CoChat
Sign Up
Cited by 3Clinical Laboratory Practices and Quality ControlDiabetes Management and ResearchAdvanced Glycation End Products research

Comparison of point‑of‑care HbA1c devices with laboratory HPLC methods: A systematic review and meta‑analysis

Open access
Wei, Yongzhong, Gu, Mingze, Ren, Yuqing, Hu, Zhongshuang·2026·Biomedical Reports
DOI 10.3892/br.2026.2120PMC PMC12977185
Open this paper in CoChat — free accountOpen PDF

Abstract

Hemoglobin A1c (HbA1c), also known as glycated hemoglobin, is a cornerstone biomarker for diagnosing and monitoring diabetes. While high-performance liquid chromatography (HPLC) is the reference standard, point-of-care (POC) devices are increasingly used due to convenience and rapid results. However, concerns remain about their analytical accuracy. This systematic review and meta-analysis evaluated the agreement between POC HbA1c devices and laboratory HPLC-based methods. A total of 11 studies involving 1,593 participants and 3,285 paired measurements were included. The pooled mean difference between POC and HPLC was 0.01% [95% confidence interval (CI), -0.08 to 0.11%; 95% prediction interval (PI), -0.443 to 0.471%; P=0.779], well within the clinically acceptable margin of ±0.5%. However, substantial heterogeneity was observed and extensive subgroup analyses across device model, analytical principle, and study characteristics did not fully explain this variability. Notably, some platforms, such as lateral flow immunoassays, exhibited larger biases approaching thresholds relevant to clinical decision-making. Although most POC devices demonstrated acceptable agreement, results highlight the importance of careful validation and interpretation. Future research should prioritize standardized protocols, comparative studies across platforms, and assessment in real-world decentralized settings to enhance confidence in POC HbA1c measurement.

Verification finished with gaps

Some checks didn’t finish. The results below cover the steps that did.

  1. Registry recordVerified
  2. Full textRead the full text
  3. Reference list49 references found
  4. Reference check50 of 50 references checked
  5. StatisticsRecomputed
  6. Claims95 claims anchored to a quote
  7. Cited sourcesChecked
  8. Self-checkDidn’t finish — the rest of the report still applies.
  9. VerificationFindings verified

AI snapshot

Written from the paper’s full text.

TL;DR

This systematic review and meta-analysis of 11 studies (1,593 participants; 3,285 paired measurements) found a pooled POC–HPLC HbA1c mean difference of 0.01% (95% CI, -0.08 to 0.11%), within the clinically acceptable ±0.5% margin. However, heterogeneity was substantial, and some platforms—notably the i-Chroma lateral-flow assay—showed larger bias, so device-specific validation remains important.

Key findings

  • The pooled mean difference between POC and HPLC was 0.01% (95% CI, -0.08 to 0.11%; P=0.779; 95% PI, -0.443 to 0.471%), corresponding to 0.1 mmol/mol on the IFCC scale; between-study heterogeneity was substantial (τ²=0.04; I²=98.96%).
  • Subgroup pooled biases were: non-chromatographic immunoassay platforms -0.02% (95% CI, -0.11 to 0.15%; τ²=0.03; I²=98.73%) and boronate affinity-based devices 0.04% (95% CI, -0.09 to 0.17%; τ²=0.05; I²=98.99%). The i-Chroma system showed the largest individual deviation, with a mean difference of -0.50% (95% CI, -0.80 to -0.20%).
  • Meta-regression found study setting explained the largest share of between-study variance (R²=20.6%), with decentralized testing associated with higher positive bias (borderline, P=0.067); POC analytical principle explained R²=6.4% (P=0.15), HPLC analytical principle R²=11.2% (P=0.10), and other moderators explained virtually none (all R²≤5%; P>0.10).
  • Sensitivity analyses were consistent: RVE-adjusted pooled mean difference was 0.03% (95% CI, -0.11 to 0.16%; P=0.683); leave-one-out re-estimated mean differences ranged from -0.01% to 0.03%; Egger’s test suggested small-study effects (P<0.001), but trim-and-fill imputed no studies and the adjusted pooled estimate remained 0.01%.

Methods

  • Systematic review and meta-analysis conducted per PRISMA 2020 and registered in PROSPERO (CRD420251063197); databases searched were PubMed, Embase, Web of Science, and Cochrane Library for January 2015–April 2025. Eligible studies were original research directly comparing POC and HPLC HbA1c measurements in the same patient population and reporting agreement metrics; 11 studies with 1,593 participants and 3,285 paired measurements were included.
  • Mean differences (POC−HPLC) in % HbA1c were pooled using a random-effects model with restricted maximum likelihood and Hartung-Knapp-adjusted confidence intervals; a 95% prediction interval was calculated. Heterogeneity was assessed with Cochran’s Q, I², and τ², with subgroup analyses and univariable random-effects meta-regression.
  • Additional analyses included robust variance estimation for within-study clustering, leave-one-out influence analysis, Egger’s test, funnel plots, and trim-and-fill; methodological quality was assessed with QUADAS-2. Analyses used Stata version 18.0.

Limitations

  • Stated in the paper: the number of included studies was relatively small, and several devices were represented by only one or two datasets, limiting statistical power and generalizability; many studies lacked detailed reporting on calibration protocols, quality control, and management of analytical interference; the analysis focused on analytical agreement rather than diagnostic accuracy, clinical outcomes, or cost-effectiveness; no formal GRADE evaluation was performed despite the original protocol; and no enzymatic POC platform studies met inclusion criteria, limiting generalizability to enzymatic systems.
  • Evident from the text: heterogeneity remained extreme (I² ~99%) and was not materially reduced by subgroup analyses or meta-regression, so the pooled estimate represents an average across diverse devices and settings rather than evidence of analytical interchangeability. The 95% prediction interval spanned nearly the full ±0.5% margin and crossed the 6.5% diagnostic cutoff, indicating individual devices or settings may classify patients differently near clinical thresholds.
  • Evident from the text: some included studies used venous blood rather than capillary fingerstick samples, which may underestimate real-world variability in decentralized settings; Egger’s test indicated possible small-study effects, though the authors interpreted this cautiously given the substantial heterogeneity.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.3892/br.2026.2120

Cites

46 of 50 references verified0 citation errors · 4 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

24 of 54 cited sources checked · 7 supported · 4 flagged · accuracy not yet measured

  • Premisecitation

    Deviations near key clinical thresholds, such as 6.5% for diagnosis or 7.0-8.0% for treatment decisions, may result in misclassification and inappropriate therapeutic interventions.

    “Notably, deviations near key clinical thresholds, such as 6.5% for diagnosis or 7.0-8.0% for treatment decisions, may result in misclassification and inappropriate therapeutic interventions (9,10).”
    Quote in the paperSubstantial claim1 of 2 sources checked
    • cites [9] Shephard M Shephard A McAteer B Regnier T Barancek K Results from 15 years of quality surveillance for a national indigenous point-of-care testing program for diabetes Clin Biochem 50 1159 1163 2017 10.1016/j.clinbiochem.2017.07.007 28734906 source
    • cites [10] Hirst JA McLellan JH Price CP English E Feakins BG Stevens RJ Farmer AJ Performance of point-of-care HbA1c test devices: Implications for use in clinical practice-A systematic review and meta-analysis Clin Chem Lab Med 55 167 180 2017 10.1515/cclm-2016-0303 27658148 source

      The Data extraction and quality assessment section identifies 6.5% as a diagnostic threshold and discusses false-positive and false-negative diagnoses. The Clinical implications section says measurement variability may change treatment decisions or lead to undertreatment. It does not identify 7.0–8.0% as treatment-decision thresholds; its 6%–8% range is an imprecision-analysis category. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
  • Premisecitation

    Immunoassay-based methods depend on antibody specificity for HbA1c, making them more vulnerable to epitope changes, masking, or variant interference.

    “By contrast, immunoassay-based methods depend on antibody specificity for HbA1c, making them more vulnerable to epitope changes, masking, or variant interference (40).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [40] Gaetani ML Pinto IC Li M O'Connor P Giorgi-Coll S Tyreman M Rumary KL Schouten JA Davis P Dixon AM Towards detection of structurally-diverse glycated epitopes in native proteins: Single-chain antibody directed to non-A1c epitope in human haemoglobin Mol Immunol 166 16 28 2024 10.1016/j.molimm.2023.12.008 38181455 source

      The Introduction and Discussion and conclusions (page numbers unavailable) describe HbA1c tests as recognizing a particular glycated epitope, while §3.2 evaluates antibody specificity in an ELISA. Section 3.6 shows that chemical alteration or denaturation of HbA1c prevented recognition by the antibody studied. The paper does not establish that immunoassays are more vulnerable than other methods, or specifically document masking or haemoglobin-variant interference. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
  • Findingcitation

    The majority of POC platforms, excluding the i-Chroma, meet the ±0.5% criterion for clinical agreement.

    “our findings, excluding the i-Chroma, indicate that the majority of POC platforms meet the ±0.5% criterion for clinical agreement (10).”
    • Also stated in: § Discussion · boosted
      “In conclusion, this systematic review and meta-analysis demonstrates that most POC HbA1c devices show acceptable analytical agreement with reference HPLC methods.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [10] Hirst JA McLellan JH Price CP English E Feakins BG Stevens RJ Farmer AJ Performance of point-of-care HbA1c test devices: Implications for use in clinical practice-A systematic review and meta-analysis Clin Chem Lab Med 55 167 180 2017 10.1515/cclm-2016-0303 27658148 source

      The Discussion and Clinical implications sections describe ±6% relative bias criteria, not a ±0.5% clinical-agreement criterion. Searches for ±0.5% and i-Chroma found no passage establishing that criterion or assessing whether a majority of platforms meet it. The sentence calls the platform results the citing authors’ own findings. (LLM, conf=0.76)

      Experimental — accuracy not yet measured
  • Premisecitation

    Analytical interference from hemoglobin variants or elevated fetal hemoglobin is a concern.

    “An additional concern is analytical interference from hemoglobin variants or elevated fetal hemoglobin.”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [45] Thom CS Dickson CF Gell DA Weiss MJ Hemoglobin variants: Biochemical properties and clinical correlates Cold Spring Harb Perspect Med 3 a011858 2013 10.1101/cshperspect.a011858 23388674 PMC3579210 source

      The Identification of Hb Variants and Their Clinical Implications section says some variants can artificially elevate HbA1c measurements, and Laboratory Testing for Hb Variants says variants can produce misleading pulse-oximetry results. These passages support analytical interference from hemoglobin variants, but our copy does not say that elevated fetal hemoglobin causes analytical interference. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [46] Yadav N Kumar Mandal A Interference of hemoglobin variants in HbA1c quantification Clin Chim Acta 539 55 65 2023 10.1016/j.cca.2022.11.031 36476843 source
  • Findingcitation

    Some included studies evaluated POC devices using venous blood instead of capillary fingerstick samples, despite the devices being designed for the latter.

    “From a practical standpoint, some studies evaluated POC devices using venous blood instead of capillary fingerstick samples, despite the devices being designed for the latter (30,33).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [30] Guadalupe Vargas M Pazmiño Gomez BJ Vera Lorenti FE Álvarez Condo GM Rodas Neira EI Veron D Fernández Veron M Cercado AG Bahar B Tufro A Veron D Assessment of two glycated hemoglobin immunoassays Endocrinol Diabetes Nutr (Engl Ed) 67 297 303 2020 10.1016/j.endinu.2019.09.011 31859182 source
    • cites [33] Rathod L Khan S Shubham S Bisne N Singh S Kumar M Tiwari R Sarma DK A comparative evaluation of point-of-care and laboratory HbA1c testing in diabetes care: An Indian perspective Cureus 16 e69956 2024 10.7759/cureus.69956 39445250 PMC11496592 source

      The Discussion says this study used venous blood for both the POC and laboratory HbA1c tests, whereas most previous studies used capillary blood for POC testing. It does not establish that the device was designed specifically for capillary fingerstick samples. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
  • May combine two claims. 24
  • Classified as probably not a substantial claim. 27
  • The quote only partly states this claim. 1

  • May combine two claims.

    Findingown data

    The systematic review and meta-analysis included 11 studies involving 1,593 participants and 3,285 paired HbA1c measurements comparing POC devices with reference HPLC-based laboratory methods.

    “A total of 11 studies involving 1,593 participants and 3,285 paired measurements were included.”
    • Also stated in: § Characteristics of included studies · plain
      “A total of 11 studies met the inclusion criteria and were included in the final meta-analysis (11,12,26-34). These studies encompassed 1,593 participants and provided 3,285 paired HbA1c measurements comparing POC devices with reference HPLC-based laboratory methods.”
    Quote in the paperSubstantial claim4 of 10 sources checked
    • cites [11] Criel M Jonckheere S Langlois M Evaluation of three hemoglobin A1c point-of-care instruments Clin Lab 62 285 291 2016 10.7754/clin.lab.2015.150301 27156315 source
    • cites [12] Grant DA Dunseath GJ Churm R Luzio SD Comparison of a point-of-care analyser for the determination of HbA1c with HPLC method Prac Lab Med 8 26 29 2017 10.1016/j.plabm.2017.04.001 28856223 PMC5575368 source

      The Abstract (Methods) and Materials and methods (Method comparison; page unavailable) describe a single comparison using 100 blood samples, not a systematic review or meta-analysis of 11 studies. The cited paper therefore does not support the attributed totals of 1,593 participants and 3,285 paired measurements. (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
    • cites [26] Menéndez-Valladares P Fernández-Riejos P Sánchez-Mora C Pérez-Pérez A Sánchez-Margalet V González-Rodríguez C Evaluation of a HbA1c point-of-care analyzer Clin Biochem 48 686 689 2015 10.1016/j.clinbiochem.2015.03.016 25835152 source
    • cites [27] Wang Y Peng W Tang J Dong L Gu C Zhang X Zhou J Jia W Verification of a novel point-of-care HbA1c device in real world clinical practice by comparison to three high performance liquid chromatography instruments Biochem Med 28 020705 2018 10.11613/BM.2018.020705 29666558 PMC5898960 source

      The Materials and methods (Study design) describe a single analytical verification study, not a systematic review or meta-analysis. Statistical analysis reports 514 subjects and 635 pairs of data, contrary to the attributed totals of 11 studies, 1,593 participants and 3,285 paired measurements. Page numbers are unavailable in our copy. (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
    • cites [28] Dupuy AM Badiou S Bargnoux AS Russelo J Cristol JP Evaluation of point of care analyzer for hemoglobin A1c J Diabetes Sci Technol 13 150 151 2019 10.1177/1932296818814544 30477338 PMC6313273 source

      The study description says this paper evaluates a point-of-care HbA1c analyzer against HPLC; it is not a systematic review or meta-analysis. Its comparison paragraph and Table 1 report 110 routine samples, not 11 studies, 1,593 participants, or 3,285 paired measurements. Page numbers are unavailable in our copy. (LLM, conf=0.89)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured
    • cites [29] Dubach IL Christ ER Diem P HbA(1c)-testing: Evaluation of two point-of-care analysers Prim Care Diabetes 13 583 587 2019 10.1016/j.pcd.2019.05.007 31175054 source
    • cites [30] Guadalupe Vargas M Pazmiño Gomez BJ Vera Lorenti FE Álvarez Condo GM Rodas Neira EI Veron D Fernández Veron M Cercado AG Bahar B Tufro A Veron D Assessment of two glycated hemoglobin immunoassays Endocrinol Diabetes Nutr (Engl Ed) 67 297 303 2020 10.1016/j.endinu.2019.09.011 31859182 source
    • cites [31] Şahingöz Erdal G Işıksaçan N Koşer M Kocamaz N Hemoglobin A1c measurement using point of care testing Istanbul Med J 21 37 41 2020 10.1353/cpr.2020.0038 33416608

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites [32] Toro-Crespo M Sánchez-Mora C Fernández-Riejos P Maesa-Márquez JM González-Rodríguez C Evaluation of 3 hemoglobin A1c point of care instruments. Point of care testing for HbA1c: Evaluation of Cobas b101, B-analyst and AfinionTM Clin Lab 63 1107 1112 2017 10.7754/Clin.Lab.2017.161212 28792719 source
    • cites [33] Rathod L Khan S Shubham S Bisne N Singh S Kumar M Tiwari R Sarma DK A comparative evaluation of point-of-care and laboratory HbA1c testing in diabetes care: An Indian perspective Cureus 16 e69956 2024 10.7759/cureus.69956 39445250 PMC11496592 source

      The Materials and methods section describes a single diagnostic-accuracy study in Bhopal, not a systematic review or meta-analysis. The Results section reports 121 participants whose HbA1c was measured by both POC and HPLC methods, contrary to the attributed totals of 11 studies, 1,593 participants, and 3,285 paired measurements. The indexed passages provide no page numbers. (LLM, conf=0.90)

      Checked by a reasoning model — may be wrong.

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    A threshold of 6.5% is recommended by international guidelines for diagnostic purposes.

    “A threshold of 6.5% is recommended by international guidelines for diagnostic purposes (4).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [4] American Diabetes Association Diagnosis and classification of diabetes mellitus Diabetes Care 33 S62 S69 2010 10.2337/dc10-S062 20042775 PMC2797383 source

  • Classified as probably not a substantial claim. May combine two claims.

    Premisecitation

    A prior meta-analysis reported substantial variability in bias across different POC platforms, with some devices showing deviations beyond clinically acceptable limits.

    “For example, a prior meta-analysis reported substantial variability in bias across different POC platforms, with some devices showing deviations beyond clinically acceptable limits (10).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites [10] Hirst JA McLellan JH Price CP English E Feakins BG Stevens RJ Farmer AJ Performance of point-of-care HbA1c test devices: Implications for use in clinical practice-A systematic review and meta-analysis Clin Chem Lab Med 55 167 180 2017 10.1515/cclm-2016-0303 27658148 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductcitation

    A formal GRADE evaluation was not performed; instead, methodological quality was assessed using QUADAS-2.

    “Although the protocol initially planned to assess certainty of evidence using the GRADE framework, a formal GRADE evaluation was not performed in the final manuscript (25). Instead, methodological quality of individual studies was assessed using the QUADAS-2 tool, which is more appropriate for method-comparison studies focusing on analytical agreement.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [25] Guyatt GH Oxman AD Vist GE Kunz R Falck-Ytter Y Alonso-Coello P Schünemann HJ GRADE: An emerging consensus on rating quality of evidence and strength of recommendations BMJ 336 924 926 2008 10.1136/bmj.39489.470347.AD 18436948 PMC2335261 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The systematic review and meta-analysis was conducted in accordance with PRISMA 2020 guidelines.

    “This systematic review and meta-analysis was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 guidelines (16).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [16] Page MJ McKenzie JE Bossuyt PM Boutron I Hoffmann TC Mulrow CD Shamseer L Tetzlaff JM Akl EA Brennan SE The PRISMA 2020 statement: An updated guideline for reporting systematic reviews BMJ 372 n71 2021 10.1136/bmj.n71 33782057 PMC8005924 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    A 95% prediction interval was calculated to reflect the expected dispersion of true effects in future clinical settings.

    “In addition to the pooled 95% CI, a 95% prediction interval (PI) was calculated to reflect the expected dispersion of true effects in future clinical settings (20,21).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [20] Hartung J Knapp G A refined method for the meta-analysis of controlled clinical trials with binary outcome Statist Med 20 3875 3889 2001 10.1002/sim.1009 11782040 source
    • cites [21] Higgins JPT Thompson SG Spiegelhalter DJ A re-evaluation of random-effects meta-analysis J R Stat Soc Ser A Stat Soc 172 137 159 2009 10.1111/j.1467-985X.2008.00552.x 19381330 PMC2667312 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The 95% prediction interval reflects the expected dispersion of true effects in future clinical settings.

    “a 95% prediction interval (PI) was calculated to reflect the expected dispersion of true effects in future clinical settings (20,21).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [20] Hartung J Knapp G A refined method for the meta-analysis of controlled clinical trials with binary outcome Statist Med 20 3875 3889 2001 10.1002/sim.1009 11782040 source
    • cites [21] Higgins JPT Thompson SG Spiegelhalter DJ A re-evaluation of random-effects meta-analysis J R Stat Soc Ser A Stat Soc 172 137 159 2009 10.1111/j.1467-985X.2008.00552.x 19381330 PMC2667312 source

  • May combine two claims.

    Premisecitation

    In decentralized settings, several additional factors may contribute to reduced accuracy, including operator training, sample collection technique, time from sampling to analysis, ambient environmental conditions, and adherence to device-specific quality control protocols.

    “In decentralized settings, several additional factors may contribute to reduced accuracy, including operator training, sample collection technique, time from sampling to analysis, ambient environmental conditions (including temperature and humidity), and adherence to device-specific quality control protocols (13).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [13] Khan AI Pratumvinit B Jacobs E Kost GJ Kary H Balla J Shaw J Milevoj Kopcinovic L Vaubourdolle M Oliver P Point-of-care testing performed by healthcare professionals outside the hospital setting: Consensus based recommendations from the IFCC committee on point-of-care testing (IFCC C-POCT) Clin Chem Lab Med 61 1572 1579 2023 10.1515/cclm-2023-0502 37267483 source

      The quality-management passage (section 4; page unavailable) calls for operator training and monitoring temperature and humidity during storage, while sections 4.1 and 5 address quality control and sample-collection competency as safeguards for reliable results. The copy does not identify time from sampling to analysis as a factor reducing accuracy, and its temperature-and-humidity recommendation concerns storage rather than ambient conditions during testing. (LLM, conf=0.85)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    Clinically prevalent hemoglobin variants include HbS, HbC, HbE, and HbD, which occur at relatively high frequencies in individuals of African, Mediterranean, and Southeast Asian descent.

    “These variants include clinically prevalent types such as HbS (sickle cell), HbC, HbE, and HbD, which occur at relatively high frequencies in individuals of African, Mediterranean, and Southeast Asian descent (45).”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites [45] Thom CS Dickson CF Gell DA Weiss MJ Hemoglobin variants: Biochemical properties and clinical correlates Cold Spring Harb Perspect Med 3 a011858 2013 10.1101/cshperspect.a011858 23388674 PMC3579210 source

      The Abstract lists HbS, HbC, and HbE among the most common and medically important variants, and ‘Variants that Affect Multiple Hemoglobin Functions’ calls HbE common in Southeast Asia. The copy does not identify HbD as a prevalent variant or establish the claimed high frequencies among people of African or Mediterranean descent; no page numbers are provided for these passages. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    Enzymatic methods are widely used in centralized laboratories and rely on endoglycosidase digestion followed by quantification of released glycated amino acids.

    “In addition, although enzymatic methods are widely used in centralized laboratories and rely on endoglycosidase digestion followed by quantification of released glycated amino acids, no studies of enzymatic POC platforms met our inclusion criteria (42-44).”
    Quote in the paperSubstantial claim2 of 3 sources checked
    • cites [42] Yonehara S Inamura N Fukuda M Sugiyama K Use of fructosyl peptide oxidase for HbA1c assay J Diabetes Sci Technol 9 200 205 2015 10.1177/1932296815569573 25633966 PMC4604587 source

      The Abstract (page unavailable) describes an enzymatic HbA1c assay on an automatic analyzer: protease releases the glycated peptide fructosyl-Val-His, which is measured through an FPOX reaction. The Conclusion (page unavailable) says the method enables high-volume measurement on a clinical chemistry analyzer. These passages support an automated laboratory assay involving digestion and measurement of a glycated product, but they do not establish that enzymatic methods are widely used, and they specify protease and a glycated peptide—not endoglycosidase and released glycated amino acids. (LLM, conf=0.86)

      Experimental — accuracy not yet measured
    • cites [43] Liu L Hood S Wang Y Bezverkov R Dou C Datta A Yuan C Direct enzymatic assay for %HbA1c in human whole blood samples Clin Biochem 41 576 583 2008 10.1016/j.clinbiochem.2008.01.013 18261468 source
    • cites [44] Nitta T Yamashiro Y Hattori Y Ezumi T Nishioka M Nakamura J The interference by HbF on HbA1c (BM test HbA1c) measurement in enzymatic method Ann Clin Biochem 52 569 575 2015 10.1177/0004563214568872 25587198 source

      The Conclusion (page unavailable) describes an enzymatic HbA1c method used with a high-throughput routine automated analyzer, but does not establish that enzymatic methods are widely used in centralized laboratories. The Discussion (page unavailable) describes endopeptidase cleavage that releases a glycated dipeptide, followed by its reaction with fructosyl peptide oxidase—not endoglycosidase digestion followed by quantification of released glycated amino acids. (LLM, conf=0.87)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Guidelines from the ADA and WHO recommend alternative diagnostic approaches, such as fasting glucose, OGTT, or short-term glycation markers, when hemoglobinopathies are suspected or confirmed.

    “Guidelines from the ADA and WHO recommend alternative diagnostic approaches, such as fasting glucose, OGTT, or short-term glycation markers (such as fructosamine and glycated albumin), when hemoglobinopathies are suspected or confirmed (47,48).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [47] Colagiuri S Glycated haemoglobin (HbA1c) in the diagnosis of diabetes mellitus Diabetes Res Clin Pract 93 299 309 2011 10.1016/j.diabres.2011.06.025 21820751 source
    • cites [48] American Diabetes Association Professional Practice Committee ElSayed NA McCoy RG Aleppo G Balapattabi K Beverly EA Early KB Bruemmer D Ebekozien O Echouffo-Tcheugui JB 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes-2025 Diabetes Care 48 S27 S49 2025 39651986 PMC11635041 source

      Recommendation 2.4 (section 2.2a; page unavailable) says to use plasma-glucose criteria for diagnosis when some hemoglobin variants alter the relationship between A1C and glycemia; section 2.1a includes fasting glucose and OGTT among those criteria. The paper does not recommend fructosamine or glycated albumin as diagnostic alternatives: “Confirming the Diagnosis” describes them as measures approved for monitoring glycemic management. It also says an A1C assay without variant interference can be used for some people with a hemoglobin variant. (LLM, conf=0.90)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Previous reviews have emphasized that variability remains a persistent challenge in standardizing HbA1c measurement across platforms and settings.

    “Previous reviews have emphasized that such variability remains a persistent challenge in standardizing HbA1c measurement across platforms and settings (10,35).”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites [10] Hirst JA McLellan JH Price CP English E Feakins BG Stevens RJ Farmer AJ Performance of point-of-care HbA1c test devices: Implications for use in clinical practice-A systematic review and meta-analysis Clin Chem Lab Med 55 167 180 2017 10.1515/cclm-2016-0303 27658148 source

      The Results and Clinical implications sections report substantial variability among point-of-care HbA1c devices compared with laboratory methods, supporting the concern about consistent measurements across platforms. However, Clinical implications says there was insufficient published data to compare performance between settings for most devices, so the claim’s extension to variability across settings is not established. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites [35] Lenters-Westra E Slingerland RJ Hemoglobin A1c point-of-care assays; a new world with a lot of consequences! J Diabetes Sci Technol 3 418 423 2009 10.1177/193229680900300303 20144277 PMC2769862 source

  • May combine two claims.

    Premisecitation

    A previous meta-analysis noted a non-significant trend toward less negative bias over time for the DCA device (coefficient 0.01% HbA1c per year; P=0.081), with no significant temporal effect for other devices.

    “This observation aligns with a previous meta-analysis which noted a non-significant trend toward less negative bias over time for the DCA device (coefficient 0.01% HbA1c per year; P=0.081), suggesting possible calibration improvements, although no significant temporal effect was observed for other devices (10,36,37).”
    Quote in the paperNot a substantial claim3 of 3 sources checked
    • cites [10] Hirst JA McLellan JH Price CP English E Feakins BG Stevens RJ Farmer AJ Performance of point-of-care HbA1c test devices: Implications for use in clinical practice-A systematic review and meta-analysis Clin Chem Lab Med 55 167 180 2017 10.1515/cclm-2016-0303 27658148 source
    • cites [36] Sacks DB Kirkman MS Little RR Point-of-care HBA1c in clinical practice: Caveats and considerations for optimal use Diabetes Care 47 1104 1110 2024 10.2337/dci23-0040 38552140 PMC11208753 source

      The section ‘POC A1C Testing Is Less Accurate Than Laboratory Testing’ summarizes a meta-analysis of device bias, but does not report a DCA-specific trend over time, a coefficient of 0.01% per year, P=0.081, or temporal results for other devices. Searches for those terms found no passage addressing the claimed temporal analysis. (LLM, conf=0.70)

      Experimental — accuracy not yet measured
    • cites [37] Little RR Rohlfing C Sacks DB The national glycohemoglobin standardization program: Over 20 years of improving hemoglobin A1c measurement Clin Chem 65 839 848 2019 10.1373/clinchem.2018.296962 30518660 PMC6693326 source

      The “Status of HbA1c Measurement” section reports that mean bias decreased over time for most methods, supporting the broad observation of improvement. It does not report a DCA-specific trend, the coefficient or P value, or a finding that other devices had no significant temporal effect; the cited work is a review of standardization rather than the specified meta-analysis. (LLM, conf=0.82)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    Across all subgroup analyses, heterogeneity remained consistently high, indicating that stratification by a single study-level characteristic did not materially reduce between-study variability.

    “Across all subgroup analyses, heterogeneity remained consistently high, indicating that stratification by a single study-level characteristic did not materially reduce between-study variability.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Study setting explained the largest share of between-study variance (R2=20.6%), with decentralized testing associated with higher positive bias (borderline, P=0.067).

    “As a result, study setting explained the largest share of between-study variance (R2=20.6%), with decentralized testing associated with higher positive bias (borderline, P=0.067).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In trim-and-fill sensitivity analyses, no studies were imputed (imputed=0), and the pooled estimate remained unchanged (observed MD=0.01% vs. adjusted MD=0.01%).

    “In trim-and-fill sensitivity analyses, no studies were imputed (imputed=0), and the pooled estimate remained unchanged (observed MD=0.01% vs. adjusted MD=0.01%; data not shown), supporting the robustness of the primary findings.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The 95% CIs consistently crossed zero, and τ2 and I2 remained high across leave-one-out iterations.

    “The 95% CIs consistently crossed zero, and τ2 and I2 remained high across iterations, indicating that no individual study drove the observed heterogeneity.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Small study effects were assessed using Egger's test, funnel plots, and trim-and-fill procedures.

    “Small study effects were assessed using Egger's test, complemented by visual inspection of funnel plots and, where applicable, trim-and-fill procedures, with results interpreted cautiously in the context of substantial heterogeneity (24).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [24] Egger M Davey Smith G Schneider M Minder C Bias in meta-analysis detected by a simple, graphical test BMJ 315 629 634 1997 10.1136/bmj.315.7109.629 9310563 PMC2127453 source

  • May combine two claims.

    Interpretationown data

    The RVE-adjusted pooled mean difference was directionally and quantitatively consistent with the primary random-effects model.

    “The RVE-adjusted pooled mean difference was 0.03% HbA1c (95% CI, -0.11 to 0.16%; P=0.683; data not shown), which was directionally and quantitatively consistent with the primary random-effects model.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    A leave-one-out influence analysis was performed to evaluate the impact of individual studies on the pooled estimate.

    “A leave-one-out influence analysis was performed to evaluate the impact of individual studies on the pooled estimate (23).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [23] Viechtbauer W Cheung MW Outlier and influence diagnostics for meta-analysis Res Synth Method 1 112 125 2010 10.1002/jrsm.11 26061377 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductown data

    Two reviewers independently screened titles, abstracts, and full texts, with disagreements resolved by discussion or third reviewer.

    “Two reviewers (YW and MG) independently screened titles, abstracts, and full texts. Disagreements were resolved by discussion or, when necessary, by consultation with a third reviewer (ZH).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The study protocol was prospectively registered in PROSPERO (registration no. CRD420251063197).

    “The study protocol was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO, registration no. CRD420251063197).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Premisecitation

    POC HbA1c devices enable rapid, on-site testing using small blood samples and can be operated by non-laboratory personnel.

    “These devices enable rapid, on-site testing using small blood samples obtained via finger prick or venous sampling, and can be operated by non-laboratory personnel (7).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [7] Paknikar S Sarmah R Sivaganeshan L Welke A Rizzo A Larson K Rendell M Long-term performance of point-of-care hemoglobin A1c assays J Diabetes Sci Technol 10 1308 1315 2016 10.1177/1932296816645362 27113451 PMC5094320 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    The methodological quality of included studies was assessed using the QUADAS-2 tool.

    “The methodological quality of the included studies was assessed using the QUADAS-2 tool (Table SII), which evaluates four domains: Patient selection, index test, reference standard, and flow and timing (17).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [17] Whiting PF Rutjes AW Westwood ME Mallett S Deeks JJ Reitsma JB Leeflang MM Sterne JA Bossuyt PM QUADAS-2 Group A revised tool for the quality assessment of diagnostic accuracy studies Ann Intern Med 155 529 536 2011 10.7326/0003-4819-155-8-201110180-00009 22007046 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    Data extraction was independently performed by two reviewers using a standardized template, with discrepancies resolved through discussion.

    “Data extraction was independently performed by two reviewers (YW and MG) using a standardized template. Any discrepancies were resolved through discussion.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    Hartung-Knapp adjustments were applied to all pooled confidence intervals to obtain robust inference under substantial heterogeneity.

    “To obtain robust inference under substantial heterogeneity, Hartung-Knapp adjustments were applied to all pooled confidence intervals.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductcitation

    Robust variance estimation (RVE) was used as a sensitivity analysis to address within-study correlation.

    “Furthermore, for studies which contributed more than one comparison, within study correlation was addressed using robust variance estimation (RVE) as a sensitivity analysis (22).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [22] Hedges LV Tipton E Johnson MC Robust variance estimation in meta-regression with dependent effect size estimates Res Synth Method 1 39 65 2010 10.1002/jrsm.5 26056092 source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    HbA1c reflects average plasma glucose over the preceding 2 to 3 months.

    “HbA1c remains the gold standard for assessing glycemic control, as it reflects average plasma glucose over the preceding 2 to 3 months (5).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [5] Bennett CM Guo M Dharmage SC HbA1c as a screening tool for detection of Type 2 diabetes: A systematic review Diabet Med 24 333 343 2007 10.1111/j.1464-5491.2007.02106.x 17367307 source

  • The quote only partly states this claim.

    Interpretationcitation

    The present study's findings reinforce the improvement in bias over time.

    “The findings of the present study reinforce this improvement and underscore the importance of device-specific evaluation, particularly for platforms such as lateral-flow chromatographic assay.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    The present study used a comprehensive search strategy.

    “The present study has several strengths, including a comprehensive search strategy, strict inclusion criteria requiring HPLC as the reference standard, and the use of standardized agreement metrics.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The number of included studies was relatively small, and several devices were represented by only one or two datasets, limiting statistical power and generalizability.

    “The number of included studies was relatively small, and several devices were represented by only one or two datasets, limiting statistical power and generalizability.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Venous sampling improves methodological consistency and reduces pre-analytical variation, but it may underestimate real-world variability.

    “While venous sampling improves methodological consistency and reduces pre-analytical variation, it may underestimate real-world variability.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Ion-exchange method and capillary electrophoresis can provide variant information on the result graph, which provides an essential clinical application.

    “Notably, ion-exchange method and capillary electrophoresis can provide variant information on the result graph, which provides an essential clinical application (49).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [49] Klingenberg O Furuset T Hestbråten CR Hallberg MH Steiro A Orset IR Berg JP HbA1c analysis by capillary electrophoresis-Comparison with chromatography and an immunological method Scand J Clin Lab Invest 77 458 464 2017 10.1080/00365513.2017.1338747 28644050 source

  • Classified as probably not a substantial claim.

    Premisecitation

    The clinically acceptable margin for HbA1c measurement is ±0.5%.

    “spanning the clinically acceptable margin of ±0.5% (35).”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [35] Lenters-Westra E Slingerland RJ Hemoglobin A1c point-of-care assays; a new world with a lot of consequences! J Diabetes Sci Technol 3 418 423 2009 10.1177/193229680900300303 20144277 PMC2769862 source

  • May combine two claims.

    Interpretationown data

    The 95% PI around the pooled estimate spanned nearly the full ±0.5% clinically acceptable margin and extended across the 6.5% diagnostic cut-off.

    “Notably, the 95% PI around the pooled estimate spanned nearly the full ±0.5% clinically acceptable margin and extended across the 6.5% diagnostic cut-off.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The present study used strict inclusion criteria requiring HPLC as the reference standard.

    “The present study has several strengths, including a comprehensive search strategy, strict inclusion criteria requiring HPLC as the reference standard, and the use of standardized agreement metrics.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Boronate affinity chromatography is generally robust and less affected by hemoglobin variants, but its broader detection scope theoretically includes multiple glycated species beyond the IFCC-defined β-N-terminal glycation.

    “Although this method is generally robust and less affected by hemoglobin variants, its broader detection scope theoretically includes multiple glycated species beyond the IFCC-defined β-N-terminal glycation (38,39).”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [38] Fluckiger R Woodtli T Berger W Quantitation of glycosylated hemoglobin by boronate affinity chromatography Diabetes 33 73 76 1984 10.2337/diab.33.1.73 6690345 source
    • cites [39] Nuttall FQ Comparison of percent total GHb with percent HbA1c in people with and without known diabetes Diabetes Care 21 1475 1480 1998 10.2337/diacare.21.9.1475 9727894 source

  • May combine two claims.

    Premisecitation

    Lateral-flow formats offer advantages in portability and ease of use, but they are generally less precise due to susceptibility to flow dynamics, operator variability, and lower analytical sensitivity.

    “While lateral-flow formats offer advantages in portability and ease of use, they are generally less precise due to their susceptibility to flow dynamics, operator variability, and lower analytical sensitivity (41).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [41] Liu Y Zhan L Qin Z Sackrison J Bischof JC Ultrasensitive and highly specific lateral flow assays for point-of-care diagnosis ACS Nano 15 3593 3611 2021 10.1021/acsnano.0c10035 33607867 source

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    Alternative markers such as glycated albumin or fructosamine may be more appropriate than HbA1c in individuals with suspected or known hemoglobinopathies.

    “Therefore, in individuals with suspected or known hemoglobinopathies, HbA1c should be interpreted with caution, and alternative markers such as glycated albumin or fructosamine may be more appropriate.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    QUADAS-2 is more appropriate for method-comparison studies focusing on analytical agreement.

    “which is more appropriate for method-comparison studies focusing on analytical agreement.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [17] Whiting PF Rutjes AW Westwood ME Mallett S Deeks JJ Reitsma JB Leeflang MM Sterne JA Bossuyt PM QUADAS-2 Group A revised tool for the quality assessment of diagnostic accuracy studies Ann Intern Med 155 529 536 2011 10.7326/0003-4819-155-8-201110180-00009 22007046 source

  • May combine two claims.

    Study design and conductown data

    Subgroup analyses were conducted, and pooled estimates remained consistent across analytic approaches, supporting the robustness of the findings.

    “Subgroup analyses were conducted, and pooled estimates remained consistent across analytic approaches, supporting the robustness of our findings.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The present study used standardized agreement metrics.

    “The present study has several strengths, including a comprehensive search strategy, strict inclusion criteria requiring HPLC as the reference standard, and the use of standardized agreement metrics.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing). May combine two claims.

    Interpretationreasoning

    A ±0.5% difference may be acceptable for population-level agreement, but such variation could accumulate over time and influence longitudinal monitoring, particularly when clinical management relies on small changes in HbA1c.

    “In addition, while a ±0.5% difference may be acceptable for population-level agreement, such variation could accumulate over time and influence longitudinal monitoring, particularly when clinical management relies on small changes in HbA1c.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationown data

    Substantial heterogeneity across studies (I2 ~99.0%) indicates that the pooled estimate represents an average effect across diverse devices and settings rather than evidence of analytical interchangeability.

    “However, the substantial heterogeneity across studies (I2 ~99.0%; Table SIV) indicates that the pooled estimate represents an average effect across diverse devices and settings rather than evidence of analytical interchangeability.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The analysis focused primarily on analytical agreement rather than diagnostic accuracy, clinical outcomes, or cost-effectiveness.

    “Additionally, our analysis focused primarily on analytical agreement rather than diagnostic accuracy, clinical outcomes, or cost-effectiveness, which are critical for real-world adoption.”
    Quote in the paperSubstantial claim
45 of 45

  • The authors state this as an assumption.

    For studies where only the 95% CI for the mean difference was available, the SD was back-calculated assuming a normal distribution and known sample size.

    “assuming normal distribution and known sample size”

  • The paper relies on this without support.

    The 11 included studies and the POC devices evaluated are representative of contemporary POC HbA1c devices in routine care.

    “suggesting that contemporary POC devices can provide clinically usable HbA1c measurements under appropriate conditions.”
2 of 2

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

This check didn’t finish.

Cites

Flagged first, then by citations

1 of the 49 references couldn’t be matched to a known paper.

Cited by

Sorted by citations

Authors
Wei, Y., Gu, M., Ren, Y., & Hu, Z.
Affiliations
Shanghai Zhongqiao Vocational and Technical University (CN)
Venue
Biomedical Reports
Publisher
Spandidos Publications
Published
2026-02-13
Volume
24
Issue
4
Pages
1-10
ISSN
  • 2049-9434 (ISSN-L)
  • 2049-9442
DOI
10.3892/br.2026.2120
PMC
PMC12977185
OpenAlex
W7128773306
Semantic Scholar
a0f6435a21c68877bc43218a9fa802340e1bc117
Cited by
3
References
49
Citations by year
2026: 3
FWCI
5.90
Type
journal article
Language
en
Open access
Licence
cc-by-nc-nd
Locations
  • Biomedical Reports (publishedVersion, cc-by-nc-nd)
  • PubMed (publishedVersion)
  • PubMed Central (submittedVersion, cc-by-nc-nd)
Topics
  • Health Sciences › Medicine › Physiology › Clinical Laboratory Practices and Quality Control
  • Health Sciences › Medicine › Endocrinology, Diabetes and Metabolism › Diabetes Management and Research
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Clinical Biochemistry › Advanced Glycation End Products research
Keywords
HbA1c, point-of-care testing, measurement agreement, diabetes monitoring, glycated hemoglobin measurement, lateral flow immunoassay, measurement bias, device validation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-10-17.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-17.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review