CoChat
Sign Up
Cited by 252Electronic Health Records SystemsMedical Coding and Health InformationPrimary Care and Health Outcomes

Adoption of Social Determinants of Health EHR Tools by Community Health Centers

Open access
Gold, Rachel, Bunce, Arwen, Cowburn, Stuart, Dambrun, Katie, Dearing, Marla, Middendorf, Mary, Mossman, Ned, Hollombe, Celine +5 more·2018·The Annals of Family Medicine
DOI 10.1370/afm.2275PMC PMC6131002
Open this paper in CoChat — free accountOpen PDF

Abstract

PURPOSE: This pilot study assessed the feasibility of implementing electronic health record (EHR) tools for collecting, reviewing, and acting on patient-reported social determinants of health (SDH) data in community health centers (CHCs). We believe it is the first such US study. METHODS: We implemented a suite of SDH data tools in 3 Pacific Northwest CHCs in June 2016, and used mixed methods to assess their adoption through July 2017. We modified the tools at clinic request; for example, we added questions that ask if the patient wanted assistance with SDH needs. RESULTS: Social determinants of health data were collected on 1,130 patients during the study period; 97% to 99% of screened patients (n = 1,098) had ≥1 SDH need documented in the EHR, of whom 211 (19%) had an EHR-documented SDH referral. Only 15% to 21% of patients with a documented SDH need indicated wanting help. Examples of lessons learned on adoption of EHR SDH tools indicate that clinics should: consider how to best integrate tools into existing workflow processes; ensure that staff tasked with SDH efforts receive adequate tool training and access; and consider that timing of data entry impacts how and when SDH data can be used. CONCLUSIONS: Our results indicate that adoption of systematic EHR-based SDH documentation may be feasible, but substantial barriers to adoption exist. Lessons from this study may inform primary care providers seeking to implement SDH-related efforts, and related health policies. Far more research is needed to address implementation barriers related to SDH documentation in EHRs.

AI snapshot

Written from the paper’s full text.

TL;DR

This pilot implemented EHR tools for collecting, reviewing, and acting on patient-reported social determinants of health (SDH) data in 3 Pacific Northwest community health centers. It found systematic EHR-based SDH documentation feasible but identified substantial adoption barriers, with high positive screening rates but low referral and patient-requested-help rates.

Key findings

  • SDH data were collected on 1,130 patients, representing 4% to 18% of adult patients with an ambulatory visit at the pilot clinics; 97% to 99% of screened patients had ≥1 SDH need documented, and 211 (19%) had an EHR-documented SDH referral.
  • After questions about wanting help were added, 15% of Clinic A patients and 21 of Clinic B patients with a documented SDH need requested help.
  • Facilitators included a trusted, EHR-savvy clinic champion, ability to customize EHR views, and reporting requirements; barriers included fragmented patient data, added workflow difficulty, paper-based data entry, time-consuming referrals, and unmanageable follow-up workload before desire-for-help questions were added.
  • The clinics successfully documented SDH in the EHR and continued post-study, but referral documentation was especially challenging.

Methods

  • Concurrent mixed-methods pilot at 3 OCHIN member community health centers in the Pacific Northwest; Epic-based SDH tools built on PRAPARE were deployed in June 2016 and assessed through July 2017, extended from 12 to 13 months.
  • Quantitative EHR data described adult patients with documented SDH data or ≥1 ambulatory visit, including demographics, SDH needs, referrals, and problem list diagnoses.
  • Qualitative data included 6 days of workflow observation, 24 interviews with care team members, and observations of clinic decision-maker discussions; analysis used deductive and inductive coding and the constant comparative method.

Limitations

  • The paper states it could not calculate SDH screening documentation rates among targeted patients because target populations changed, change dates were unclear, and target criteria were not easily extracted from the EHR; screened vs unscreened differences therefore cannot be interpreted as related to EHR-based SDH screening.
  • The paper states screened-patient counts may be low and SDH need prevalence may be higher than in the overall population because some paper surveys were likely not entered into the EHR; referral rates are likely underestimated because referrals were often documented in chart notes rather than discrete fields.
  • The paper states it could not assess changes in SDH status, care quality, or biomarkers; could not quantify how often the SDH summary was accessed or by whom; the SDH screening measures have not been validated as predictive of specific outcomes; and the single EHR system limits transferability. Evident from the text: this was a small 3-clinic pilot with clinic-selected pragmatic target groups and no control or comparison group.

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

Integrity signals

Registry

Crossref · 10.1370/afm.2275

Cites

55 of 55 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Claims the paper makes

0 of 25 cited sources checked · 0 supported · 0 flagged

  • Findingown data

    A key barrier was perceptions that EHR-based SDH data tools could add a layer of difficulty to collecting and acting on SDH data.

    “(2) could add a layer of difficulty to collecting and acting on SDH data; (eg, due to lack of staff EHR expertise, the tools' customized nature, differences in EHR security access by staff role, and the need for new EHR competencies from some staff)”
    Quote in the paperSubstantial claim
  • Findingown data

    Only 15% to 21% of patients with a documented SDH need indicated wanting help.

    “Only 15% to 21% of patients with a documented SDH need indicated wanting help.”
    • Also stated in: § SDH Screening Documentation and Referrals · plain
      “15% of Clinic A patients and 21% of Clinic B patients with a documented need requested help.”
    Quote in the paperNot a substantial claim
  • Findingown data

    97% to 99% of screened patients had at least one SDH need documented in the EHR.

    “97% to 99% of screened patients (n = 1,098) had ≥1 SDH need documented in the EHR”
    • Also stated in: § SDH Screening Documentation and Referrals · hedged
      “Most screened patients (97% to 99%) indicated a potential SDH need in ≥1 SDH domain”
    Quote in the paperNot a substantial claim
  • Findingnone given

    Referral documentation proved especially challenging.

    “Referral documentation proved especially challenging.”
    Quote in the paperNot a substantial claim
  • Findingnone given

    All 3 clinics sought to minimize the primary care provider's role in SDH documentation, and in making SDH-related referrals.

    “(3) sought to minimize the primary care provider's role in SDH documentation, and in making SDH-related referrals.”
    Quote in the paperNot a substantial claim
  • May combine two claims. 26
  • Classified as probably not a substantial claim. 23

  • May combine two claims.

    Findingown data

    The study clinics' patient populations are primarily publicly insured or uninsured and aged <65 years; 10% to 31% are non-white and 10% to 30% are of Hispanic ethnicity.

    “Like most CHCs, the study clinics' patient populations are primarily publicly insured or uninsured and aged <65 years; 10% to 31% are non-white and 10% to 30% are of Hispanic ethnicity (Table 2).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingnone given

    Despite the many identified challenges to adopting all of the SDH data tools, the study clinics did successfully document SDH in the EHR, and continued to do so post-study.

    “Despite the many identified challenges to adopting all of the SDH data tools, the study clinics did successfully document SDH in the EHR, and continued to do so post-study.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Clinics B and C routed most SDH referrals to community health workers or similar staff; these internal referrals generally lacked the information required to match them to a specific SDH need.

    “Clinics B and C routed most SDH referrals to community health workers or similar staff; these internal referrals generally lacked the information required to match them to a specific SDH need.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    In Clinic B, 379 patients were screened and 367 (96.8%) had a positive SDH screen.

    “B 379 367 (96.8) 26 (6.8)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In Clinic A, 602 patients were screened and 583 (96.8%) had a positive SDH screen.

    “A 602 583 (96.8) 141 (23.4)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    In Clinic C, 149 patients were screened and 148 (99.3%) had a positive SDH screen.

    “C 149 148 (99.3) 44 (29.5)”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingnone given

    Facilitators to using EHR tools to document SDH included a clinic champion who was trusted, EHR-savvy, able to customize EHR views to support workflows, and willing to use adoption data to iterate workflows.

    “Facilitators to using EHR tools to document SDH included a clinic champion who was trusted, EHR-savvy, able to customize EHR views to support workflows, and willing to use adoption data to iterate workflows.”
    Quote in the paperSubstantial claim

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

    Findingnone given

    Clinic A first tested their workflows with a few randomly chosen patients per day, then added all new patients plus those completing annual insurance reauthorizations, and eventually added patients receiving care coordination, HIV services, and behavioral health services.

    “Clinic A first tested their workflows with a few randomly chosen patients per day, then added all new patients plus those completing annual insurance reauthorizations, and eventually added patients receiving care coordination, HIV services, and behavioral health services.”
    Quote in the paperNot a substantial claim

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

    Findingnone given

    Clinic B began with patients in their diabetes and Hepatitis C case management programs, then added patients aged 65 years and older.

    “Clinic B began with patients in their diabetes and Hepatitis C case management programs, then added patients aged 65 years and older.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    A key barrier was perceptions that EHR-based SDH data tools created a fragmented view of the patient, with relevant data in multiple places, and did not readily support documenting a narrative about a given patient.

    “Key barriers included perceptions that EHR-based SDH data tools: (1) created a fragmented view of the patient, with relevant data in multiple places, and did not readily support documenting a narrative about a given patient”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing), S4 (based). May combine two claims.

    Findingnone given

    The completed paper form does not reveal how a patient scored on the SDH domains; this scoring is not easily identifiable until the data are in the EHR.

    “Consider that the completed paper form does not reveal how a patient scored on the SDH domains; this scoring is not easily identifiable until the data are in the EHR.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    SDH data were collected on 1,130 patients during the study period, representing 4% to 18% of adult patients with an ambulatory visit at the pilot clinics.

    “Social determinants of health data were collected on 1,130 patients during the study period”
    • Also stated in: § SDH Screening Documentation and Referrals · plain
      “Social determinants of health data were collected on 1,130 patients, representing 4% to 18% of adult patients with an ambulatory visit at the pilot clinics during the study period (Table 2 and Table 3).”
    Quote in the paperNot a substantial claim

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

    Findingnone given

    Text shortcut tools for documenting SDH referrals do not provide agency details but are a fast and easy way to document SDH referrals.

    “Consider using text shortcut tools to document provision of community resource materials or referrals provided by SDH topic (this does not provide agency details, but is a fast and easy way to document SDH referrals).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    All 3 clinics principally used the paper-based SDH screening questionnaire, necessitating a data entry workflow step that impacted when the data were available in the EHR.

    “(2) principally used the paper-based SDH screening questionnaire, necessitating a data entry workflow step that impacted when the data were available in the EHR”
    Quote in the paperSubstantial claim

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

    Findingnone given

    Each clinic first screened a limited population of patients, then scaled up.

    “Each clinic first screened a limited population of patients, then scaled up.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S4 (based). May combine two claims.

    Premisenone given

    This pilot study is the first to formally test adoption of EHR-based SDH data documentation, review, and action in CHCs, and is one of the first in any clinical setting.

    “This pilot study is the first to formally test adoption of EHR-based SDH data documentation, review, and action in CHCs, and is one of the first in any clinical setting.”
    Quote in the paperNot a substantial claim

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

    Interpretationcitation

    While some aspects of the tools were not widely adopted, this is common to such efforts; adoption of new EHR tools can take time.

    “While some aspects of the tools were not widely adopted, this is common to such efforts [58][59][60][61] ; adoption of new EHR tools can take time.”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [58]

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

    • cites [59]

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

    • cites [60]

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

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationreasoning

    For the benefits of systematic EHR-based SDH documentation to be realized, barriers to adoption of EHR tools must be addressed.

    “however, for the benefits of systematic EHR-based SDH documentation to be realized, barriers to adoption of EHR tools must be addressed.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Interpretationcitation

    SDH documentation, review, and referral-making in EHRs can be feasible and useful, but is challenging to adopt.

    “SDH documentation, review, and referral-making in EHRs can be feasible and useful, but is challenging to adopt.63”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites [63]

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

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

    Interpretationnone given

    Some barriers to SDH screening are likely to occur regardless of the technology used, for example, staff concerns about SDH screening when little intervention is possible.

    “Some barriers to SDH screening are likely to occur regardless of the technology used, for example, staff concerns about SDH screening when little intervention is possible.”
    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 conductcitation

    Analysis followed the constant comparative method; new data were compared with previously collected/coded data to detect commonalities/differences, and to develop nuanced understandings of clinic experiences.

    “Analysis followed the constant comparative method 46 ; new data were compared with previously collected/coded data to detect commonalities/differences, and to develop nuanced understandings of clinic experiences.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [46]

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

  • May combine two claims.

    Study design and conductown data

    Some clinic counts of patients screened did not match the quantitative data from the EHR, and data entry workflows were not always standardized.

    “For example, some clinic counts of patients screened did not match our quantitative data from the EHR; further, we observed that data entry workflows were not always standardized.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    Regular study team meetings and discussions with stakeholders informed each step of analysis and results interpretation.

    “Regular study team meetings and discussions with stakeholders informed each step of analysis and results interpretation.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    SDH documentation in EHRs could enable care plan adjustments/referrals to social services, and tracking such referrals' results.

    “Social determinants of health documentation in EHRs could enable care plan adjustments/referrals to social services, and tracking such referrals' results.”
    Quote in the paperNot a substantial claim

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

    Study design and conductnone given

    The study was approved by the Kaiser Permanente Northwest Institutional Review Board.

    “This study was approved by the Kaiser Permanente Northwest Institutional Review Board.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Study design and conductown data

    The study could not calculate rates of SDH screening documentation among targeted patients because the study clinics' target populations changed during the study period, the dates of these changes were unclear, and target criteria were not easily extracted from the EHR.

    “We could not calculate rates of SDH screening documentation among targeted patients because the study clinics' target populations changed during the study period, the dates of these changes were unclear, and target criteria were not easily extracted from the EHR.”
    Quote in the paperSubstantial claim

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

    Study design and conductown data

    Quantitative and qualitative data were collected simultaneously, and preliminary results from each informed subsequent data collection.

    “Quantitative and qualitative data were collected simultaneously.Preliminary results from each informed subsequent data collection (eg, discrepancies between interview descriptions of clinic workflows and corresponding quantitative data were further explored in site visits).”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    The systematic EHR documentation and management of SDH needs could impact CHC patients' health, but optimizing such EHR tools and integrating them smoothly into clinic workflows will require addressing substantial knowledge gaps.

    “The systematic EHR documentation and management of SDH needs could impact CHC patients' health, but optimizing such EHR tools and integrating them smoothly into clinic workflows will require addressing substantial knowledge gaps.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationcitation

    Before benefits can be fully realized, the unique challenges of EHR-based SDH documentation and referral-making must be understood and addressed.

    “Before benefits can be fully realized, the unique challenges of EHR-based SDH documentation and referral-making must be understood and addressed.1,39,48”
    Quote in the paperNot a substantial claim0 of 3 sources checked
    • cites [1]

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

    • cites [39]

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

    • cites [48]

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

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

    Study design and conductnone given

    Data collection and analysis were iterative and inductive; emergent understanding and questions from preliminary findings were explored.

    “Data collection and analysis were iterative and inductive; emergent understanding and questions from preliminary findings were explored.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    To meet the growing national emphasis on EHR-based SDH documentation and referral, numerous factors must be considered.

    “Therefore, to meet the growing national emphasis on EHR-based SDH documentation and referral, numerous factors must be considered.”
    Quote in the paperNot a substantial claim

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

    Interpretationcitation

    Our findings are consistent with those of a prior case study on EHR-based SDH documentation.

    “Our findings are consistent with those of a prior case study on EHR-based SDH documentation.62”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites [62]

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

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

    Study design and conductnone given

    Final results were compared to validate and confirm analytic interpretations.

    “Final results were compared to validate and confirm analytic interpretations.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationown data

    Our results raise important questions about the feasibility of EHR-based SDH documentation per national recommendations.

    “Our results raise important questions about the feasibility of EHR-based SDH documentation per national recommendations.”
    Quote in the paperNot a substantial claim

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

    Interpretationreasoning

    Results may be informative to CHCs and other primary care providers seeking to implement SDH-related efforts, especially if SDH documentation becomes required or associated with financial incentives.

    “Results may be informative to CHCs and other primary care providers seeking to implement SDH-related efforts, especially if SDH documentation becomes required or associated with financial incentives.”
    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 conductreasoning

    The count of screened patients may be low, and prevalence of SDH needs among screened patients may be higher than in the clinics' overall population, because some data from paper surveys were likely not entered into EHRs, especially if no SDH needs were identified.

    “It is possible that our count of screened patients may be low, and prevalence of SDH needs among screened patients may be higher than in the clinics' overall population, as some data from paper surveys were likely not entered into EHRs, especially if no SDH needs were identified.”
    Quote in the paperNot a substantial claim
36 of 36

  • The authors state this as an assumption.

    The study is the first US study on the collection of SDH data using EHR-based tools in community health centers.

    “We believe it is the first such US study.”
    “We believe this is the first US study on the collection of SDH data using EHR-based tools in CHCs.”

  • The paper relies on this without support.

    Most health care systems document and track most patient information in EHRs.

    “Since most health care systems document and track most patient information in EHRs, SDH data should also be in EHRs.”
2 of 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?

  • Prior studies do not include measures that reflect whether SDOH data are reviewed after being documented.

    “They also do not include measures that reflect whether SDOH data are reviewed after being documented.”

    From Documentation and review of social determinants of health data in the EHR: measures and associated insights (2021) · cites this paper as [13]

    The Limitations section says this study could not quantify how often the SDH summary was accessed or by whom, supporting the narrower point that it lacked a measure of review after documentation. The paper does not establish that prior studies generally lacked such measures. (LLM, conf=0.84)

    Experimental — accuracy not yet measured

  • Prior studies reveal the value of measuring different approaches to SDOH documentation in the EHR but do not offer a generalizable set of measures that can be used to compare different EHR-based SDOH documentation options.

    “These studies reveal the value of measuring different approaches to SDOH documentation in the EHR but do not offer a generalizable set of measures that can be used to compare different EHR-based SDOH documentation options.”

    From Documentation and review of social determinants of health data in the EHR: measures and associated insights (2021) · cites this paper as [13]

  • Simply adding dedicated SDOH fields and tools to EHRs is insufficient to drive increased documentation of social needs.

    “This suggests that simply adding dedicated SDOH fields and tools to EHRs is insufficient to drive increased documentation of social needs, a conclusion consistent with other publications.13,21”

    From Documentation and review of social determinants of health data in the EHR: measures and associated insights (2021) · cites this paper as [13]

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. U. S. Department of Health & Human Services (DHHS). Healthy People 2010, 2nd Edition. Washington, DC: US Government Printing Office; 2000.

Cited by

Sorted by citations

Related

Authors
Gold, R., Bunce, A., Cowburn, S., Dambrun, K., Dearing, M., Middendorf, M., Mossman, N., Hollombe, C., Mahr, P., Melgar, G., Davis, J., Gottlieb, L., & Cottrell, E.
Affiliations
  • Kaiser Permanente Center for Health Research (US)
  • Ochin (US)
  • Multnomah County Health Department (US)
  • University of California, San Francisco (US)
ORCID
Venue
The Annals of Family Medicine
Publisher
Annals of Family Medicine
Published
2018-09
Volume
16
Issue
5
Pages
399-407
ISSN
  • 1544-1709 (ISSN-L)
  • 1544-1717
DOI
10.1370/afm.2275
PMC
PMC6131002
OpenAlex
W2891941547
Semantic Scholar
ec90bbff459f6d579a5de4a0ee6586f1aebfa662
Cited by
252
References
55
Citations by year
2026: 21 · 2025: 32 · 2024: 25 · 2023: 38 · 2022: 38 · 2021: 42 · 2020: 31 · 2019: 23 · 2018: 2
FWCI
31.11
Type
journal article
Language
en
Open access
Locations
  • The Annals of Family Medicine (publishedVersion)
  • Europe PMC (PubMed Central) (publishedVersion)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
  • eScholarship (California Digital Library) (submittedVersion)
Funders
National Institute of Diabetes and Digestive and Kidney Diseases (R18DK105463)
Topics
  • Health Sciences › Health Professions › Health Information Management › Electronic Health Records Systems
  • Health Sciences › Health Professions › Health Information Management › Medical Coding and Health Information
  • Health Sciences › Health Professions › General Health Professions › Primary Care and Health Outcomes
Keywords
social determinants of health, electronic health records, community health centers, social needs screening, referral documentation, implementation barriers, primary care, health policy
MeSH terms
Community Health Centers, Documentation, Feasibility Studies, Health Plan Implementation, Humans, Pilot Projects, Referral and Consultation, Electronic Health Records, Social Determinants of Health

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 2026-10-28.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-10-28.

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