Assessment of accuracy, clinical validity, and analytical linearity in point-of-care glucose monitoring devices for diabetes mellitus: A systematic review and meta-analysis
{"title":"Assessment of accuracy, clinical validity, and analytical linearity in point-of-care glucose monitoring devices for diabetes mellitus: A systematic review and meta-analysis","authors":"Michael Pius Ukpe , Anastecia Chinasa Ezeanuka","doi":"10.1016/j.clinbiochem.2025.110911","DOIUrl":null,"url":null,"abstract":"<div><div>Point-of-care technology (POCT) is utilized in diabetes management due to its fast turnaround time. However, the accuracy and reliability of its results are of concern. This review aims to evaluate the (i) accuracy, (ii) clinical validity and (iii) linearity of POCT for blood glucose concentration following ISO 15197. The study was conducted following the PRISMA framework. Searches with the search term “point of care test” OR “POCT” AND “Accuracy” AND “glucose” were made on Scopus, PubMed, DOAJ, and Cochrane Library. Quality assessments were performed using the QUADAS-2 tool. Proportion and Coefficient <em>meta</em>-analyses were used. Recent studies (2019–2023) evaluating the accuracy of POC devices in blood glucose estimation were included. 4918 participants and 56 POC brands were identified from the 30 included studies. The pooled percentage of POCT results within ±15 mg/dL (±0.83 mmol/L) for glucose concentrations <100 mg/dL (<5.6 mmol/L) was 98.83 % (95 % CI: 96.52–99.92) for professional use and 87.70 % (95 % CI: 62.65–99.65) for home-use glucometers. For glucose concentrations ≥100 mg/dL (≥5.6 mmol/L), the pooled percentages were 98.59 % (95 % CI: 95.90–99.88) and 88.79 % (95 % CI: 67.35–99.44), respectively. Measurements in zones A and B of the consensus error grid analysis were 98.06 % (95 % CI: 94.59–99.80) for professional-use and 98.70 % (95 % CI: 95.85–99.95) for home-use glucometers. The pooled correlation coefficient for professional-use glucometers was 0.988 (95 % CI: 0.980–0.993) while home-use was 0.930 (95 % CI: 0.869–0.963). While professional-use glucometers met ISO 15197 accuracy criteria but not clinical validity standards, home-use devices failed to meet both accuracy and clinical validity criteria. Professional-use glucometers demonstrated superior accuracy compared to home-use glucometers. Based on the variability in analytical performance of POC devices, it is recommended that proper care be taken when selecting POCT devices for optimal use in diabetic management.</div></div>","PeriodicalId":10172,"journal":{"name":"Clinical biochemistry","volume":"137 ","pages":"Article 110911"},"PeriodicalIF":2.5000,"publicationDate":"2025-03-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Clinical biochemistry","FirstCategoryId":"3","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0009912025000402","RegionNum":3,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"MEDICAL LABORATORY TECHNOLOGY","Score":null,"Total":0}
引用次数: 0
Abstract
Point-of-care technology (POCT) is utilized in diabetes management due to its fast turnaround time. However, the accuracy and reliability of its results are of concern. This review aims to evaluate the (i) accuracy, (ii) clinical validity and (iii) linearity of POCT for blood glucose concentration following ISO 15197. The study was conducted following the PRISMA framework. Searches with the search term “point of care test” OR “POCT” AND “Accuracy” AND “glucose” were made on Scopus, PubMed, DOAJ, and Cochrane Library. Quality assessments were performed using the QUADAS-2 tool. Proportion and Coefficient meta-analyses were used. Recent studies (2019–2023) evaluating the accuracy of POC devices in blood glucose estimation were included. 4918 participants and 56 POC brands were identified from the 30 included studies. The pooled percentage of POCT results within ±15 mg/dL (±0.83 mmol/L) for glucose concentrations <100 mg/dL (<5.6 mmol/L) was 98.83 % (95 % CI: 96.52–99.92) for professional use and 87.70 % (95 % CI: 62.65–99.65) for home-use glucometers. For glucose concentrations ≥100 mg/dL (≥5.6 mmol/L), the pooled percentages were 98.59 % (95 % CI: 95.90–99.88) and 88.79 % (95 % CI: 67.35–99.44), respectively. Measurements in zones A and B of the consensus error grid analysis were 98.06 % (95 % CI: 94.59–99.80) for professional-use and 98.70 % (95 % CI: 95.85–99.95) for home-use glucometers. The pooled correlation coefficient for professional-use glucometers was 0.988 (95 % CI: 0.980–0.993) while home-use was 0.930 (95 % CI: 0.869–0.963). While professional-use glucometers met ISO 15197 accuracy criteria but not clinical validity standards, home-use devices failed to meet both accuracy and clinical validity criteria. Professional-use glucometers demonstrated superior accuracy compared to home-use glucometers. Based on the variability in analytical performance of POC devices, it is recommended that proper care be taken when selecting POCT devices for optimal use in diabetic management.
期刊介绍:
Clinical Biochemistry publishes articles relating to clinical chemistry, molecular biology and genetics, therapeutic drug monitoring and toxicology, laboratory immunology and laboratory medicine in general, with the focus on analytical and clinical investigation of laboratory tests in humans used for diagnosis, prognosis, treatment and therapy, and monitoring of disease.