Temporal associations between treated and untreated hearing loss and mild behavioral impairment in older adults without dementia
Penny Gosselin, Dylan X. Guan, Eric E. Smith, Zahinoor Ismail- Psychiatry and Mental health
- Cellular and Molecular Neuroscience
- Geriatrics and Gerontology
- Neurology (clinical)
- Developmental Neuroscience
- Health Policy
- Epidemiology
Abstract
Background
Hearing loss and mild behavioral impairment (MBI) are non‐cognitive markers of dementia that are cross‐sectionally linked to one another. Longitudinal approaches are necessary to further understand mechanisms underlying their association, which may inform interventions for hearing loss and MBI. This study investigated the relationship between hearing loss, with or without hearing aids (HAs), and MBI in dementia‐free older adults both cross‐sectionally and longitudinally.
Method
Data were analyzed from dementia‐free National Alzheimer’s Coordinating Center participants, aged ≥50, collected between 2005‐2022. Three self‐report questions were used to generate a three‐level categorical hearing status variable: No‐HL, Untreated‐HL (not wearing hearing aids), and Treated‐HL (regularly wearing hearing aids). MBI status was derived from the informant‐rated Neuropsychiatric Inventory Questionnaire using a published algorithm. At baseline (n = 7992), age‐, sex‐, education‐, and cognitive status‐adjusted logistic regression was used to examine the association between hearing status and presence of global and domain‐specific MBI. Cox proportional hazards models examined the effect of: 1) hearing status as exposure on rate of incident MBI (n = 6639); and 2) MBI status as exposure on rate of incident hearing loss (n = 7343).
Result
Cross‐sectionally, participants with Untreated‐HL (not wearing hearing aids) were more likely to exhibit global MBI (OR = 1.58) and individual MBI domains of social inappropriateness (OR = 2.08), affective dysregulation (OR = 1.64), and impulse dyscontrol (OR = 1.68), compared to those with No‐HL (Figure 1). Treated‐HL (regularly wearing hearing aids) did not differ from No‐HL for presence of global‐ or domain‐specific MBI. Longitudinally, older adults with Treated‐HL developed MBI at higher rates than No‐HL (HR = 1.42), and older adults with MBI at baseline developed Treated‐HL at higher rates than those without MBI (HR = 1.34) [Table 1].
Conclusion
Our cross‐sectional results indicate that HA use in dementia‐free hearing‐impaired older adults is associated with lower likelihood of having MBI. However, hearing loss requiring hearing aids was associated with incident MBI and vice versa. This bidirectional relationship may indicate a positive feedback loop between hearing loss and MBI, with Treated‐HL and MBI reflecting greater hearing loss severity and disease burden, increasing the rate of development of the complementary marker and of dementia.