Hearing loss has been proposed as a potentially modifiable risk factor for brain health, but whether hearing-related features act as causal levers for cognitive or psychiatric outcomes remains unresolved. This study examined how reported hearing symptoms fit into an ear-disease comorbidity network, and whether perceived or measured hearing deficits show evidence of causal relationships with brain-health outcomes using triangulated methods.
The authors constructed an ear-disease comorbidity network using NHANES data from 2011–2020 (N = 18,939). The resulting network included 16 nodes and 62 edges. Within that network, subjective hearing symptoms had a prevalence of 6.1% and occupied hub positions, indicating many comorbid connections with other ear-disease nodes. In contrast, objective hearing impairment had a prevalence of 8.0% and was sparsely connected in the same network.
To assess causality, the study performed bidirectional Mendelian randomization (MR) across 24 exposure–outcome pairs drawn from available genome-wide association summary statistics (sources cited include OpenGWAS and FinnGen). Forward MR tested whether hearing-related exposures influenced brain-health outcomes; reverse MR tested whether brain-health phenotypes influenced hearing outcomes. The authors report that all forward MR estimates were null after adjustment for multiple testing: inverse-variance weighted (IVW) P-values were greater than .05 for 9 of 9 forward pairs specifically noted. One reverse MR association (cognition → objective hearing) was nominally significant (beta = –0.15, P = .013) but did not survive multiple-testing correction.
As a longitudinal complement, the authors used CHARLS data (N = 17,101) to examine prospective associations between perceived hearing symptoms and incident brain-health outcomes. In this analysis, subjective hearing symptoms predicted incident depression with a hazard ratio (HR) of 1.57 and P = .00004, indicating a robust prospective association for this outcome in the dataset used.
Based on the pattern of results across network mapping, MR, and longitudinal analyses, the authors propose a "flag, not lever" conceptualization. Under this framework, perceived hearing symptoms function as clinical flags—markers that indicate systemic multimorbidity and the need for clinical attention—rather than as causal levers whose modification would directly improve downstream brain-health outcomes such as cognition or dementia risk.
Clinically, the findings suggest that when patients report hearing symptoms, clinicians should consider comprehensive assessment for comorbid conditions and mental-health risk (for example, depression), given the symptom's hub position in a comorbidity network and its prospective association with incident depression. From a research standpoint, the null MR results indicate that genetic-instrumented variation in hearing traits did not provide evidence for a causal effect on the brain-health phenotypes examined in this study, so trials targeting hearing as a strategy for dementia prevention remain unsupported by these genetic data. The authors frame subjective hearing as a marker to prompt broader evaluation rather than a proven intervention target for preventing cognitive decline.
All datasets used in the study are publicly available: NHANES 2011–2020 from the U.S. CDC/NCHS, CHARLS from Peking University (accessed under a data use agreement), and GWAS summary statistics from OpenGWAS and FinnGen. Analysis code and canonical network JSON files are provided in the supplementary materials. The study used de-identified secondary data and was determined to be exempt from IRB review by Chengdu University of Traditional Chinese Medicine. The authors declared no competing interests.
The source article reports the analytic approaches and main quantitative results cited above. Specific methodological details (for example, the full list of exposure–outcome pairs tested in MR, instrument strength metrics, and multiple-testing correction method) are provided in the article and supplementary materials; where not restated here, those details were not exhaustively reported in the abstract and summary text used as the source for this rewrite.
Perceived hearing symptoms are centrally placed in an ear-disease comorbidity network and predict incident depression prospectively, but triangulated genetic and longitudinal analyses do not support subjective hearing symptoms as a causal lever for brain-health outcomes. The authors recommend treating subjective hearing as a clinical flag that warrants attention to multimorbidity rather than as an established target for dementia-prevention interventions.