A Mobile Health Intervention (LifeBuoy App) to Help Young People Manage Suicidal Thoughts: Protocol for a Mixed-Methods Randomized Controlled Trial
Open accessPublished version of a preprint
The DOI matches its Crossref record; 54 of the 54 references it cites were found in a registry or an index; 11 of the 22 sources its claims cite were checked for support.
Abstract
Background: Self-help smartphone apps offer a new opportunity to address youth suicide prevention by improving access to support and by providing potentially high fidelity and cost-effective treatment. However, there have been very few smartphone apps providing evidence-based support for suicide prevention in this population. To address this gap, we developed the LifeBuoy app, a self-help smartphone app informed by dialectical behavior therapy (DBT), to help young people manage suicidal thoughts in their daily life. Objective: This study describes the protocol for a randomized controlled trial to evaluate the efficacy of the LifeBuoy app for reducing suicidal thoughts and behaviors, depression, anxiety, and psychological distress, and improving general mental well-being in young adults aged 18 to 25 years. Methods: This is a randomized controlled trial recruiting 378 young adults aged between 18 and 25 years and comparing the LifeBuoy app with a matched attention control (a placebo app with the same display but no DBT components). The primary outcome is suicidal thoughts measured by the Suicidal Ideation Attributes Scale (SIDAS). The secondary outcomes are suicidal behavior, depression, anxiety, psychological distress, and general mental well-being. The changes in the levels of insomnia, rumination, suicide cognitions, distress tolerance, loneliness, and help seeking before and after using the app are evaluated in this study. The study also addresses risk factors and responses to the intervention. A series of items assessing COVID-19 experiences is included in the trial to capture the potential impact of the pandemic on this study. Assessments will occur on the following three occasions: baseline, postintervention, and follow-up at 3 months postintervention. A qualitative interview about user experience with the LifeBuoy app will take place within 4 weeks of the final assessment. Using linear mixed models, the primary analysis will compare the changes in suicidal thoughts in the intervention condition relative to the control condition. To minimize risks, participants will receive a call from the team clinical psychologist by clicking a help button in the app or responding to an automated email sent by the system when they are assessed with elevated suicide risks at the baseline, postintervention, and 3-month follow-up surveys. Results: The trial recruitment started in May 2020. Data collection is currently ongoing. Conclusions: This is the first trial examining the efficacy of a DBT-informed smartphone app delivered to community-living young adults reporting suicidal thoughts. This trial will extend knowledge about the efficacy and acceptability of app-based support for suicidal thoughts in young people. ©Jin Han, Lauren McGillivray, Quincy JJ Wong, Aliza Werner-Seidler, Iana Wong, Alison Calear, Helen Christensen, Michelle Torok.
Overview
CoChat Summary
TL;DR
This paper is a protocol for a randomized controlled trial of LifeBuoy, a DBT-informed self-help smartphone app for young adults aged 18–25 years with suicidal thoughts. It tests whether the app reduces suicidal thoughts versus a matched attention control, and also evaluates secondary mental health outcomes, app adherence, satisfaction, and acceptability.
Key findings
- The planned trial recruits 378 young adults aged 18 to 25 years, with 189 per condition, powered at 0.80 and alpha = .05 to detect an expected effect size of 0.30 (Cohen d) for suicidal thoughts.
- The primary outcome is suicidal thoughts measured by the Suicidal Ideation Attributes Scale (SIDAS), scored 0 to 50, with higher scores indicating more severe suicidal thoughts.
- Assessments occur at baseline (day 0), postintervention (day 42), and 3-month follow-up (day 132); a qualitative interview is conducted within 4 weeks of the final assessment.
- The trial also measures secondary outcomes including suicidal behaviors, depression, anxiety, psychological distress, and mental well-being, plus tertiary outcomes such as insomnia, rumination, suicide cognitions, distress tolerance, loneliness, and help-seeking.
Methods
- Design: Randomized controlled superiority trial with a matched attention control; participants are allocated in a 2:2 block design (block size 4), stratified by gender and age group, using an automated web-based platform.
- Sample/data: Community-based sample of 18–25-year-old young adults in Australia who reported suicidal thoughts in the past 12 months; inclusion also required no suicide attempt in the past month, no psychosis or bipolar diagnosis, and fluency in English.
- Analysis: Mixed model repeated measures analyses with maximum likelihood estimation and an unstructured covariance matrix; intention-to-treat is used under a missing-at-random assumption. Qualitative interviews are analyzed using Braun & Clarke thematic analysis with two independent coders and a third researcher resolving discrepancies.
Limitations
- The paper states that participants may not be blinded to allocation because of the app content.
- The sample is recruited through targeted Facebook advertising, which the authors note as a limitation of the recruitment approach.
- The trial is conducted during the COVID-19 period, which the authors say may elevate distress and symptoms.
- Evident from the text: this is a protocol, so no efficacy results are reported yet; the paper explicitly says recruitment started in May 2020 and data collection is ongoing.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
6 Valid · 3 Not valid · 5 Not checked · 4 Own results and interpretations
Some cited sources support this claim and some don’t — read each source’s verdict below.
In the paper
“Based on an attrition rate of 30% [28,29] and a 0.50 correlation between pre- and posttest scores of the primary outcome, a sample size of 189 for each condition (total N=378) will be required to detect the expected effect size with a power of 0.80 and alpha of .05.”
§ A Mobile Health Intervention (LifeBuoy App) to Help Young People Manage Suicidal Thoughts: Protocol for a Mixed-Methods Randomized Controlled Trial > Methods > Sample Size
What the sources say
- Disputedcites [28] sourceExperimental — accuracy not yet measured
The citation is attached to the assumed 30% attrition rate. The cited paper’s Methods > Sample size section instead specifies an expected drop-out rate of 20%; it does not support the cited assumption of 30%. Page number is unavailable in our copy. (LLM, conf=0.91)
Checked by a reasoning model — may be wrong.
- Supportedcites [29] source
The citation is attached to the 30% attrition assumption, not the citing authors’ calculated sample size. In Methods > Sample Size, the source explicitly allows for up to 30% attrition. It also assumes a 0.5 correlation, although its own calculation calls for 285 participants per arm rather than 189. (LLM, conf=0.92)
plainn = 378- 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.
- Disputed
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Traditional face-to-face psychotherapies, including DBT, usually carry high economic costs and personal barriers that may prevent youth from accessing them [20].”
§ A Mobile Health Intervention (LifeBuoy App) to Help Young People Manage Suicidal Thoughts: Protocol for a Mixed-Methods Randomized Controlled Trial > Introduction
What the sources say
- Partly supportedcites [20] sourceExperimental — accuracy not yet measured
The Results (Correlates of barriers to treatment; page not provided) report that younger respondents were more likely to encounter structural barriers. The Discussion (page not provided) identifies possible financial and time barriers for younger respondents. The paper addresses mental-health treatment generally, not the costs of face-to-face psychotherapy or DBT specifically, and its younger respondents are not identified as youth. (LLM, conf=0.88)
hedgedstates a possible cause- 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
“Finally, the sample is being recruited during the COVID-19 period, which is likely to elevate participants’ levels of psychological distress and mental health symptoms [54].”
§ A Mobile Health Intervention (LifeBuoy App) to Help Young People Manage Suicidal Thoughts: Protocol for a Mixed-Methods Randomized Controlled Trial > Discussion
What the sources say
- Supportedcites [54] source
The Abstract reports elevated depression, anxiety and stress among respondents surveyed during the COVID-19 outbreak, and says 78% reported that their mental health had worsened since the outbreak. This supports the citing sentence’s hedged expectation of elevated distress and mental health symptoms during recruitment, without establishing a causal effect for its particular sample. (LLM, conf=0.88)
hedgedstates a possible cause- 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 that could be checked couldn’t confirm or dispute this claim.
In the paper
“For the qualitative interviews, we will recruit a 10% subsample of individuals who participate in the intervention condition to assess their experience with the app (N=20), which is likely to reach data saturation according to prior studies [30,31].”
§ A Mobile Health Intervention (LifeBuoy App) to Help Young People Manage Suicidal Thoughts: Protocol for a Mixed-Methods Randomized Controlled Trial > Methods > Sample Size
What the sources say
- Could not verifycites [30] sourceExperimental — accuracy not yet measured
The Results (Participant Characteristics; page unavailable) report 21 adolescents in three focus groups, and the Methods (Design; page unavailable) describe how those groups were conducted. Neither section says that 20 interviews, or a 10% subsample of an intervention group, is likely to reach data saturation; a search for data and thematic saturation found no such passage. (LLM, conf=0.75)
- Could not verifycites [31] sourceExperimental — accuracy not yet measured
The Methods > Statistical Analysis > Sample Size passage describes a target of 25 people for app review and says prior studies guided that choice. The Methods > Step 2: Participatory Engagement passage reports six focus groups, but neither passage addresses whether interviews with a 10% subsample (N=20) would reach data saturation. Searches for qualitative interviews and data saturation found no supporting passage in our copy. (LLM, conf=0.76)
hedgedn = 20- 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.
- Could not verify
Every source the paper cites for this claim was checked and supports it.
In the paper
“Based on a systematic review and meta-analysis on online and mobile apps for suicidal thoughts and self-harm [27], an effect size of 0.30 (Cohen d) is expected between the intervention condition and the control condition at posttest for the primary outcome suicidal thoughts.”
§ A Mobile Health Intervention (LifeBuoy App) to Help Young People Manage Suicidal Thoughts: Protocol for a Mixed-Methods Randomized Controlled Trial > Methods > Sample Size
What the sources say
- Supportedcites [27] source
The Results (Suicidal ideation) report a post-intervention standardized mean difference of −0.26 for suicidal ideation scores across five randomized controlled trials. Its magnitude is close to the citing authors’ expected effect size of 0.30; the expectation is their planning estimate, not a prediction made by the review. (LLM, conf=0.88)
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.
- Supported
4 more claims
Create a free account to see all 18 claims and every source verdict.
- Probably not a substantial claim. 6
- It may state two separate claims as one. 6
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
“The trial protocol (S1 Protocol) was prospectively registered on the Australian New Zealand Clinical Trials Registry (ACTRN12619001671156) and has also been published elsewhere [24].”
From The effect of a therapeutic smartphone application on suicidal ideation in young adults: Findings from a randomized controlled trial in Australia. (2022) · cites this paper as [24]
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
“Due to delays to the development of the LifeBuoy application, the final assessment was changed to 3-month postintervention to ensure the trial would complete in the funding timeline. This change was approved at March 14, 2020, prior to the start of recruitment, by the University of New South Wales Human Research Ethics Committee and is accurately reflected in our published protocol [24].”
From The effect of a therapeutic smartphone application on suicidal ideation in young adults: Findings from a randomized controlled trial in Australia. (2022) · cites this paper as [24]
Experimental — accuracy not yet measured
Citations
Cites
The PHQ-9: Validity of a brief depression severity measure
Kroenke, K., Spitzer, R. L., & Williams, J. B. W. (2001) · Journal of General Internal Medicine · doi:10.1046/j.1525-1497.2001.016009606.x · Cited by 44,910
A Brief Measure for Assessing Generalized Anxiety Disorder: The GAD-7
Spitzer, R. L., Kroenke, K., Williams, J. B. W., & Löwe, B. (2006) · Archives of Internal Medicine · doi:10.1001/archinte.166.10.1092 · Cited by 31,765
Grit: perseverance and passion for long-term goals
Duckworth, A. L., Peterson, C., Matthews, M. D., & Kelly, D. R. (2007) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.92.6.1087 · Cited by 6,911 · cited 1×
“The Short Grit Scale (SGS [47]) is an eight-item measure assessing perseverance and passion for pursuing long-term goals.”
BackgroundThe Revised UCLA Loneliness Scale: concurrent and discriminant validity evidence
Russell, D., Peplau, L. A., & Cutrona, C. E. (1980) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.39.3.472 · Cited by 5,173 · cited 2×
“The three items that compose this scale were selected from the R-UCLA Loneliness Scale [44] and include the following: “How often do you feel that you lack companionship?””
MethodThe Warwick-Edinburgh Mental Well-being Scale (WEMWBS): development and UK validation
Tennant, R., Hiller, L., Fishwick, R., Platt, S., Joseph, S., Weich, S., Parkinson, J., Secker, J., & Stewart-Brown, S. (2007) · Health and Quality of Life Outcomes · doi:10.1186/1477-7525-5-63 · Cited by 5,031
A Short Scale for Measuring Loneliness in Large Surveys: Results From Two Population-Based Studies
Hughes, M. E., Waite, L. J., Hawkley, L. C., & Cacioppo, J. T. (2004) · Research on Aging · doi:10.1177/0164027504268574 · Cited by 4,781 · cited 1×
“The Three-Item Loneliness Scale (TILS [43]) is a brief self-report measure of loneliness.”
BackgroundMorin, C. M., Belleville, G., Bélanger, L., & Ivers, H. (2011) · Sleep · doi:10.1093/sleep/34.5.601 · Cited by 4,620 · cited 1×
“The Insomnia Severity Index (ISI [39]) is a psychometrically sound, seven-item, self-report measure assessing the perceived”
BackgroundQualitative Psychology: A Practical Guide to Research Methods
(2008) · QMiP Bulletin · doi:10.53841/bpsqmip.2008.1.6.47 · Cited by 4,351
Linear Mixed Models for Longitudinal Data
Verbeke, G. (1997) · Lecture Notes in Statistics · doi:10.1007/978-1-4612-2294-1_3 · Cited by 3,506
Cognitive-behavioral treatment of chronically parasuicidal borderline patients
Linehan, M. M. (1991) · Archives of General Psychiatry · doi:10.1001/archpsyc.1991.01810360024003 · Cited by 2,203 · cited 1×
“DBT was initially developed for persons diagnosed with borderline personality disorder [16], and there is increasing evidence to suggest that DBT is also effective in reducing suicidal thoughts, nonsuicidal self-injurious behavior, and suicide attempts in both adults [17] and adolescents [18,19].”
Background
Research works printed but not linked: 3
- 2014, Preventing Suicide: A Global Imperative
- 2019, WHO Guideline: recommendations on digital interventions for health system strengthening
- Wilson, C, 2005, Can J Couns, 39, 15
Cited by
Torok, M., Han, J., McGillivray, L., Wong, Q., Werner-Seidler, A., O’Dea, B., Calear, A., & Christensen, H. (2022) · PLoS Medicine · doi:10.1371/journal.pmed.1003978 · Cited by 71
“The trial protocol (S1 Protocol) was prospectively registered on the Australian New Zealand Clinical Trials Registry (ACTRN12619001671156) and has also been published elsewhere [24].”
BackgroundMethodSelf-citationThe Application of e-Mental Health in Response to COVID-19: Scoping Review and Bibliometric Analysis
Ellis, L. A., Meulenbroeks, I., Churruca, K., Pomare, C., Hatem, S., Harrison, R., Zurynski, Y., & Braithwaite, J. (2021) · JMIR Mental Health · doi:10.2196/32948 · Cited by 59
“However, the flexible and remote nature of e-mental health may offer a more viable option to certain populations who may have historically fallen through the cracks or struggled to engage with traditional psychology service provision, for example, marginalized populations, such as those identifying as LGBTQ+ (lesbian,…”
BackgroundSarubbi, S., Rogante, E., Erbuto, D., Cifrodelli, M., Sarli, G., Polidori, L., Lester, D., Berardelli, I., & Pompili, M. (2022) · Journal of Clinical Medicine · doi:10.3390/jcm11195616 · Cited by 42
“The main suicidal outcomes are suicidal ideation [53,55–58], self-harm [54] and suicide behaviors [55].”
BackgroundJiménez-Muñoz, L., Peñuelas‐Calvo, I., Díaz-Oliván, I., Gutiérrez-Rojas, L., Baca‐García, E., & Porras‐Segovia, A. (2022) · Harvard Review of Psychiatry · doi:10.1097/hrp.0000000000000331 · Cited by 42
Resilience to suicidal behavior in young adults: a cross-sectional study
Han, J., Wong, I., Christensen, H. M., & Batterham, P. J. (2022) · Scientific Reports · doi:10.1038/s41598-022-15468-0 · Cited by 38
Self-citationBaka, E., Tan, Y.-R., Wong, B. L. H., Xing, Z., & Yap, P. (2025) · Oxford Open Digital Health · doi:10.1093/oodh/oqaf005 · Cited by 31
E-Mental Health for People with Personality Disorders: A Systematic Review
Xie, Q., Torous, J. B., & Goldberg, S. B. (2022) · Current Psychiatry Reports · doi:10.1007/s11920-022-01360-1 · Cited by 23
Onie, S., Armstrong, S. O., Josifovski, N., Berlinquette, P., Livingstone, N., Holland, S., Finemore, C., Gale, N., Elder, E., Laggis, G., et al. (2023) · JMIR Mental Health · doi:10.2196/50283 · Cited by 7
“For example, apps can help people keep themselves safe during a suicidal crisis [1]; phone booths are installed at frequently used locations, which give rapid access to a suicide hotline [2]; or an online banner containing a suicide hotline number may appear XSL • FO RenderX if individuals search…”
BackgroundSelf-citationMcGillivray, L., Keng-Meng Hui, N., Wong, Q. J. J., Han, J., Qian, J., & Torok, M. (2023) · JMIR Mental Health · doi:10.2196/44862 · Cited by 7
“This paper is a secondary analysis of a clinical trial of the LifeBuoy smartphone intervention, with the full trial details, safety protocols [37], and main outcomes reported elsewhere [25].”
Self-citationLi, S., Achilles, M. R., Subotic-Kerry, M., Werner‐Seidler, A., Newby, J. M., Batterham, P. J., Christensen, H. M., Mackinnon, A. J., & O’Dea, B. (2022) · BMC Psychiatry · doi:10.1186/s12888-022-04383-3 · Cited by 9
“Attention-matched control conditions account for nonspecific intervention effects and are frequently used in other studies within the digital health field [39].”
BackgroundSelf-citationSotelo, K. V., Cárdenas, F. P., González, B. Z., Orozco, C. M. V., Rodríguez, I. H., & Pérez, C. L. (2023) · Frontiers in Psychology · doi:10.3389/fpsyg.2022.984374 · Cited by 9
“For this reason, disorders related to mental health, particularly those related to suicide, represent a global public health problem affecting adolescents and young adults, which has been considered the second leading cause of death in these age groups in recent years (Han et al., 2020; Penninx et al., 2021).”
BackgroundOkuboyejo, S. R., Han, S., Jha, S., Eneja, C., & Orji, R. (2025) · International Journal of Human-Computer Interaction · doi:10.1080/10447318.2025.2476711 · Cited by 3
“Furthermore, the scalability and accessibility of mobile apps make them an ideal platform for delivering suicide prevention interventions, especially for populations with limited access to traditional mental health services (Han et al., 2020).”
Yosep, I., Kurniawan, K., Maulana, I., Mardhiyah, A., Hazmi, H., & Hikmat, R. (2026) · International Journal of Africa Nursing Sciences · doi:10.1016/j.ijans.2026.101085 · Cited by 2
Resilience To Suicidal Behaviour In Young Adults: A Cross-Sectional StudyPreprint
Han, J., Wong, I., Christensen, H. M., & Batterham, P. J. (2022) · Research Square · doi:10.21203/rs.3.rs-1233434/v1
“Cognitive and behavioural responses to emotional distress were measured by 18 questions assessing participants’ likelihood to engage in the listed activities 53,54 using a four-point Likert scale, ranging from “Highly unlikely (1)” to “Highly likely (4)”.”
MethodSelf-citationDigital Mental Health Interventions for Children and Youth
Machado, A. S., Ferreira, B., Pereira, C. Y., Carrola, C., de Melo, B. B. V., & Campos, C. (2023) · Advances in medical technologies and clinical practice book series · doi:10.4018/978-1-6684-9251-2.ch002 · Cited by 2
Psychological Impact of COVID-19 on individuals: Through Active Choices and Passive Adaption
Fang, J., Lau, C. K. M., & Cheng, Y. (2024) · Frontiers research topics · doi:10.3389/978-2-8325-4481-5
Ludin, N. M., Torok, M. H., McGillivray, L., Wong, Q. J. J., Fortune, S. A., Meinhardt, I., Thabrew, H., Cargo, T., & Hetrick, S. E. (2026) · BMJ Open · doi:10.1136/bmjopen-2025-114884 · Cited by 1
“Users receive a push notification once a day to log their mood on a half-circle (with editable descriptors allowing young people to use their own words, based on a 0–10 scale) with an optional journal-style reflection feature.”
Self-citation
Metadata
- Authors
- Han, J., McGillivray, L., Wong, Q. J., Werner-Seidler, A., Wong, I., Calear, A., Christensen, H., & Torok, M.
- Affiliations
- Black Dog Institute (AU)
- Western Sydney University (AU)
- Australian National University (AU)
- Corresponding author
- Han, Jin
- ORCID
- Venue
- JMIR Research Protocols
- Publisher
- JMIR Publications Inc.
- Published
- 2020-10-27
- Volume
- 9
- Issue
- 10
- Pages
- e23655
- ISSN
- 1929-0748 (ISSN-L)
- PMC
- PMC7655466
- OpenAlex
- W3090389920
- Cited by
- 17
- References
- 49
- Citations by year
- 2026: 2 · 2025: 2 · 2024: 1 · 2023: 4 · 2022: 7 · 2021: 1
- FWCI
- 2.71
- Type
- journal article
- Language
- en
- Open access
- gold
- doi.org
- Licence
- cc-by
- Locations
- JMIR Research Protocols (publishedVersion, cc-by)
- PubMed Central (publishedVersion, cc-by)
- PubMed (publishedVersion)
- UNSWorks (University of New South Wales, Sydney, Australia) (submittedVersion, cc-by-nc-nd)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- ANU Open Research (Australian National University) (submittedVersion, cc-by)
- Funders
- Australian Rotary Health
- National Health and Medical Research Council (1122544, 1173146, 1155614, 1138710)
- Topics
- Social Sciences › Psychology › Applied Psychology › Digital Mental Health Interventions
- Social Sciences › Psychology › Clinical Psychology › Suicide and Self-Harm Studies
- Health Sciences › Health Professions › General Health Professions › Mobile Health and mHealth Applications
- Keywords
- mobile health intervention, suicidal ideation, youth suicide prevention, dialectical behavior therapy, young adults, depression, anxiety, psychological distress, psychological well-being, insomnia, rumination, distress tolerance, loneliness, help-seeking, COVID-19 pandemic, app acceptability
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 2026-10-11, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-11 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (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
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.
