CoChat
Sign Up
Cited by 3,617COVID-19 and Mental HealthHealthcare professionals’ stress and burnoutMaternal Mental Health During Pregnancy and Postpartum

Prevalence of depression, anxiety, and insomnia among healthcare workers during the COVID-19 pandemic: a systematic review and meta-analysis

Open access
Pappa, Sofia, Ntella, Vasiliki, Giannakas, Timoleon, Giannakoulis, Vassilis G., Papoutsi, Eleni, Katsaounou, Paraskevi·2020·Brain Behavior and Immunity
DOI 10.1016/j.bbi.2020.05.026PMC PMC7206431
Open this paper in CoChat — free account

Abstract

Background COVID-19 pandemic has the potential to significantly affect the mental health of healthcare workers (HCWs), who stand in the frontline of this crisis. It is, therefore, an immediate priority to monitor rates of mood, sleep and other mental health issues in order to understand mediating factors and inform tailored interventions. The aim of this review is to synthesize and analyze existing evidence on the prevalence of depression, anxiety and insomnia among HCWs during the Covid-19 outbreak.Methods A systematic search of literature databases was conducted up to April 17th, 2020. Two reviewers independently assessed full-text articles according to predefined criteria. Risk of bias for each individual study was assessed and data pooled using random-effects meta-analyses to estimate the prevalence of specific mental health problems. The review protocol is registered in PROSPERO and is available online.Findings Thirteen studies were included in the analysis with a combined total of 33,062 participants. Anxiety was assessed in 12 studies, with a pooled prevalence of 23·2% and depression in 10 studies, with a prevalence rate of 22·8%. A subgroup analysis revealed gender and occupational differences with female HCPs and nurses exhibiting higher rates of affective symptoms compared to male and medical staff respectively. Finally, insomnia prevalence was estimated at 38·9% across 5 studies.Interpretation Early evidence suggests that a considerable proportion of HCWs experience mood and sleep disturbances during this outbreak, stressing the need to establish ways to mitigate mental health risks and adjust interventions under pandemic conditions.

AI snapshot

Written from the paper’s full text.

TL;DR

This systematic review and meta-analysis pooled 13 cross-sectional studies of healthcare workers during the COVID-19 pandemic and found substantial pooled prevalence of anxiety, depression, and insomnia. It reports roughly one in four workers with anxiety or depression and about one in three with insomnia, with higher estimates among women and nursing staff.

Key findings

  • Pooled prevalence was 23.21% for anxiety (95% CI 17.77–29.13, I² = 99%; 12 studies), 22.8% for depression (95% CI 15.1–31.51, I² = 99.62%; 10 studies), and 34.32% for insomnia (95% CI 27.45–41.54, I² = 98%; 5 studies).
  • Subgroup estimates were higher for women than men for anxiety (29.06% vs 20.92%) and depression (26.87% vs 20.34%), and higher for nurses than doctors for anxiety (25.80% vs 21.73%) and depression (30.30% vs 25.37%).
  • Mild symptoms were more common than moderate/severe symptoms: anxiety 17.93% vs 6.88%; depression 24.60% vs 16.18%. Scale-specific pooled estimates varied, e.g., GAD-7 anxiety 36.92% vs SAS anxiety 16.47%, and PHQ-9 depression 36.72 vs SDS depression 32.81.

Methods

  • The review followed PRISMA and MOOSE guidance and was registered in PROSPERO (CRD42020180313). Searches covered MEDLINE, PubMed, Google Scholar, MedRxiv, and SSRN through April 17, 2020, with no language restrictions, and included cross-sectional studies of healthcare workers in COVID-19-affected areas that used validated measures for depression, anxiety, or insomnia.
  • The meta-analysis included 13 cross-sectional studies with 33,062 participants; 12 were conducted in China (2 in Wuhan) and 1 in Singapore. Median study size was 1,563 (range 134–11,118), median male representation was 18%, and median questionnaire participation was 85.3%.
  • Pooled prevalence was estimated with a random-effects model in MetaXL using double arcsine transformation. Subgroup analyses examined gender, rating scales, symptom severity, and professional group; sensitivity analyses excluded each study in turn. Study quality was assessed with a modified Newcastle-Ottawa scale.

Limitations

  • Stated in the paper: heterogeneity was high across studies because of different assessment scales and cut-offs; some studies may have sampled overlapping populations from the same region or country; most studies were from China, limiting generalizability; all included studies were cross-sectional, so long-term mental health effects remain unknown; and the number of available studies was relatively small early in the pandemic.
  • Evident from the text: the very high I² values (often near 99%) indicate substantial unexplained between-study variability, and the inclusion of preprints means some data may not have been peer-reviewed.

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

Integrity signals

Registry

Crossref · 10.1016/j.bbi.2020.05.026

Cites

48 of 48 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 15 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    The pooled prevalence of anxiety among healthcare workers during COVID-19 was 23.21% (95% CI 17.77-29.13, I2=99%) across 12 studies.

    “The pooled prevalence was 23·21% (95% CI 17·77-29·13, I2 = 99%) as presented in Fig. 2 .”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.
  • Findingown data

    When Lai et al. (2020) and Zhang et al. (2020a) were excluded, the recalculated pooled depression prevalence was 16.94% (95% CI 10.38-24.67, I2=99.56%).

    “when both were excluded, the recalculated pooled prevalence was 16·94% (95% CI, 10·38-24·67, I2 = 99·56%).”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.
  • Findingown data

    Among nine low risk of bias studies, the pooled anxiety prevalence was 24.06% (95% CI 16.84-32.09, I2=99%).

    “low risk of bias studies (n = 9) revealed a total pooled anxiety prevalence of 24·06% (95% CI 16·84-32·09, I2 = 99%).”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.
  • Findingown data

    Four studies using the GAD-7 scale had a pooled anxiety prevalence of 36.92% (95% CI 26.06-48.23, I2=99%).

    “four studies (Huang and Zhao, 2020, Lai et al., 2020, Zhang et al., 2020a, Zhu et al., 2020) used the GAD-7 scale with a pooled prevalence of 36·92% (95% CI 26·06-48·23, I2 = 99%).”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.
  • Findingown data

    The pooled prevalence of insomnia among healthcare workers during COVID-19 was 34.32% (95% CI 27.45-41.54, I2=98%) across five studies.

    “The pooled prevalence was calculated as 34·32% (95% CI 27·45–41·54, I2 = 98%).”
    Quote in the paperSubstantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.
  • May combine two claims. 17
  • Classified as probably not a substantial claim. 16

  • May combine two claims.

    Findingown data

    The review included 13 cross-sectional studies with a total of 33,062 participants.

    “After de-duplication and screening, thirteen studies (Du et al., 2020, Guo et al., 2020, Huang et al., 2020b, Huang and Zhao, 2020, Lai et al., 2020, Liu et al., 2020, Liu et al., 2020, Lu et al., 2020, Qi et al., 2020, Tan et al., 2020, Zhang et al., 2020a, Zhang et al., 2020b, Zhu et al., 2020) with a total of 33,062 participants were included in the analysis.”
    Quote in the paperSubstantial claim

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

    Findingown data

    Of the 13 included studies, 12 were conducted in China, two of them in Wuhan, and one in Singapore.

    “Out of the 13 studies, 12 were undertaken in China, two of which were in Wuhan (Lai et al., 2020, Zhu et al., 2020), while one took place in Singapore (Tan et al., 2020).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In six studies with severity data, pooled anxiety prevalence was 17.93% for mild anxiety and 6.88% for moderate/severe anxiety.

    “Regarding the severity of the anxiety, data were available in six studies with a pooled prevalence of 17·93% for mild anxiety and 6·88% for moderate/severe”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.

  • May combine two claims.

    Findingown data

    The majority of healthcare workers experienced mild symptoms of depression and anxiety, while moderate and severe symptoms were less common.

    “it appears that the majority of the HCWs experienced mild symptoms both for depression and anxiety, while moderate and severe symptoms were less common among the participants.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In five studies with doctor and nurse data, pooled depression prevalence was 30.30% for nurses and 25.37% for doctors.

    “Between doctors and nurses, in the five studies with available data on depression, the pooled prevalence was calculated as 30.30% for nurses and 25.37% for doctors”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.

  • May combine two claims.

    Findingown data

    In six studies with doctor and nurse data, pooled anxiety prevalence was 21.73% for doctors and 25.80% for nurses.

    “In doctor and nurse groups, prevalence could be calculated in six studies, with respective values of 21·73%and 25·80%”
    Quote in the paperNot a substantial claim
    • The authors state this as an assumption: One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.

  • May combine two claims.

    Findingown data

    The prevalence of anxiety and depression was higher in female healthcare workers than in male healthcare workers.

    “The prevalence rate of anxiety and depression appeared to be higher in females”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Nursing staff had higher prevalence estimates for both anxiety and depression compared with doctors.

    “nursing staff exhibited higher prevalence estimates both for anxiety and depression compared to doctors.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Interpretationcitation

    The higher prevalence of anxiety and depression in nurses may be partly confounded by nurses being mostly female, and may also be due to greater risk of exposure to COVID-19 patients through more time on wards, direct care, and sputum collection.

    “These results may be partly confounded by the fact that nurses are mostly female but could be also attributed to the fact they may face a greater risk of exposure to COVID-19 patients as they spend more time on wards, provide direct care to patients and are responsible for the collection of sputum for virus detection”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Liu et al., 2020

      Not checked in the free run.

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

    Premisecitation

    COVID-19 is severely affecting the wellbeing of healthcare professionals.

    “a number of other studies published in recent weeks provide emerging evidence that COVID-19 is severely affecting the wellbeing of healthcare professionals.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites Cheung et al., 2020

      Not checked in the free run.

    • cites Bohlken et al., 2020

      Not checked in the free run.

    • cites Xu et al., 2020

      Not checked in the free run.

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

    Study design and conductcitation

    The review was conducted in accordance with the PRISMA statement.

    “The systematic review was conducted in accordance with the PRISMA statement”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Physicians are at increased risk of suicide compared to the general population.

    “physicians are already at an increased risk of suicide compared to the general population (West et al., 2018).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites West et al., 2018 source

      Not checked in the free run.

  • May combine two claims.

    Study design and conductown data

    The included studies were all cross-sectional, so long-term implications warrant further research.

    “the studies included in our meta-analysis were all cross-sectional, thus the long-term implications of COVID-19 pandemic on HCW’s mental health warrant further research.”
    Quote in the paperSubstantial claim

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

    Study design and conductown data

    Two authors independently identified records for inclusion.

    “Two authors independently identified records published until April 17th 2020 that reported on the prevalence of depression, anxiety, and insomnia in HCWs during the coronavirus pandemic through systematically searching MEDLINE, PubMed and Google Scholar databases.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    A large proportion of healthcare workers in Wuhan were affected and mental health support was necessary even for mild psychological reactions.

    “another study performed in the original center of the epidemic, Wuhan, showed that a large proportion of HCW in Wuhan were affected and that mental health support was necessary even for mild psychological reactions (Kang et al., 2020).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Kang et al., 2020 source

      Not checked in the free run.

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

    Study design and conductreasoning

    Several studies might have included the same population because they were broadly conducted in the same region or country.

    “Another limitation is that several studies might have included the same population as they were broadly conducted in the same region/country.”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    Data were extracted independently by two reviewers.

    “The following data were extracted from each article by two reviews independently”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The MOOSE checklist was followed.

    “The MOOSE (Meta-analyses Of Observational Studies in Epidemiology) Checklist was followed.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationreasoning

    Despite limited generalizability, the findings provide a reliable indication of the potential of the COVID-19 pandemic to affect the mental health of healthcare workers because China was severely affected.

    “considering the fact that China was severely affected, they provide a reliable indication of the potential of COVID-19 pandemic to affect the mental health of HCWs.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: China was severely affected by COVID-19.

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

    Interpretationreasoning

    Differences between COVID-19 and previous outbreaks could be explained by the extremely high infectious potential and mortality rate of the former and by experience acquired in the interim.

    “Potential differences, however, between these outbreaks and the COVID-19 pandemic could be explained on the basis of the extremely high infectious potential and mortality rate of the former but also the experience acquired in the interim in these areas.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The accuracy of extracted or calculated data was confirmed by comparing the collection forms of the two investigators.

    “The accuracy of the extracted or calculated data was confirmed by comparing the collection forms of the two investigators.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductreasoning

    Different assessment scales and cut-offs were used across studies, and threshold criteria for case definition varied.

    “Different assessment scales were utilized for population screening and different cut offs set even though several studies used the same tests. Thus, threshold criteria for case definition varied”
    Quote in the paperSubstantial claim

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

    Study design and conductown data

    The review protocol was registered in PROSPERO (CRD42020180313).

    “The review protocol is registered in PROSPERO and is available online (CRD42020180313).”
    Quote in the paperNot a substantial claim

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

    Study design and conductown data

    Two authors independently evaluated risk of bias using a modified Newcastle-Ottawa scale, with disagreements resolved by a third author.

    “two authors independently evaluated the risk of bias of the included cross-sectional studies using a modified form of the Newcastle-Ottawa scale. Potential disagreements were resolved by a third author.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Interpretationown data

    The review provides a timely and comprehensive synthesis of existing evidence highlighting high prevalence rates of depression, anxiety and insomnia among healthcare professionals.

    “our systematic review and meta-analysis provide a timely and comprehensive synthesis of the existing evidence highlighting the high prevalence rates of depression, anxiety and insomnia of healthcare professionals.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning

    The review's findings can help quantify staff support needs and inform tiered and tailored interventions under pandemic conditions.

    “Findings can help to quantify staff support needs and inform tiered and tailored interventions under pandemic conditions that enhance resilience and mitigate vulnerability.”
    Quote in the paperNot a substantial claim

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

    Premisecitation

    Level of social support positively correlates with self-efficacy and sleep quality and negatively with anxiety and stress.

    “level of social support was found to positively correlate with self-efficacy and sleep quality and negatively with anxiety and stress (Xiao et al., 2020).”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Xiao et al., 2020

      Not checked in the free run.

27 of 27

  • The paper relies on this without support.

    The impact of the COVID-19 crisis on the psychological well-being of medical and nursing staff had not yet been established at the time of the review.

    “However, to date, the impact of the current unprecedented crisis on the psychological well-being of medical and nursing staff is yet to be established.”

  • The paper relies on this without support.

    This is the first systematic review and meta-analysis to examine the pooled prevalence of depression, anxiety and insomnia in healthcare workers during the COVID-19 outbreak.

    “To our knowledge, this is the first systematic review and meta-analysis to examine the pooled prevalence of depression, anxiety and insomnia on HCW during the COVID-19 outbreak.”

  • The authors state this as an assumption.

    One true effect size cannot be assumed across studies due to different patient populations, regions, and assessment methods.

    “Due to the different patient populations, regions, and assessment methods across studies, one true effect size cannot be assumed; therefore, a random effects model (DerSimonian & Laird) was used to extract the pooled prevalence.”

  • The paper relies on this without support.

    China was severely affected by COVID-19.

    “considering the fact that China was severely affected, they provide a reliable indication of the potential of COVID-19 pandemic to affect the mental health of HCWs.”

  • The paper relies on this without support.

    Healthcare systems vary greatly between countries.

    “generalizing this type of results could pose severe flaws as healthcare systems vary greatly between countries.”
5 of 5

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?

  • Across 13 studies of healthcare staff, pooled estimates showed 23.2% with raised COVID-19 anxiety, 22.8% with depression, and 38.9% with insomnia.

    “A meta-analysis of 13 studies of mental health among healthcare workers found that almost a quarter exhibited elevated COVID-19 related symptoms of anxiety (23.2%) and depression (22.8%), and more than a third (38.9%) experienced problems with insomnia (Pappa et al. 2020).”

    From Risk and protective factors for prospective changes in adolescent mental health during the COVID-19 pandemic (2021) · cites this paper as Pappa et al. 2020

    The Results report 13 included studies, with pooled anxiety prevalence of 23.21% and depression prevalence of 22.8%. The Results > Insomnia prevalence section instead reports 34.32%, not 38.9%; only five of the 13 studies contributed insomnia data. (LLM, conf=0.94)

    Experimental — accuracy not yet measured

  • Frontline work has been reported to increase the risk for psychological symptoms in healthcare workers.

    “Several factors have been reported to increase the risk for psychological symptoms in healthcare workers, including being a nurse, female gender, and frontline work (Muller et al., 2020; Pappa et al., 2020).”

    From Beyond the physical risk: Psychosocial impact and coping in healthcare professionals during the COVID‐19 pandemic. (2025) · cites this paper as Pappa et al., 2020

    The Discussion says frontline physicians and nurses experience psychological effects during the COVID-19 crisis. It does not report that frontline work increases the risk of symptoms relative to other healthcare work; it says the impact is also felt by workers in other specialties. No page number is available in our copy. (LLM, conf=0.82)

    Experimental — accuracy not yet measured

  • Being a nurse has been reported to increase the risk for psychological symptoms in healthcare workers.

    “Several factors have been reported to increase the risk for psychological symptoms in healthcare workers, including being a nurse, female gender, and frontline work (Muller et al., 2020; Pappa et al., 2020).”

    From Beyond the physical risk: Psychosocial impact and coping in healthcare professionals during the COVID‐19 pandemic. (2025) · cites this paper as Pappa et al., 2020

    The Discussion (page not supplied) reports higher estimated prevalence of anxiety and depression among nurses than doctors. It also notes possible confounding by gender, so the observational finding does not establish that being a nurse itself increases risk. (LLM, conf=0.86)

    Experimental — accuracy not yet measured

  • The COVID-19 pandemic led to increased stress, distress, and burnout across the health and social care workforce, including palliative care.

    “The COVID-19 pandemic led to increased stress, distress, and burnout across the health and social care workforce,9–12 including palliative care13–16 and exacerbated the need for healthcare staff support internationally.13–17”

    From Feasibility of RESTORE: An online Acceptance and Commitment Therapy intervention to improve palliative care staff wellbeing. (2023) · cites this paper as [9]

    The Discussion reports high rates of anxiety and depression among healthcare professionals during COVID-19 and cites emerging evidence of burnout among medical and nursing staff. The review does not establish an increase from pre-pandemic levels or examine the broader social care workforce. The palliative-care clause has separate citations and is not attributed to this source. (LLM, conf=0.85)

    Experimental — accuracy not yet measured

  • Female gender has been reported to increase the risk for psychological symptoms in healthcare workers.

    “Several factors have been reported to increase the risk for psychological symptoms in healthcare workers, including being a nurse, female gender, and frontline work (Muller et al., 2020; Pappa et al., 2020).”

    From Beyond the physical risk: Psychosocial impact and coping in healthcare professionals during the COVID‐19 pandemic. (2025) · cites this paper as Pappa et al., 2020

Cites

Flagged first, then by citations

14 of the 62 references couldn’t be matched to a known paper.

Research works printed but not linked: 1

  1. Bohlken, 2020, COVID-19 pandemic: stress experience of healthcare workers – a short current review, Psychiatr. Prax, 47, 190

Cited by

Sorted by citations

Related

Authors
Pappa, S., Ntella, V., Giannakas, T., Giannakoulis, V. G., Papoutsi, E., & Katsaounou, P.
Affiliations
  • West London Mental Health NHS Trust (GB)
  • National and Kapodistrian University of Athens (GR)
Corresponding author
Pappa, Sofia
ORCID
Venue
Brain Behavior and Immunity
Publisher
Elsevier BV
Published
2020-08
Volume
88
Pages
901-907
ISSN
0889-1591 (ISSN-L)
DOI
10.1016/j.bbi.2020.05.026
PMC
PMC7206431
OpenAlex
W3021065879
Cited by
3,617
References
62
Citations by year
2026: 161 · 2025: 231 · 2024: 357 · 2023: 568 · 2022: 885 · 2021: 1070 · 2020: 342 · 2017: 1 · 2015: 1
FWCI
56.00
Type
journal article
Language
en
Update notices
erratum (2021-02-01)
Open access
Licence
Locations
  • Brain Behavior and Immunity (publishedVersion)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
Topics
  • Social Sciences › Psychology › Clinical Psychology › COVID-19 and Mental Health
  • Health Sciences › Health Professions › General Health Professions › Healthcare professionals’ stress and burnout
  • Health Sciences › Medicine › Public Health, Environmental and Occupational Health › Maternal Mental Health During Pregnancy and Postpartum
Keywords
COVID-19 pandemic, healthcare workers, depression, anxiety, insomnia, mental health, prevalence, nurses, gender differences, occupational differences, mood disorders, sleep disorders
MeSH terms
Betacoronavirus, COVID-19, SARS-CoV-2, Anxiety, Depression, Health Personnel, Humans, Sleep Initiation and Maintenance Disorders, Nurses, Physicians, Pneumonia, Viral, Prevalence, Sex Distribution, Coronavirus Infections, Pandemics

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 under definition version 3 (extract-v3) on 2026-09-28.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-09-28.

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