Skip to main content
AAO-HNSF 2025 Annual Meeting & OTO EXPO Main banner
Icon Legend
This session is not in your schedule.
This session is in your schedule. Click again to remove it.

Late-Breaking Scientific Oral Presentations

Comprehensive Otolaryngology

Session: Late Breaking Scientific Orals 1

The Predictive Role of Hearing Loss in Dementia

Saturday, October 11, 2025
2:48 PM - 2:54 PM EDT
Location: 245
Introduction: To determine the predictive role of hearing loss in dementia using machine learning algorithm.

Methods: The All of Us database was queried for participants with hearing loss and impaired cognition. These participants were then matched to controls who reflected the demographic makeup of the United States census. Demographics, medical history data, hearing loss and dementia diagnoses, and all hearing loss-related survey questions were extracted and compared among participants. A logistic regression model using optimized parameters was used to identify variables that were strongly associated with the development of dementia.

Results: Of the 16,298 participants included, 5,958 had sensorineural hearing loss, and 1,252 were diagnosed with dementia. Among the variables with significant p values (p < 0.001), nicotine dependence, gender, having a child with severe hearing loss, surveyed severe hearing loss, sensorineural hearing loss, and smoking had the largest effect sizes. Of the variables with large effect sizes, the strongest contributors to a prediction of dementia diagnoses were surveyed severe hearing loss (OR=3.565, 95% CI: 3.207-3.964, p< 0.001), sensorineural hearing loss (OR=3.466, 95% CI: 3.084-3.894, p< 0.001), and smoking (OR=3.260, 95% CI: 2.712-3.918, p< 0.001). The logistic regression classifier considering hearing loss related survey questions, hearing loss diagnosis, and demographic variables performed with an overall accuracy of 72.2%, sensitivity of 63%, specificity of 96%, and had an AUC-ROC area of 0.71.

Conclusions: Sensorineural hearing loss was strongly associated with dementia, with severe hearing loss being the single strongest predictive factor for a dementia diagnosis. Qualitative factors such as surveys related to hearing loss in addition to demographic variables were effective in predicting a diagnosis of dementia.

    Presenting Author(s)

  • AD

    Abigail I. Dichter, BA

    Medical Student
    University of California, Irvine
    Newport Beach, California, United States

    Disclosure information not submitted.

    • Co-Author(s)

  • TP

    Timothy Park, BA

    University of California, Irvine, United States

    Disclosure information not submitted.

  • EG

    Elham Ghanbarian, MD, PhD

    University of California, Irvine, United States

    Disclosure information not submitted.

  • YY

    Yalda Yazdani, MD

    Postdoctoral Research Fellow
    University of California, Irvine
    Orange, CA, United States

    Disclosure(s):

    faculty for this accredited education activity has no relationship(s) with ineligible companies to disclose.

  • SN

    Saharnaz Nedjat, MD, PhD

    University of California, Irvine, United States

    Disclosure information not submitted.

    • Senior Author(s)

  • Mehdi Abouzari, MD, PhD

    Assistant Professor
    University of California, Irvine
    Orange, California, United States

    Disclosure information not submitted.