View in EDS

Supporting trend detection in the cumulative display of electronic laboratory reports from multiple laboratories while preserving measurement provenance.

Saved in:
Bibliographic Details
Title: Supporting trend detection in the cumulative display of electronic laboratory reports from multiple laboratories while preserving measurement provenance.
Authors: Bietenbeck, Andreas1 (AUTHOR) lab@bietenbeck.net, Adler, Jakob2,3 (AUTHOR), Durner, Jürgen4 (AUTHOR), Gebauer, Julian5 (AUTHOR), Lüdemann, Sascha6 (AUTHOR), Müller, Burkhardt7 (AUTHOR), Orth, Matthias8 (AUTHOR), Streichert, Thomas9 (AUTHOR), Tolios, Alexander10 (AUTHOR), Wiegel, Bernhard7 (AUTHOR), von Meyer, Alexander11 (AUTHOR)
Source: Clinical Chemistry & Laboratory Medicine. Sep2025, p1. 6p. 2 Illustrations.
Subject Terms: *DATA visualization, *MEDICAL informatics, *PATTERN perception
Abstract: Electronic health records will increasingly aggregate longitudinal laboratory results from multiple providers, but availability alone does not guarantee safe interpretation. We present guidance, developed by laboratory professionals with the DGKL medical informatics division, for cumulative displays that are clinically meaningful. The core principle is to group medically comparable analyses while preserving laboratory provenance so that clinicians can follow true patient trends without conflating them with laboratory-induced variation. Comparability is defined algorithmically from Logical Observation Identifiers Names and Codes (LOINC) axis: analyses estimating the same patient property (allowing serum/plasma system equivalence and mathematically convertible properties such as substance vs. mass concentration) are grouped; coding of units is harmonized via Unified Code for Units of Measure (UCUM) with consistent conversion of numeric results and corresponding reference intervals, including inequality qualifiers. Analyte-specific conversion factors should come from authoritative sources; for poorly standardized measurands (e.g., tumor markers) or when conversions are inappropriate (e.g., Lp(a)), results remain separated. Methodological distinctions that affect interpretation – such as screening vs. confirmatory drug testing and point-of-care testing – are displayed independently to signal potential analytical discontinuities. A standardized, medically meaningful default result sequence – derived from LOINC metadata and clinical nomenclatures, with alphabetic naming as a pragmatic fallback – supports cross-laboratory aggregation; rare or novel tests lacking robust standardization remain as free text. The rules-based approach updates seamlessly with LOINC releases and remains compatible with the Nomenclature for Properties and Units (NPU), facilitating cross-border exchange within the European Health Data Space. While harmonized presentation improves trend analysis, true comparability ultimately requires measurement procedures traceable to reference methods and materials. [ABSTRACT FROM AUTHOR]
Database: Academic Search Index
Description
Abstract:Electronic health records will increasingly aggregate longitudinal laboratory results from multiple providers, but availability alone does not guarantee safe interpretation. We present guidance, developed by laboratory professionals with the DGKL medical informatics division, for cumulative displays that are clinically meaningful. The core principle is to group medically comparable analyses while preserving laboratory provenance so that clinicians can follow true patient trends without conflating them with laboratory-induced variation. Comparability is defined algorithmically from Logical Observation Identifiers Names and Codes (LOINC) axis: analyses estimating the same patient property (allowing serum/plasma system equivalence and mathematically convertible properties such as substance vs. mass concentration) are grouped; coding of units is harmonized via Unified Code for Units of Measure (UCUM) with consistent conversion of numeric results and corresponding reference intervals, including inequality qualifiers. Analyte-specific conversion factors should come from authoritative sources; for poorly standardized measurands (e.g., tumor markers) or when conversions are inappropriate (e.g., Lp(a)), results remain separated. Methodological distinctions that affect interpretation – such as screening vs. confirmatory drug testing and point-of-care testing – are displayed independently to signal potential analytical discontinuities. A standardized, medically meaningful default result sequence – derived from LOINC metadata and clinical nomenclatures, with alphabetic naming as a pragmatic fallback – supports cross-laboratory aggregation; rare or novel tests lacking robust standardization remain as free text. The rules-based approach updates seamlessly with LOINC releases and remains compatible with the Nomenclature for Properties and Units (NPU), facilitating cross-border exchange within the European Health Data Space. While harmonized presentation improves trend analysis, true comparability ultimately requires measurement procedures traceable to reference methods and materials. [ABSTRACT FROM AUTHOR]
ISSN:14346621
DOI:10.1515/cclm-2025-1160