16-year study links worsening vision to higher dementia risk

By tracking both eyesight and cognitive health, researchers examined whether the severity and pace of vision loss could help identify older adults who need closer monitoring.

Study: Severity and progression of visual impairment and risk of dementia. Image Credit: Kunlathida6242 / Shutterstock

In a recent study published in the journal Alzheimer’s & Dementia: Diagnosis, Assessment & Disease Monitoring, researchers investigated whether visual impairment is statistically associated with incident dementia risk.

The long-term (16-year follow-up) study evaluated 15,576 community-dwelling older adults to determine whether visual impairment severity and longitudinal rates of visual acuity decline were associated with subsequent dementia.

Moderate to severe visual impairment was independently associated with heightened dementia risk among participants. Faster visual decline added to this association at poorer visual acuity levels. Severe visual impairment was linked to an estimated 2.1 fewer dementia-free years over the study period compared with normal vision.

These findings suggest that visual acuity could help assess subsequent dementia risk, but correcting vision has not been shown to prevent dementia.

Background

Visual impairment is common in older adults. Research identifies sensory deficits as potential dementia risk factors, yet visual impairment frequently remains underdiagnosed and undertreated.

Previous investigations of associations between visual impairment and dementia risk have largely relied on cross-sectional or retrospective datasets or treated visual impairment as a binary variable (present versus absent), leaving the roles of impairment severity and progressive vision loss poorly understood.

Standard cognitive screening instruments often rely on visual tasks, increasing the risk of misclassifying visual deficits as dementia.

In addition, neurodegenerative pathologies can have a preclinical phase lasting a decade or more, meaning visual disturbances could reflect reverse causation in which preclinical neurodegeneration precedes both visual and cognitive decline.

About the study

The present study aimed to address this knowledge gap through a territory-wide longitudinal study in Hong Kong that tracked participants for up to 16 years.

Participants were community-dwelling Chinese adults aged 65 years or older (n = 15,576; mean age 74.5 years; 63.8% female) attending 18 government Elderly Health Centers (EHCs) who were clinically confirmed to be free of dementia at baseline. Dementia diagnoses were established using International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) criteria or a Clinical Dementia Rating (CDR) score of 1–3. Assessments included nonvisual cognitive tests and clinical history, and geriatric psychiatrists independently reviewed diagnoses.

Participants’ habitual distance visual acuity (at baseline and follow-up) was assessed using a standard 6-meter Snellen E chart (expressed in logarithm of the minimum angle of resolution [LogMAR] units). Participants wore their usual optical correction; higher LogMAR values indicate poorer vision.

Visual acuity results were used to categorize participants into four severity groups: 1. Normal vision (<0.3 LogMAR), 2. Mild impairment (0.3 to 0.47 LogMAR), 3. Moderate impairment (0.48 to 0.99 LogMAR), and 4. Severe impairment (≥1.00 LogMAR).

Statistical analyses included multivariable Cox proportional hazards models. To reduce reverse causation concerns, the final model excluded dementia cases occurring within the first 6 years and participants whose dementia-free status at year 6 could not be confirmed. It assessed dementia during years 7–16.

Study Findings

During the study’s 168,434 person-years of follow-up, 3,643 participants (23.4%) developed incident dementia. Poorer baseline visual acuity was significantly associated with incident dementia (hazard ratio [HR] = 1.37 per one-unit LogMAR increase, p = 0.02), after adjustment and the six-year exclusion. The association persisted when accounting for death as a competing event.

Severity analyses suggested a graded association, in which participants with moderate visual impairment had a 42% higher dementia hazard (adjusted HR = 1.42, p = 0.001), while severe impairment was associated with a 57% higher dementia hazard (adjusted HR = 1.57, p = 0.009) compared with participants with normal vision.

Mild visual impairment was not statistically associated with dementia risk after adjustment and the six-year exclusion (p = 0.06). The association was statistically significant in the adjusted model before that exclusion, so the final result does not establish an absence of risk.

Restricted mean survival time analyses found that moderate and severe visual impairments were associated with an estimated 1.5 and 2.1 fewer dementia-free years, respectively, over 16 years, based on group-level survival curves. Joint trajectory modeling estimated a doubled dementia hazard (HR ≥ 2.0) at 0.55 LogMAR or worse, even with stable vision, with rapid visual decline further magnifying risk around 0.61 LogMAR or worse. These are study-specific modeled thresholds.

An improving visual trajectory was associated with a weaker relationship between poor visual acuity and dementia than continued deterioration. Finally, cataract and glaucoma were not shown to be independently associated with dementia after adjusting for visual acuity. Disease severity, subtype, duration, and treatment were unavailable, limiting interpretation.

Associations were stronger at ages 65–74 and no longer statistically significant at age 75 or older. Interaction tests supported an age difference but found no difference by diabetes status.

Conclusions

The present study is among the largest and longest community-based studies examining the association between visual acuity loss and increased dementia risk. Its observational design cannot establish causation or treatment benefits. Residual confounding and missed dementia cases remain concerns. The findings may not generalize to other populations because participants were older, community-dwelling Chinese adults attending Hong Kong health centers. Relatively few participants had normal or improving vision, and habitual acuity testing could not fully separate inadequate correction from eye disease.

These findings highlight that incorporating visual monitoring into primary healthcare could improve dementia risk stratification. Moving forward, clinical trials should aim to determine whether timely interventions, such as cataract surgery or corrective lenses, can preserve cognitive health in aging populations.

Sources:

Journal reference: