{
  "abstract": "ObjectiveLarge language models (LLMs) are increasingly used to generate, summarize, and evaluate clinical documentation, yet standardized methods for evaluating nuanced, context-dependent clinical language remain limited. Existing approaches rely primarily on annotated clinical corpora or keyword-based methods, restricting reproducibility, public dissemination, and systematic comparison across models. We present ClinStigma, a clinician-guided framework for constructing, annotating, and evaluating synthetic benchmarks of contextual stigmatizing language, demonstrated through ClinStigma-SUD, its inaugural implementation for substance use disorder documentation.\n\nMaterials and MethodsCandidate clinical sentences were generated using a frontier language model and iteratively reviewed according to a clinician-developed annotation rubric refined through expert review in stigmatizing language. The resulting benchmark intentionally incorporates explicit stigmatizing, neutral, and contextually ambiguous (\"gray-zone\") documentation requiring clinical reasoning rather than keyword recognition. Three locally deployed open-weight LLMs were evaluated using zero-shot inference, followed by external evaluation using independently annotated clinical documentation from MIMIC-IV.\n\nResultsClinStigma-SUD consists of 500 clinician-adjudicated synthetic clinical documentation sentences. Inter-rater agreement was 98.8% across the full benchmark (Cohens {kappa} = 0.976) and 94.0% within the intentionally ambiguous subset ({kappa} = 0.880). Under zero-shot prompting, binary accuracy ranged from 84.8% to 91.8% across evaluated models, while F1 for the ambiguous (gray) class ranged from 0.038 to 0.423. Zero-shot external evaluation on 3,716 annotated examples from the independently annotated Harrigian MIMIC-IV corpus yielded 84.4% accuracy (84.4% balanced accuracy; macro F1 0.824), against a 69.9% majority-class baseline. Stigmatizing-class F1 was 0.879 in the substance-related subgroup and 0.883 in the full external corpus.\n\nDiscussion and ConclusionClinStigma establishes a reproducible clinician-guided framework for developing synthetic evaluation benchmarks for contextual clinical language. By combining structured clinical expertise, explicit annotation guidance, synthetic benchmark generation, and external evaluation, the framework enables open, privacy-preserving evaluation of language models while providing a scalable foundation for benchmark development across additional clinical specialties, including obstetrics, emergency medicine, and mental health.",
  "abstract_shown": true,
  "affiliations": [
    "Hospital Administration and Clinical Informatics, San Joaquin General Hospital, French Camp, CA, USA"
  ],
  "article_type": "PUBLISHAHEADOFPRINT",
  "arxiv": null,
  "author_total": 6,
  "authors": [
    {
      "affs": [
        0
      ],
      "name": "J. A. Izzo"
    },
    {
      "name": "A. M. McIntyre"
    },
    {
      "name": "J. Nguyen"
    },
    {
      "name": "D. Bashaw"
    },
    {
      "name": "C. A. Torrance"
    },
    {
      "name": "J. Foster"
    }
  ],
  "bibtex": "https://research.draconitic.com/paper/5gyw7y6jsp1cp.bib",
  "cited_by": 0,
  "doi": "10.64898/2026.10.10.26365233",
  "fulltext_url": "https://www.medrxiv.org/content/10.64898/2026.10.10.26365233v1.full",
  "id": "5gyw7y6jsp1cp",
  "issns": [],
  "landing_url": "https://www.medrxiv.org/content/10.64898/2026.10.10.26365233v1",
  "language": null,
  "licence": "CC-BY-NC",
  "licence_label": "CC BY-NC",
  "licence_source": "article",
  "licence_url": "https://creativecommons.org/licenses/by-nc/4.0/",
  "metadata_licence": null,
  "native_subjects": [
    [
      "medrxiv",
      "health informatics"
    ],
    [
      "openalex",
      "1702|Machine Learning in Healthcare"
    ]
  ],
  "openalex": "W7221538322",
  "pdf_url": "https://www.medrxiv.org/content/10.64898/2026.10.10.26365233v1.full.pdf",
  "pmcid": null,
  "pmid": null,
  "published": "2026-10-10",
  "publisher": null,
  "ris": "https://research.draconitic.com/paper/5gyw7y6jsp1cp.ris",
  "sources": [
    {
      "first_seen": "2026-10-30T12:20:42Z",
      "native_id": "10.64898/2026.10.10.26365233",
      "source": "medrxiv",
      "url": "https://www.medrxiv.org/content/10.64898/2026.10.10.26365233v1"
    }
  ],
  "subjects": [
    {
      "field": "medicine",
      "origin": "medrxiv:health informatics",
      "subfield": "epidemiology-public-health"
    }
  ],
  "title": "ClinStigma: A Clinician-Guided Framework for Developing Synthetic Benchmarks of Contextual Stigmatizing Language in Clinical Documentation",
  "title_text": "ClinStigma: A Clinician-Guided Framework for Developing Synthetic Benchmarks of Contextual Stigmatizing Language in Clinical Documentation",
  "type": "preprint",
  "updated_at": "2026-10-30T12:20:42Z",
  "url": "https://research.draconitic.com/paper/5gyw7y6jsp1cp",
  "venue": "medRxiv",
  "versions": []
}