Search

The amount of data generated in the field of laboratory medicine has grown to an extent that conventional laboratory information systems (LISs) are struggling to manage and analyze this complex, entangled information (“Big Data”). Statistical learning, a generalized framework from machine learning (ML) and artificial intelligence (AI) is predestined for processing “Big Data” and holds the potential to revolutionize the field of laboratory medicine. Personalized medicine may in particular benefit from AI-based systems, especially when coupled with readily available wearables and smartphones which can collect health data from individual patients and offer new, cost-effective access routes to healthcare for patients worldwide. The amount of personal data collected, however, also raises concerns about patient-privacy and calls for clear ethical guidelines for “Big Data” research, including rigorous quality checks of data and algorithms to eliminate underlying bias and enable transparency. Likewise, novel federated privacy-preserving data processing approaches may reduce the need for centralized data storage. Generative AI-systems including large language models such as ChatGPT currently enter the stage to reshape clinical research, clinical decision-support systems, and healthcare delivery. In our opinion, AI-based systems have a tremendous potential to transform laboratory medicine, however, their opportunities should be weighed against the risks carefully. Despite all enthusiasm, we advocate for stringent added-value assessments, just as for any new drug or treatment. Human experts should carefully validate AI-based systems, including patient-privacy protection, to ensure quality, transparency, and public acceptance. In this opinion paper, data prerequisites, recent developments, chances, and limitations of statistical learning approaches are highlighted.

Transportation of blood samples is a major part of the preanalytical pathway and can be crucial in delaying laboratory results to the clinicians. A variety of aspects however makes sample transportation a complex, challenging and often overlooked task that needs thorough planning and dedicated resources. The purpose of this review is to outline the options available for this task and to emphasize the preanalytical aspects that need consideration in this process, e.g. performance specifications for sample transportation as stated in ISO standards 15189 and 20658, quality control of automated transportation systems, monitoring of sample integrity parameters and temperature surveillance in general and for external samplers in particular. All these are tasks that the laboratory must assure on a daily basis in terms of continuous quality control, and simultaneously the laboratory must remain alert to alterations in clinical demands (sample frequency, turn-around-times) and new regulations within this area (e.g. the recent General Data Protection Regulation from the EU).

Several lines of evidence now confirm that the vast majority of errors in laboratory medicine occur in the extra-analytical phases of the total testing processing, especially in the preanalytical phase. Most importantly, the collection of unsuitable specimens for testing (either due to inappropriate volume or quality) is by far the most frequent source of all laboratory errors, thus calling for urgent strategies for improving blood sample quality and managing data potentially generated measuring unsuitable specimens. A comprehensive overview of scientific literature leads us to conclude that hemolyzed samples are the most frequent cause of specimen non-conformity in clinical laboratories (40–70%), followed by insufficient or inappropriate sample volume (10–20%), biological samples collected in the wrong container (5–15%) and undue clotting (5–10%). Less frequent causes of impaired sample quality include contamination by infusion fluids (i.e. most often saline or glucose solutions), cross-contamination of blood tubes additives, inappropriate sample storage conditions or repeated freezing-thawing cycles. Therefore, this article is aimed to summarize the current evidence about the most frequent types of unsuitable blood samples, along with tentative recommendations on how to prevent or manage these preanalytical non-conformities.

The new generation of clinical chemistry and coagulation analyzers is equipped with technical features allowing a systematic check of sample quality, including an assessment of the so-called HIL (“hemolysis”, “icterus”, “lipemia”) indices. These measures enable an accurate and reproducible assessment of sample hemolysis in serum or plasma, hence the hemolysis index (H-index) is now also increasingly used for monitoring and benchmarking phlebotomy performance. Reliable evidence attests that intravascular hemolysis is not such a rare phenomenon, and its prevalence may be especially higher in geographical areas where congenital hemolytic diseases are endemic, as well as in healthcare settings where patients with acquired hemolytic disorders are more frequently visited or hospitalized. It is hence conceivable that monitoring phlebotomy performance based on the rate of hemolyzed specimens received by the laboratory may not be so straightforward, provided that specimens drawn from patients with intravascular hemolysis can be identified and excluded from the analysis. The aim of this article is to provide an overview of potential drawbacks in using the H-index alone for monitoring phlebotomy performance, and to offer potential solutions to improve its efficiency for this scope. We therefore suggest that the H-index may only be used for purposes of benchmarking phlebotomy performance when the overall number of diagnoses of hemolytic diseases or the haptoglobin values measured by the laboratories are comparable across different healthcare settings or geographic areas.

Diagnostic pathways are an essential subset of clinical pathways and a logical consequence of DRG-based reimbursement. They combine the principle of stepwise reflex and reflective testing with a management concept that helps to fulfill medical needs with organizational and economic efficacy. The two most common formats describing diagnostic pathways are graphical decision trees on paper and “if…then…else” rules on computers. From a laboratory point of view, diagnostic pathways represent “smart” test profiles, which – in contrast to conventional (inflexible) profiles – are not necessarily worked off completely, but just to a point, where a diagnostic decision can be made. This improves the cost-effectiveness of laboratory testing, while making sure that no essential tests are missed. The paper describes benefits and limitations of diagnostic pathways from a medical, organizational, and economic point of view. Their major advantage is also their major drawback, since they make the diagnostic process on the one hand extremely straight-forward and transparent, while on the other hand oversimplifying the underlying medical decision principles. This may provoke the abuse of their primarily medical intentions for mere economic purposes.

Supplemental Digital Content is available in the text