Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses
Open accessPublished version of a preprint
The DOI matches its Crossref record; 57 of the 57 references it cites were found in a registry or an index; 12 of the 87 sources its claims cite were checked for support.
Abstract
Background: Mental health apps have great potential to help people needing support to cope with distress or specific symptoms. In fact, there is an exponential increase in the number of mental health apps available on the internet, with less than 5% being actually studied. Objective: This study aimed to assess the quality of the available evidence regarding the use of mental health apps and to summarize the results obtained so far. Methods: Systematic reviews and meta-analyses were searched, specifically for mobile apps on mental health issues or symptoms, and rated using the Grading of Recommendations Assessment, Development and Evaluation system. Results: A total of 7 meta-analyses were carefully reviewed and rated. Although some meta-analyses looked at any mental health issue and analyzed the data together, these studies were of poorer quality and did not offer strong empirical support for the apps. Studies focusing specifically on anxiety symptoms or depressive symptoms were of moderate to high quality and generally had small to medium effect sizes. Similarly, the effects of apps on stress and quality of life tended to offer small to medium effects and were of moderate to high quality. Studies looking at stand-alone apps had smaller effect sizes but better empirical quality than studies looking at apps with guidance. The studies that included follow-ups mostly found a sustained impact of the app at an 11-week follow-up. Conclusions: This meta-review revealed that apps for anxiety and depression hold great promise with clear clinical advantages, either as stand-alone self-management or as adjunctive treatments. More meta-analyses and more quality studies are needed to recommend apps for other mental health issues or for specific populations.
Overview
CoChat Summary
TL;DR
This meta-review of 7 meta-analyses found that apps for anxiety and depression show promise, with small-to-medium effects and moderate-to-high quality evidence, while evidence for other mental health issues remains limited. Stand-alone apps produced smaller effects but better empirical quality than apps with guidance or adjunctive apps.
Key findings
- The search retrieved 2558 papers; 24 meta-analyses were reviewed and 7 were included. Across subanalyses, 14 results were poor or poor-to-moderate, 15 were moderate or moderate-to-high, and 8 were high-quality evidence.
- For anxiety, Firth et al. reported overall g=0.33 (95% CI 0.17-0.48), and Linardon et al. reported g=0.30 (95% CI 0.2-0.4), with a 7-11 week follow-up effect of g=0.52 (95% CI 0.41-0.63). For depression, Firth et al. reported overall g=0.38 (95% CI 0.242-0.524), and Linardon et al. reported g=0.28 (95% CI 0.21-0.36), with a 7-11 week follow-up effect of g=0.46 (95% CI 0.36-0.55).
- Apps with guidance had larger effects than stand-alone apps: Linardon et al. found anxiety guidance g=0.53 vs stand-alone g=0.21, and depression guidance g=0.48 vs stand-alone g=0.23; Versluis et al. found guidance g=0.73 vs stand-alone g=0.45. However, stand-alone apps were supported by better-quality evidence.
- For specific anxiety symptoms, social anxiety showed a significant medium effect (g=0.58), while panic (g=-0.05) and PTSD (g=0.18) did not improve significantly. For suicidal ideation, RCTs showed d=-0.26; self-harm frequency was nonsignificant.
Methods
- This was a meta-review of systematic reviews reporting quantitative pooled data (meta-analyses) on app technology for mental health issues, published in English or French. MEDLINE, EMBASE, Current Contents, PsycINFO, and Google Scholar were searched; final inclusion decisions were made by consensus of 2 researchers.
- Evidence quality was rated using the GRADE system, considering sample size, precision, directness, consistency, study design, follow-up, publication bias, and confounding factors. Effect sizes were presented separately from quality ratings; 2 expert raters achieved over 95% initial agreement.
Limitations
- Stated in the paper: the review focused on mental health and excluded apps focused on a specific intervention or model without symptom outcomes; no meta-analyses were found for severe mental illness, addiction, eating disorders, or obsessive-compulsive disorder; few reviews considered confounding factors such as actual frequency or time of app exposure; control conditions varied; a digital placebo effect could not be excluded; several studies used nonclinical populations, limiting generalizability; attrition was a repeated problem.
- Evident from the text: only 7 meta-analyses were included, many subanalyses had heterogeneity, imprecision, or publication bias, follow-up data were limited, and actual app usage was not measured.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
10 Valid · 1 Not valid · 67 Not checked · 41 Own results and interpretations
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, the quality of the evidence was rated as poor to moderate (see Tables 1 and 2), given that the effects were imprecise, the samples were very heterogeneous, and the effects were indirect (no subanalyses by diagnosis or problem and all mixed together).”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Mental Health (Multiple Problems)
What the sources say
- Partly supportedcites [44] sourceExperimental — accuracy not yet measured
The Results report significant heterogeneity across 25 studies, and Limitations and Future Directions says the small number of studies limited moderator analyses. Our copy does not rate the evidence as poor to moderate or establish the citing authors’ imprecision assessment. It also reports comparisons among medical conditions, addictions, and psychological disorders, so the claim that there were no subanalyses by problem is not supported. (LLM, conf=0.83)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“However, only a small proportion of these technologies have undergone any form of empirical assessment [3].”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Mobile Health and Apps
What the sources say
- Partly supportedcites [3] sourceExperimental — accuracy not yet measured
The Discussion (page unavailable) says the literature review found few results compared with the commercial-app review, supporting the general point that research has not kept pace with app development. It does not establish what proportion of mobile health technologies underwent empirical assessment; a publication about an app is not necessarily an empirical assessment of it. (LLM, conf=0.82)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
Every source the paper cites for this claim was checked and supports it.
In the paper
“Firth et al [39], on the other hand, included only RCTs and compared apps with waitlist or active controls and found a small to medium effect size overall when compared with waitlist and small effect when compared with active controls.”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Anxiety
What the sources say
- Supportedcites [39] source
Section 3.3 reports a small effect of smartphone interventions on anxiety compared with active controls (g=0.189), alongside a moderate effect compared with waitlist controls (g=0.452). It also states that the analysis used data from RCTs, supporting the cited sentence. (LLM, conf=0.95)
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“The quality of the evidence was rated as moderate, given the small sample size and the lack of control for biases (publication or otherwise).”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Depression
What the sources say
- Partly supportedcites [48] sourceExperimental — accuracy not yet measured
The Discussion 'Limitations of the included studies' (and the Abstract Conclusions) state that most included studies were of 'low to moderate quality' on the Cochrane risk-of-bias tool / ROBINS–I, with bias concentrated in participant, clinical personnel and outcome assessor blinding — so a 'moderate' quality characterisation is partly there. But this is a study-level risk-of-bias judgement across all 13–14 studies, not a GRADE-style 'quality of evidence' rating made specifically for the suicidal ideation RCTs (Methods 'Risk of bias' lists only the Cochrane and ROBINS–I tools; no per-outcome evidence certainty rating appears anywhere in our copy). Nor does the paper attribute the rating to small sample size — the stated basis is unblinding (performance and detection bias), while 'small number of studies included in any one meta-analysis' is raised only as a reason subgroup/sensitivity analyses could not be undertaken — and publication bias is not assessed or mentioned. (LLM, conf=0.72)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“followed up with a similar effect size”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Depression
What the sources say
- Partly supportedcites [48] sourceExperimental — accuracy not yet measured
In Results > Suicidal ideation, our copy gives the two RCT-only pooled estimates for suicidal ideation scores: SMD -0.26 (95% CI -0.44 to -0.08, 5 RCTs, p = 0.005) at post-intervention and SMD -0.34 (95% CI -0.70 to 0.01, 2 studies, p = 0.06) at the final follow-up, so the magnitudes are indeed comparable and the numeric core of the claim is in the source. However, the paper does not describe the follow-up result as a 'similar effect size' or as a maintained effect — it states the interventions were 'not associated with a significant treatment effect for suicidal ideation by this time point', with a confidence interval crossing zero, and notes that because both follow-up studies were RCTs no subgroup-difference or sensitivity testing was possible. The citing sentence therefore upgrades a non-significant, two-trial estimate into a demonstrated similar effect at follow-up. (LLM, conf=0.70)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Partly supported
50 more claims · 1 with their cited sources checked
Create a free account to see all 119 claims and to verify the 75 references still open.
- Probably not a substantial claim. 15
- It may state two separate claims as one. 51
- The quote only partly states this claim. 2
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 source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Research has shown that mental health applications have distinct clinical benefits for users, with a meta-analysis synthesizing the effectiveness of more than twenty mental health applications suggesting that using applications to relieve symptoms and self-manage illness effectively reduced depressive symptoms [14].”
From A Review of Current Digital Mental Health Care Applications for Anxiety Symptoms and Future Prospects (2024) · cites this paper as [14]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“However, research on the efficacy and effectiveness of blended care is limited compared to the evidence base for self-guided and guided DMHIs [8,9,23] and is confined mainly to Europe and North America [14,24].”
From Needs and expectations for the myNewWay blended digital and face-to-face psychotherapy model of care for depression and anxiety (part 1): participatory design study including people with lived and living experience (2025) · cites this paper as [23]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Several systematic reviews have recently been published focusing on various aspects and outcomes of mental health apps [8–13].”
From Artificial intelligence and machine learning in mobile apps for mental health: A scoping review (2022) · cites this paper as [8]
The cited source says what the claim attributes to it.
In the citing paper
“Many mobile health apps have been developed and made available to the public to try and address this need [8].”
From Artificial intelligence and machine learning in mobile apps for mental health: A scoping review (2022) · cites this paper as [8]
The cited source says what the claim attributes to it.
In the citing paper
“In addition to their feasibility, there is a growing body of evidence supporting the efficacy of app-based mental health interventions [44,45].”
From App-Based Mindfulness Meditation for People of Color Who Experience Race-Related Stress: Protocol for a Randomized Controlled Trial (2022) · cites this paper as [44]
5 more claims about this paper
Create a free account to see every claim other papers make about this one.
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.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Other Mental Health Concepts
“For depression, the effect size was medium for apps with guidance versus small for stand-alone apps, with the quality of the evidence being moderate to high for stand-alone apps and moderate for guidance.”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Current Knowledge on Mobile Apps for Mental Health Problems
“| | RCT | 978, controls: 918 | | | | medium | poor-moderate | | | RCT | 2489, controls: 2522 | | | | small | moderate |”
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Other Mental Health Concepts
“Versluis [45] found a medium to large effect size when guidance was offered, compared with medium effect size for stand-alone apps, with stand-alone apps being supported by poor evidence compared with poor to moderate evidence for guidance.”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Current Knowledge on Mobile Apps for Mental Health Problems
“| | PC+RCT | 474 | No follow-up described | No | Unknown for this subanalysis | medium-large | moderate |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Depression
“When looking at noncontrolled studies for suicidal ideation, the quality of the evidence drops to very poor, with small sample size, high heterogeneity, and imprecise effect.”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Current Knowledge on Mobile Apps for Mental Health Problems
“| | PC | 149 | No follow-up described | No | unknown | small-medium | poor |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Current Knowledge on Mobile Apps for Mental Health Problems
“| | N/A | outcome: general distress | g=0.40 | 95% CI: 0.24-0.56, Yes | I²: 60% No | No |”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Current Knowledge on Mobile Apps for Mental Health Problems
“| Authors | Technology used | Intervention type and subtype or outcome type | Effect size d, g, OR, RR | Precise (less than 0.25 = precise) 95% CI, Yes or No | Consistent Q or I², Yes or No | Direct Yes or No |”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Results > Mental Health (Multiple Problems)
“As a consequence, the results are indirect, highly heterogeneous, and imprecise with a total quality score of moderate.”
§ Mobile Apps for Mental Health Issues: Meta-Review of Meta-Analyses > Introduction > Current Knowledge on Mobile Apps for Mental Health Problems
“| Stratton et al., 2017 | RCT | 2399, controls 2265 (total but only 3 app studies) | follow-up described (g=0.23), but not specific for apps | No | Yes, adjusted effect size = g=0.12, 95% CI = 0.01-0.25 | small | poor-moderate |”
Citations
Cites
Grading quality of evidence and strength of recommendations
(2004) · BMJ · doi:10.1136/bmj.328.7454.1490 · Cited by 8,477 · cited 3×
“The GRADE system was used to assess evidence quality [42].”
MethodInfluentialGRADE guidelines: 3. Rating the quality of evidence
Balshem, H., Helfand, M., Schünemann, H. J., Oxman, A. D., Kunz, R., Brozek, J., Vist, G. E., Falck-Ytter, Y., Meerpohl, J., & Norris, S. (2011) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2010.07.015 · Cited by 8,285 · cited 1×
“The magnitude of the impact of the app is determined based on the estimated effect size (the larger the value, the better) [43].”
BackgroundEcological momentary assessment
Shiffman, S., Stone, A. A., & Hufford, M. R. (2008) · Annual Review of Clinical Psychology · doi:10.1146/annurev.clinpsy.3.022806.091415 · Cited by 6,454
Andrews, G., Cuijpers, P., Craske, M. G., McEvoy, P., & Titov, N. (2010) · PLoS ONE · doi:10.1371/journal.pone.0013196 · Cited by 1,311 · cited 1×
“Furthermore, several studies were conducted with nonclinical populations, who presented with symptoms but perhaps not a diagnosed disorder, limiting the generalizability of the results for clinical populations [54].”
BackgroundRahmani, A. M., Gia, T. N., Negash, B., Anzanpour, A., Azimi, I., Jiang, M., & Liljeberg, P. (2018) · Future Generation Computer Systems · doi:10.1016/j.future.2017.02.014 · Cited by 1,198
Firth, J., Torous, J., Nicholas, J., Carney, R., Pratap, A., Rosenbaum, S., & Sarris, J. (2017) · World Psychiatry · doi:10.1002/wps.20472 · Cited by 1,097 · cited 11×
“As was mentioned for anxiety disorders, Linardon et al’s [47] meta-analysis is the most recent but does not include all the studies reviewed in either Versluis et al’s [45] or Firth et al’s [40] meta-analyses, justifying the need to keep all 3 in this meta-review.”
BackgroundInfluentialLinardon, J., Cuijpers, P., Carlbring, P., Messer, M., & Fuller‐Tyszkiewicz, M. (2019) · World Psychiatry · doi:10.1002/wps.20673 · Cited by 860 · cited 10×
“As for indirect mental health outcomes, namely, outcomes that were considered but were not the main focus of the app intervention (such as distress, stress, or quality of life), the effects are consistently small to medium, with greater quality evidence for the most recent meta-analysis [47].”
BackgroundResultInfluentialA Smartphone Application to Support Recovery From Alcoholism: A Randomized Clinical Trial
Gustafson, D. H., McTavish, F. M., Chih, M.-Y., Atwood, A. K., Johnson, R. A., Boyle, M. G., Levy, M. S., Driscoll, H., Chisholm, S. M., Dillenburg, L., et al. (2014) · JAMA Psychiatry · doi:10.1001/jamapsychiatry.2013.4642 · Cited by 692 · cited 1×
“Most of these studies have been conducted in the last few years and assessed either the feasibility and acceptability of mental health apps and, in some cases, their efficacy for a broad spectrum of mental disorders, including depressive disorders, posttraumatic stress disorders, schizophrenia, bipolar disorders, or a…”
BackgroundFirth, J., Torous, J., Nicholas, J., Carney, R., Rosenbaum, S., & Sarris, J. (2017) · Journal of Affective Disorders · doi:10.1016/j.jad.2017.04.046 · Cited by 675 · cited 6×
“Multiple meta-analyses on apps focusing on a single or multiple mental health problems have been conducted [39,40], with very different results at times.”
BackgroundResultInfluentialVirtual reality exposure therapy in anxiety disorders: a quantitative meta-analysis
Opriş, D., Pintea, S., García-Palacios, A., Botella, C., Szamosközi, Ş., & David, D. (2012) · Depression and Anxiety · doi:10.1002/da.20910 · Cited by 636 · cited 1×
“The effect sizes of interventions are typically lower when compared with active controls instead of inactive controls [49,50].”
Background
Cited by
Torous, J., Bucci, S., Bell, I. H., Kessing, L. V., Faurholt‐Jepsen, M., Whelan, P., Carvalho, A. F., Keshavan, M., Linardon, J., & Firth, J. (2021) · World Psychiatry · doi:10.1002/wps.20883 · Cited by 1,199
Stein, D. J., Shoptaw, S. J., Vigo, D. V., Lund, C., Cuijpers, P., Bantjes, J., Sartorius, N., & Maj, M. (2022) · World Psychiatry · doi:10.1002/wps.20998 · Cited by 331
Deniz-Garcia, A., Fabelo, H., Rodriguez-Almeida, A. J., Zamora-Zamorano, G., Castro-Fernandez, M., Alberiche Ruano, M. D. P., Solvoll, T., Granja, C., Schopf, T. R., Callico, G. M., et al. (2023) · Journal of Medical Internet Research · doi:10.2196/44030 · Cited by 268
“In general, apps have been found to exert a small but notable effect on enhancing healthy lifestyles, specifically regarding nutrition [12], physical activity [13], management of chronic conditions [14-17] and mental health [18].”
Goldberg, S. B., Lam, S. U., Simonsson, O., Torous, J., & Sun, S. (2022) · PLOS Digital Health · doi:10.1371/journal.pdig.0000002 · Cited by 237
“[25] conducted the only meta-review investigating the effects of mobile phone-based interventions on mental health.”
BackgroundInfluentialKowal, M., Conroy, E., Ramsbottom, N., Smithies, T., Toth, A., & Campbell, M. (2021) · JMIR Serious Games · doi:10.2196/26575 · Cited by 172
“Despite promising findings in this area, a recent meta-analysis by Lecomte et al [46] noted that to date, only approximately 3% of mental health apps have been thoroughly validated by research.”
BackgroundResult
Metadata
- Authors
- Lecomte, T., Potvin, S., Corbière, M., Guay, S., Samson, C., Cloutier, B., Francoeur, A., Pennou, A., & Khazaal, Y.
- Affiliations
- Institut universitaire en santé mentale de Montréal (CA)
- Université du Québec à Montréal (CA)
- Université de Montréal (CA)
- Centre Hospitalier Universitaire Vaudois (CH)
- Corresponding author
- Lecomte, Tania
- ORCID
- Venue
- JMIR mHealth and uHealth
- Publisher
- JMIR Publications Inc.
- Published
- 2020-05-29
- Volume
- 8
- Issue
- 5
- Pages
- e17458
- ISSN
- 2291-5222 (ISSN-L)
- PMC
- PMC7293054
- OpenAlex
- W3008344128
- Cited by
- 336
- References
- 57
- Citations by year
- 2026: 37 · 2025: 61 · 2024: 84 · 2023: 54 · 2022: 52 · 2021: 42 · 2020: 5
- FWCI
- 6.03
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- JMIR mhealth and uhealth (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- IRIS (submittedVersion, cc-by)
- SERVAL (Université de Lausanne) (submittedVersion, cc-by)
- Topics
- Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
- Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Mental Health Research Topics
- Keywords
- mental health apps, mobile health, digital interventions, anxiety symptoms, depressive symptoms, stress, quality of life, self-management, adjunctive therapy, quality of evidence, follow-up outcomes
- MeSH terms
- Anxiety, Humans, Mental Health, Quality of Life, Internet, Mobile Applications
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-06 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol and qwen3.8-omni-flash).
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.
