Readability of Chatbot Responses in Prostate Cancer and Urological Care: Objective Metrics Versus Patient Perceptions.

Simple Summary: Large language models (LLMs) are being used more often in healthcare, but we still know little about how well they help patients understand medical information. Until now, no study has asked patients directly how clear they find information generated by LLMs. To address this, we cond...

Descripción completa

Detalles Bibliográficos
Publicado en:Current Oncology Vol. 32; no. 10; pp. 582 - 595
Autores principales: Maywald, Lasse, Nguyen, Lisa, Winterstein, Jana Theres, Hetz, Martin Joachim, Mangold, Maurin Helen, Renner, Luisa Vivienne, Brinker, Titus Josef, Wessels, Frederik, Carl, Nicolas
Formato: Journal Article
Publicado: MDPI Oct2025
Acceso en línea:Ver este registro en EBSCOhost
fields @attributes:
  recordID: 1
pdfLink:
plink: https://search.ebscohost.com/login.aspx?direct=true&db=ccm&AN=188957488&site=ehost-live
header:
  @attributes:
    shortDbName: ccm
    uiTerm: 188957488
    longDbName: CINAHL Complete
    uiTag: AN
  controlInfo:
    bkinfo:
    dissinfo:
    jinfo:
      jid:
        11980052
        5EKK
      jtl: Current Oncology
      issn: 11980052
      maglogo: N
    pubinfo:
      dt: Oct2025
      vid: 32
      iid: 10
      pid: 97109
      pub: MDPI
    artinfo:
      ui:
        188957488
        10.3390/curroncol32100582
        188957488
      ppf: 582
      ppct: 13
      formats:
        fmt:
          – @attributes:
              type: T
          – @attributes:
              type: P
      tig:
        atl: Readability of Chatbot Responses in Prostate Cancer and Urological Care: Objective Metrics Versus Patient Perceptions.
      aug:
        au:
          Maywald, Lasse
          Nguyen, Lisa
          Winterstein, Jana Theres
          Hetz, Martin Joachim
          Mangold, Maurin Helen
          Renner, Luisa Vivienne
          Brinker, Titus Josef
          Wessels, Frederik
          Carl, Nicolas
        affil: Department of Urology, University Medical Center Mannheim, 68167 Mannheim, Germany
      sug:
      ab: Simple Summary: Large language models (LLMs) are being used more often in healthcare, but we still know little about how well they help patients understand medical information. Until now, no study has asked patients directly how clear they find information generated by LLMs. To address this, we conducted a survey comparing how readable chatbot responses were with how easy patients said they were to understand. Knowledge retention was not assessed. Urology patients used a voice-based chatbot to ask medical questions during regular clinic visits. We analyzed transcripts from 231 conversations. While chatbot answers were written at a level higher than typically recommended for patient education, most patients still found the information understandable. We believe this may be due to how patients received the information, by both reading and listening, which might make it easier to understand. These findings suggest the benefit of multimodal designs in health tools to meet patients' diverse needs. Large language models (LLMs) are increasingly explored as chatbots for patient education, including applications in urooncology. Since only 12% of adults have proficient health literacy and most patient information materials exceed recommended reading levels, improving readability is crucial. Although LLMs could potentially increase the readability of medical information, evidence is mixed, underscoring the need to assess chatbot outputs in clinical settings. Therefore, this study evaluates the measured and perceived readability of chatbot responses in speech-based interactions with urological patients. Urological patients engaged in unscripted conversations with a GPT-4-based chatbot. Transcripts were analyzed using three readability indices: Flesch–Reading-Ease (FRE), Lesbarkeitsindex (LIX) and Wiener-Sachtextformel (WSF). Perceived readability was assessed using a survey covering technical language, clarity and explainability. Associations between measured and perceived readability were analyzed. Knowledge retention was not assessed in this study. A total of 231 conversations were evaluated. The most frequently addressed topics were prostate cancer (22.5%), robotic-assisted prostatectomy (19.9%) and follow-up (18.6%). Objectively, responses were classified as difficult to read (FRE 43.1 ± 9.1; LIX 52.8 ± 6.2; WSF 11.2 ± 1.6). In contrast, perceived readability was rated highly for technical language, clarity and explainability (83–90%). Correlation analyses revealed no association between objective and perceived readability. Chatbot responses were objectively written at a difficult reading level, exceeding recommendations for optimized health literacy. Nevertheless, most patients perceived the information as clear and understandable. This discrepancy suggests that perceived comprehensibility is influenced by factors beyond measurable linguistic complexity.
      pubtype: Academic Journal
      doctype: Journal Article
      ougenre: Article
    language: English
    refInfo:
    holdings:
      @attributes:
        islocal: N