CoChat
Sign Up
Cited by 6Treatment of Major DepressionMental Health Research TopicsAnxiety, Depression, Psychometrics, Treatment, Cognitive Processes

Predicting Relapse of Depressive Episodes During Maintenance Treatment: The Canadian Biomarker Integration Network in Depression (CAN-BIND) Wellness Monitoring in Major Depressive Disorder Study: Prédire la rechute d’épisodes dépressifs pendant le traitement d’entretien : Une étude de suivi du bien-être dans les troubles dépressifs majeurs du Réseau canadien d’intégration des biomarqueurs pour la dépression (CAN-BIND)

Open access
Lam, Raymond W., Rnic, Katerina, Nunez, John-Jose, Ho, Keith, LeMoult, Joelle, Nunes, Abraham, Chakrabarty, Trisha, Foster, Jane A. +12 more·2025·The Canadian Journal of Psychiatry
DOI 10.1177/07067437251337603PMC PMC12043620
Open this paper in CoChat — free account

Abstract

Background: Relapse rates in major depressive disorder (MDD) remain high even after treatment to remission. Identifying predictors of relapse is, therefore, crucial for improving maintenance strategies and preventing future episodes. Remote data collection and sensing technologies may allow for more comprehensive and longitudinal assessment of potential predictors. Methods: The Canadian Biomarker Integration Network in Depression Wellness Monitoring for MDD (CBN-WELL) study was a prospective, multicentre observational study with an aim to identify biomarkers associated with relapse in patients on maintenance treatment for MDD. Participants had a DSM-5-TR diagnosis of MDD in remission and a Montgomery–Åsberg Depression Rating Scale (MADRS) score ≤14. Participants remained on their baseline medication regimens and were followed bimonthly for up to 2 years. Relapse criteria included MADRS > 22 for 2 consecutive weeks, suicidality or hospitalization, and initiation or change in medication for worsening symptoms. Data collection included clinical assessments, self-report questionnaires, and remote monitoring using wrist-worn actigraphs and smartphones. Results: A total of 96 participants had follow-up data. Of these, 28.9% experienced a depressive relapse during the study period, with an average time to relapse of 211 days. Baseline depressive severity, as measured by MADRS, was higher in participants who relapsed compared to those who did not, but few other baseline clinical measures differentiated these groups. Conclusions: Individuals with MDD in remission continued to have high relapse rates despite maintenance treatment. The paucity of clinical factors that predict relapse underscores the need for biomarkers. The CBN-WELL database can be used for future research to integrate multiple predictive factors and to identify objective measures to predict relapse in individuals.

AI snapshot

Written from the paper’s full text.

TL;DR

In the CBN-WELL prospective observational study, 28.9% of 96 participants with major depressive disorder in remission relapsed during maintenance treatment, with an average time to relapse of 211 days. Few baseline clinical measures differentiated relapsers from nonrelapsers, supporting the need for biomarker-based prediction.

Key findings

  • Among 96 participants with follow-up data, 28 individuals (29.2%; abstract/discussion report 28.9%) experienced a depressive relapse, with an average time to relapse of 211 days (SD = 121).
  • Baseline depressive severity was higher in the relapse group (MADRS M = 6.89, SD = 4.28) than in nonrelapse without residual symptoms (M = 3.90, SD = 2.84) and nonrelapse with residual symptoms (M = 4.50, SD = 3.16); small group differences also appeared for QIDS-SR and SDS.
  • Residual depressive symptoms during follow-up did not distinguish relapsers from nonrelapsers: 39.3% of the relapse group vs. 46.7% of the nonrelapse group had mild residual symptoms at a study visit (χ² = 0.03, p = 0.86).
  • Relapse criteria met were criterion 1a in 16/28 (57%), criterion 2 in 1/28 (4%), criterion 3 in 2/28 (7%), and criterion 5 in 9/28 (32%).

Methods

  • Prospective, multicentre, longitudinal observational study (CBN-WELL; NCT02934334) conducted July 2016–January 2019 at 5 Canadian centres, enrolling outpatients aged 18–65 with DSM-5 MDD in remission and MADRS ≤14.
  • Participants remained on baseline medication regimens and were followed bimonthly for up to 2 years, with clinical assessments, self-report questionnaires, smartphone self-reports, wrist-worn actigraphy, and relapse verification/adjudication.
  • Analyses compared demographic and clinical variables across relapse, nonrelapse, and dropout groups using ANOVA and chi-square tests; Kaplan–Meier survival analysis visualized time to relapse.

Limitations

  • Stated: the sample was heterogeneous in remission duration and treatment type, which may make a common predictor harder to identify but reflects a real-world population.
  • Stated: relapse criteria were broad and included medication changes for worsening symptoms; some participants may not have met full relapse criteria with watchful waiting.
  • Evident from the text: the outcome analysis included 96 participants, and the study reports few baseline clinical predictors, so it cannot establish causal predictors or rule out unmeasured or combined longitudinal factors.

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

Integrity signals

Registry

Crossref · 10.1177/07067437251337603

Cites

49 of 49 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

0 of 42 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    Except for small differences in baseline MADRS, QIDS-SR, and SDS (all within the remission range), no baseline clinical characteristics differentiated those who relapsed from those who did not.

    “Except for a small difference in the baseline scores in the MADRS, QIDS-SR, and SDS (all of which were within the remission range), there were no baseline clinical characteristics that differentiated those who relapsed from those who did not.”
    Quote in the paperSubstantial claim
  • Findingown data

    There was no difference between the relapse and nonrelapse groups in the proportion of participants having mild residual depressive symptoms at a study visit (39.3% vs. 46.7%, χ2 = 0.03, p = 0.86).

    “There was no difference between the relapse and nonrelapse groups in the proportion of participants having mild residual depressive symptoms at a study visit (39.3% vs. 46.7%, respectively; χ2 = 0.03, p = 0.86).”
    Quote in the paperNot a substantial claim
  • Findingown data

    There were small overall differences among the groups in QIDS-SR (F(3,96) = 4.55, p = 0.015) and SDS (F(3,96) = 3.97, p = 0.02).

    “Similarly, there were small overall differences among the groups in the QIDS-SR (F(3, 96) = 4.55, p = 0.015) and SDS (F(3, 96) = 3.97, p = 0.02).”
    Quote in the paperNot a substantial claim
  • Findingown data

    There was a significant group difference in baseline MADRS scores among the three participant groups (F(3,96) = 4.71, p = 0.004).

    “there was a significant group difference in baseline depressive symptom severity as assessed with the MADRS (F(3, 96) = 4.71, p = 0.004)”
    Quote in the paperNot a substantial claim
  • Findingown data

    Many participants with residual symptoms in this study did not have a subsequent relapse event.

    “but many participants with residual symptoms in this study did not have a subsequent relapse event”
    Quote in the paperSubstantial claim
  • Classified as probably not a substantial claim. 6
  • May combine two claims. 12

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

    Findingown data

    The nonrelapse group was followed for an average of 451 days (SD = 203).

    “Those in the nonrelapse group were followed for an average of 451 days (SD = 203).”
    Quote in the paperNot a substantial claim

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

    Findingown data

    There were no other significant differences among groups (all ps ≥ 0.470).

    “There were no other significant differences among groups (all ps ≥ 0.470).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The study sample includes a diverse group of participants with MDD who were in variable periods of remission after treatment with a range of depression interventions and who were on different medication regimens during maintenance.

    “our study sample includes a diverse group of participants with MDD who were in variable periods of remission after treatment with a range of depression interventions, including medications, psychotherapy, and neurostimulation, and who were on different medication regimens during maintenance.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The criteria for relapse met by participants were: criterion 1a (n = 16, 57%), criterion 2 (n = 1, 4%), criterion 3 (n = 2, 7%), and criterion 5 (n = 9, 32%).

    “The criteria for relapse met by the participants were: criterion 1a (n = 16 out of 28, 57%), criterion 2 (n = 1, 4%), criterion 3 (n = 2, 7%), and criterion 5 (n = 9, 32%).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    The relapse rate in this cohort is similar to those reported in other clinical settings and in meta-analyses of RCTs, which have shown relapse rates of 21% to 49%.

    “This relapse rate is similar to those reported in other clinical settings7,22 and in meta-analyses of RCTs,6,8 which have shown that those who remain on antidepressants following remission had relapse rates of 21% to 49%.”
    Quote in the paperNot a substantial claim0 of 4 sources checked
    • cites [7] Hardeveld F Spijker J De Graaf R, et al. Recurrence of major depressive disorder across different treatment settings: results from the NESDA study. J Affect Disord. 2013; 147(1–3): 225- 231. 23218899 source

      Not checked in the free run.

    • cites [22] Lye MS Tey YY Tor YS, et al. Predictors of recurrence of major depressive disorder. PLoS One. 2020; 15(3): e0230363. PMC7082055 32191745 source

      Not checked in the free run.

    • cites [6] Gueorguieva R Chekroud AM Krystal JH. Trajectories of relapse in randomised, placebo-controlled trials of treatment discontinuation in major depressive disorder: an individual patient-level data meta-analysis. Lancet Psychiatry. 2017; 4(3): 230- 237. 28189575 PMC5340978 source

      Not checked in the free run.

    • cites [8] Kato M Hori H Inoue T I, et al. Discontinuation of antidepressants after remission with antidepressant medication in major depressive disorder: a systematic review and meta-analysis. Mol Psychiatry. 2021; 26(1): 118- 133. 32704061 PMC7815511 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationreasoning

    The alignment of the CBN-WELL protocol with other observational studies allows for pooling of data to increase power, or for using one study as an independent external validation sample for predictors found in another study.

    “This allows for pooling of data to increase power, or for using one study as an independent external validation sample for predictors found in another study.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premisecitation

    Remote recording tools via the internet or smartphones allow patients to complete self-assessments at home and reduce participant burden and the need for frequent clinic visits.

    “Remote recording tools via the internet or smartphones allow patients to complete self-assessments at home27,28 and reduce participant burden and the need for frequent clinic visits.”
    Quote in the paperNot a substantial claim0 of 2 sources checked
    • cites [27] He Y Wang X Wang Z, et al. Comparison of the efficacy between standard measurement-base care (MBC) and enhanced MBC for major depressive disorder: a pilot study. Neuropsychiatr Dis Treat. 2024; 20: 1465- 1473. 39100573 PMC11296504 source

      Not checked in the free run.

    • cites [28] Li VW Sahota J Dev DK, et al. A randomized evaluation of MoodFX, a patient-centred e-Health tool to support outcome measurement for depression. Can J Psychiatry. 2024; 69(7): 493- 502. 38600892 PMC11168344 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    The authors have developed a method for longitudinal long-term tracking of actigraphy data using an open-source data processing pipeline that accounts for missing data, nonwear periods, and sleep-wake scoring, making it suitable for long-term relapse prediction studies.

    “We have developed a method for longitudinal long-term tracking of actigraphy data using an open-source data processing pipeline that accounts for issues such as missing data, nonwear periods, and sleep-wake scoring, 44 making it suitable for long-term relapse prediction studies.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [44] Slyepchenko A Uher R Ho K, et al. A standardized workflow for long-term longitudinal actigraphy data processing using one year of continuous actigraphy from the CAN-BIND Wellness Monitoring Study. Sci Rep. 2023; 13(1): 15300. 37714910 PMC10504311 source

      Not checked in the free run.

  • May combine two claims.

    Premisecitation

    After treatment, a higher risk of relapse has been associated with the number and severity of previous episodes, the persistence of residual symptoms, and a history of childhood maltreatment.

    “after treatment, a higher risk of relapse has been associated with the number and severity of previous episodes, the persistence of residual symptoms and a history of childhood maltreatment.9,15,19–23”
    Quote in the paperNot a substantial claim0 of 7 sources checked
    • cites [9] Klein NS Holtman GA Bockting CLH, et al. Development and validation of a clinical prediction tool to estimate the individual risk of depressive relapse or recurrence in individuals with recurrent depression. J Psychiatr Res. 2018; 104: 1- 7. 29908389 source

      Not checked in the free run.

    • cites [15] Buckman JEJ Underwood A Clarke K, et al. Risk factors for relapse and recurrence of depression in adults and how they operate: a four-phase systematic review and meta-synthesis. Clin Psychol Rev. 2018; 64: 13- 38. 30075313 PMC6237833 source

      Not checked in the free run.

    • cites [19] Berwian IM Walter H Seifritz E, et al. Predicting relapse after antidepressant withdrawal – a systematic review. Psychol Med. 2017; 47: 426- 437. 27786144 PMC5244448 source

      Not checked in the free run.

    • cites [20] Duffy L Lewis G Marston L, et al. Clinical factors associated with relapse in depression in a sample of UK primary care patients who have been on long-term antidepressant treatment. Psychol Med. 2024; 54(5): 951- 961. 37753652 source

      Not checked in the free run.

    • cites [21] Hardeveld F Spijker J De Graaf R, et al. Prevalence and predictors of recurrence of major depressive disorder in the adult population. Acta Psychiatr Scand. 2010; 122(3): 184- 191. 20003092 source

      Not checked in the free run.

    • cites [22] Lye MS Tey YY Tor YS, et al. Predictors of recurrence of major depressive disorder. PLoS One. 2020; 15(3): e0230363. PMC7082055 32191745 source

      Not checked in the free run.

    • cites [23] Wojnarowski C Firth N Finegan M, et al. Predictors of depression relapse and recurrence after cognitive behavioural therapy: a systematic review and meta-analysis. Behav Cogn Psychother. 2019; 47: 514- 529. 30894231 source

      Not checked in the free run.

  • May combine two claims.

    Interpretationcitation

    While individual baseline measures may not be accurate predictors of relapse, combinations, longitudinal changes, or relationships between several measures may be more informative.

    “While individual baseline measures may not be accurate predictors of relapse, combinations, longitudinal changes, or relationships between several measures may be more informative. 42”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [42] Moriarty AS Meader N Snell KIE, et al. Predicting relapse or recurrence of depression: systematic review of prognostic models. Br J Psychiatry. 2022; 221(2): 448- 458. 35048843 source

      Not checked in the free run.

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

    Premisecitation

    Each relapse/recurrence is associated with an increased risk of future recurrence, as well as reduced quality-of-life and functioning.

    “Each relapse/recurrence is associated with an increased risk of future recurrence, 11 as well as reduced quality-of-life 12 and functioning. 13”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [11] Solomon DA Keller MB Leon AC, et al. Multiple recurrences of major depressive disorder. Am J Psychiatry. 2000; 157(2): 229- 233. 10671391 source

      Not checked in the free run.

    • cites [12] Ishak WW Mirocha J James D, et al. Quality of life in major depressive disorder before/after multiple steps of treatment and one-year follow-up. Acta Psychiatr Scand. 2015; 131(1): 51- 60. 24954156 PMC4267902 source

      Not checked in the free run.

    • cites [13] Schwarz R Munkholm K Christensen MS, et al. Functioning in patients with major depressive disorder in remission: a systematic review and meta-analysis. J Affect Disord. 2024; 363: 112- 123. 39025440 source

      Not checked in the free run.

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

    Premisereasoning

    These data can now be reliably recorded and may be important factors in depressive relapse in addition to, or combined with, clinical information.

    “These data can now be reliably recorded and may be important factors in depressive relapse in addition to, or combined with clinical information.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    All relapses were adjudicated using the study criteria and verified, prior to any analysis, by consensus of a panel of investigators using all available data.

    “All relapses were adjudicated using these criteria and verified, prior to any analysis, by consensus of a panel of investigators (CNS, KH, RWL, RVM, and RU) using all available data.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Premiseown data

    Greater self-depressed associations at baseline and longitudinal increases across follow-up predicted a shorter time to relapse in this cohort (DIAT).

    “We found that greater self-depressed associations at baseline and longitudinal increases across follow-up predicted a shorter time to relapse in this cohort. 46”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [46] Rnic K LeMoult J Torres IJ, et al. Predicting recurrence of major depressive episodes using the Depression Implicit Association Test: a Canadian biomarker integration network in depression (CAN-BIND) report. Psychiatry Res. 2023; 330: 115606. 37979318 source

      Not checked in the free run.

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

    Study design and conductnone given

    Ethics approval was obtained from institutional review boards at each site and all participants provided written informed consent.

    “Ethics approval was obtained from institutional review boards at each site and all participants provided written informed consent.”
    Quote in the paperNot a substantial claim
15 of 15

  • The authors state this as an assumption.

    Relapse and recurrence cannot be empirically differentiated, so the term relapse is used to indicate both.

    “because the definitions vary widely and there is no empirical evidence to differentiate between the original and a new depressive episode,14–16 we use the term relapse to indicate both relapse and recurrence.”
1 of 1

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • The second criterion with which most authors agree is antidepressant treatment, with some mentioning it in general and others specifying change, upward adjustment, review, or potentiation.

    “The second criterion with which most authors agree is that related to antidepressant treatment, although some mention it in general (27, 33) and others specify whether they consider a change of antidepressant (8, 15, 26, 34, 40, 42, 48, 50), an upward adjustment (10, 17, 40, 45), a review of the treatment (24) or a potentiation (17).”

    From Administrative criteria in the relapse of depressive disorder: a critical appraisal (2026) · cites this paper as [48]

    The Methods > Relapse Definition passage (page not provided) includes medication change suggestive of worsening depression as a relapse criterion, with examples of increasing an antidepressant dose or adding an adjunctive agent. It does not specifically identify switching to a different antidepressant, the particular type of change for which this source is cited. (LLM, conf=0.83)

    Experimental — accuracy not yet measured

  • 4 studies use the Clinical Global Impression (CGI-S), 2 use the Beck's Depression Inventory, and 1 uses the Goldberg Anxiety and Depression Scale (GADS).

    “Finally, 4 studies use the Clinical Global Impression (CGI-S) (15, 33, 43, 48), 2 use the Beck`s Depression Inventory (4, 9) and 1 uses the Goldberg Anxiety and Depression Scale (GADS) (20).”

    From Administrative criteria in the relapse of depressive disorder: a critical appraisal (2026) · cites this paper as [48]

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • ConsistentF(3, 96) = 4.71reported p = .004 · recomputed p = .004

    “F(3, 96) = 4.71, p = .004”

  • ConsistentF(3, 96) = 4.55reported p = .015 · recomputed p = .005

    “F(3, 96) = 4.55, p = .015”

    The paper states a correction for multiple tests

  • ConsistentF(3, 96) = 3.97reported p = .02 · recomputed p = .010

    “F(3, 96) = 3.97, p = .02”

    The paper states a correction for multiple tests

3 of 3

Self-check

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

Each contradiction between two passages listed here was confirmed by an independent check.

  • Baseline MADRS mean differs between the table and results

    Major

    For the nonrelapse group without residual symptoms, Table 1 gives a baseline MADRS mean of 6.63, whereas the results describe the same group as having a mean of 3.90. This changes the stated comparison underlying the baseline-severity finding.

  • Abstract and methods disagree on the relapse threshold

    Minor

    The abstract specifies MADRS greater than 22, while the methods specify MADRS greater than or equal to 22. A score of exactly 22 is treated differently by these statements.

  • Associate-degree percentage is inconsistent with its group size

    Minor

    The nonrelapse group without residual symptoms has 40 participants. Six participants with an associate degree therefore represent 15.0%, not the reported 25.0%.

    Reported 25 · recomputed 15

  • Female total does not equal the subgroup counts

    Minor

    The three outcome groups contain 23, 17, and 19 female participants, totaling 59 rather than the 62 listed in Table 1's total column.

    Reported 62 · recomputed 59

  • Three-group comparison is reported with four-group degrees of freedom

    Minor

    The results call this a comparison of three participant groups, but report F(3, 96). A three-group ANOVA has a numerator degree of freedom of 2, not 3.

  • Conclusion claims an anxiety-score difference not found in the results

    Minor

    The conclusion says baseline anxiety scores differentiated participants who relapsed, whereas the results identify differences in MADRS, QIDS-SR, and SDS and state that there were no other significant group differences.

  • Residual-symptom percentage uses the wrong denominator

    Minor

    Among 68 nonrelapse participants, 28 had mild residual symptoms, which is 41.2%, not 46.7%. The latter percentage is repeated in the discussion.

    Reported 46.7 · recomputed 41.176471

  • Residual-symptom definitions disagree at MADRS 22

    Minor

    The methods exclude a score of exactly 22 from mild residual symptoms, but the results include it.

8 of 8

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Lam, R. W., Rnic, K., Nunez, J.-J., Ho, K., LeMoult, J., Nunes, A., Chakrabarty, T., Foster, J. A., Frey, B. N., Harkness, K. L., Hassel, S., Kennedy, S. H., Li, Q. S., Milev, R. V., Quilty, L. C., Rotzinger, S., Soares, C. N., Taylor, V. H., Turecki, G., & Uher, R.
Affiliations
  • University of British Columbia (CA)
  • Department of Psychiatry, University of British Columbia, Vancouver, Canada
  • University of Toronto (CA)
  • Dalhousie University (CA)
  • The University of Texas Southwestern Medical Center (US)
  • McMaster University (CA)
  • Queen's University (CA)
  • University of Calgary (CA)
  • Neuroscience, Janssen Research & Development, LLC, Titusville, NJ, USA
  • Department of Psychiatry, Queen's University, Kingston, Canada
  • McGill University (CA)
Corresponding author
Lam, Raymond W.
ORCID
Venue
The Canadian Journal of Psychiatry
Publisher
SAGE Publications
Published
2025-04-29
Volume
70
Issue
7
Pages
565-573
ISSN
  • 0706-7437 (ISSN-L)
  • 1497-0015
DOI
10.1177/07067437251337603
PMC
PMC12043620
OpenAlex
W4409973087
Cited by
6
References
49
Citations by year
2026: 5 · 2025: 1
FWCI
3.39
Type
journal article
Language
en
Open access
Licence
Locations
  • The Canadian Journal of Psychiatry (publishedVersion, cc-by-nc)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by-nc)
Funders
  • Ontario Brain Institute (n/a)
  • Janssen Pharmaceutica (n/a)
Topics
  • Health Sciences › Medicine › Pharmacology › Treatment of Major Depression
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Mental Health Research Topics
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Anxiety, Depression, Psychometrics, Treatment, Cognitive Processes
Keywords
major depressive disorder, recurrence prediction, maintenance therapy, biomarkers, remote monitoring, actigraphy, smartphone data, wellness monitoring, remission, relapse risk factors
MeSH terms
Adult, Antidepressive Agents, Canada, Major Depressive Disorder, Depressive Disorder, Major, Female, Humans, Male, Middle Aged, Prospective Studies, Recurrence, Biomarkers

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 2026-10-10.

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

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