeria to guide strategies to ensure best practices and improve the quality of surgical specimen handling.

METHODS

We developed structured pathology survey to assess tissue handling, sample processing, and immunohistochemistry (IHC) capabilities. The survey was distributed electronically to 22 medical centers in Nigeria that are part of established cancer consortia. Data were collected between September and October 2017.

RESULTS

Sixteen of 22 centers completed the survey in full. All 16 institutions had at least one board-certified pathologist and at least one full-time laboratory scientist/technologist. The majority of responding institutions (75%) reported processing fewer than 3,000 samples per year. For sample processing, 38% of institutions reported manual tissue processing and 75% processed biopsies and surgical specimens together. The average tissue fixation time ranged from 5 to more than 72 hours before processing and paraffin embedding. Half of the institutions reported having no quality assurance processes to evaluate hematoxylin and eosin–stained slides, and 25% reported having no written operating procedures. Half of the participating institutions have a facility for routine IHC staining, and among these there was considerable variability in processes and validation procedures. External proficiency testing was not common among surveyed sites (38%).

CONCLUSION

Data from 16 Nigerian medical institutions indicate deficiencies in standardization, quality control, and IHC validation that could affect the reliability of pathology results. These findings highlight addressable gaps in pathology services that can ensure accurate diagnosis and follow-up for the growing number of patients with cancer in this region.

A Retrospective Analysis of Corrected Laboratory Results in a Large Academic Medical Center

OBJECTIVES

To evaluate the total number of corrected laboratory results that occurred as a consequence of errors in the preanalytical, analytical, and postanalytical phases of testing and to identify areas for improvement in patient care.

METHODS

Revised laboratory results were retroactively reviewed for a 6-month period. The total number of revised results were categorized by department and the phase in which the resulting error occurred.

RESULTS

Of 1,278,783 reportable tests, a total of 156 revisions were noted. Errors were predominately observed in the postanalytical phase of testing; 57.69% (n = 156) of all revisions were called and documented to a care provider.

CONCLUSION

Monitoring revised laboratory results is an important quality indicator that helps to identify areas within the laboratory that warrant improvement. Education, training, and ongoing process improvement initiatives are essential parts of a laboratory's quality management system in order to limit the total number of future revisions.