EUVIMEDOpdag EUVIMED
Open access · CC BY-NC-ND 4.0

What Patients are Reading About Prenatal Cell‐Free DNA: An Analysis of Leading Publicly Available AI‐Chatbots

Miller S Finkelstein iD, Susannah L Rose iD, Lenicia Jenkins iD, Christina Collart iD, Ruth M Farrell iD

Prenatal Diagnosis · 2026-11-14

DOI: 10.1002/pd.70291 ↗

At a glance

Journal article in Prenatal Diagnosis, published online on 2026-11-14. Written by Miller S Finkelstein and 4 co-authors. Institution: Department of Subspeciality OB/GYN, Women and Children's Institute, Cleveland Clinic, Cleveland, Ohio, USA. 29 references. The full text is freely available (open access, CC BY-NC-ND 4.0). Funded by: NHGRI NIH HHS, National Human Genome Research Institute.

AbstractSubjectsBibliografiske oplysningerCiteRelated studies

Abstract (original language)

Objective

To characterize the quality, consistency, and framing of information about prenatal cell‐free DNA (cfDNA) screening generated by publicly available chatbots, including variation across simulated user age profiles.

Methods

We conducted a qualitative content analysis of outputs from three publicly available chatbots. To reduce carryover effects and personalization, prompts were entered using a VPN, private browsing, and cleared browser cookies. Outputs were generated using distinct simulated user age profiles. Analysis examined informational completeness, framing of benefits and limitations, terminology, source transparency, and age‐related differences in responses.

Results

Chatbot outputs varied substantially across platforms and user profiles, with misinformation, omissions, and incomplete information identified. Common themes included inconsistent explanations of the purpose and scope of cfDNA screening, frequent omission of test limitations and downstream implications, and variable readability scores (at a 9th–14th grade level) and terminology that sometimes conflate screening with diagnostic testing. Citations to high‐quality sources were limited or absent. Age‐based profile variation was associated with differences in tone, complexity, and emphasis, suggesting implicit tailoring of information presentation.

Conclusion

Publicly available chatbots may provide inaccurate or inconsistent information about cfDNA screening. As these tools are increasingly used outside clinical settings, improving accuracy, transparency, discussion of limitations, and source citation is essential to support reliable patient‐facing health information.

Subjects

Author keywords

cell‐free DNAcomplete informationcontent analysisgeneticshealth informationincomplete informationlarge language modelsmisinformationobstetricssource quality

Authors and institutions

  • Department of Subspeciality OB/GYN, Women and Children's Institute, Cleveland Clinic, Cleveland, Ohio, USA.
  • Department of Biomedical Informatics & Center for Biomedical Ethics and Society, Vanderbilt University Medical Center, Nashville, Tennessee, USA.
  • Center for Personalized Genetic Healthcare, Cleveland Clinic, Cleveland, Ohio, USA.
  • Center for Bioethics, Cleveland Clinic, Cleveland, Ohio, USA.

ORCID profiles: Miller S Finkelstein, Susannah L Rose, Lenicia Jenkins, Christina Collart, Ruth M Farrell

Funding

  • NHGRI NIH HHS
  • National Human Genome Research Institute

Cite

APA

Finkelstein, M. S., Rose, S. L., Jenkins, L., Collart, C., & Farrell, R. M. (2026). What Patients are Reading About Prenatal Cell‐Free DNA: An Analysis of Leading Publicly Available AI‐Chatbots. Prenatal Diagnosis. https://doi.org/10.1002/pd.70291

Vancouver

Finkelstein MS, Rose SL, Jenkins L, Collart C, Farrell RM. What Patients are Reading About Prenatal Cell‐Free DNA: An Analysis of Leading Publicly Available AI‐Chatbots. Prenatal Diagnosis. 2026. doi:10.1002/pd.70291

BibTeX
@article{Finkelstein2026,
  author = {Finkelstein, Miller S and Rose, Susannah L and Jenkins, Lenicia and Collart, Christina and Farrell, Ruth M},
  title = {What Patients are Reading About Prenatal Cell‐Free DNA: An Analysis of Leading Publicly Available AI‐Chatbots},
  journal = {Prenatal Diagnosis},
  year = {2026},
  doi = {10.1002/pd.70291}
}
Start for free →