Dementia Prevalence May Be Reduced by Improving Vision

Nearly 100,000 dementia cases in the U.S. could have been prevented by improving vision, a cross-sectional analysis suggested.

The population attributable fraction — the proportion of dementia cases that potentially would be prevented if a risk factor was eliminated — of vision impairment was 1.8% in the 2018 Health and Retirement Study, representing more than 100,000 prevalent cases, wrote Joshua Ehrlich, MD, MPH, of the University of Michigan in Ann Arbor, and co-authors in a JAMA Neurology brief report.

“We’ve known for some time that vision impairment is a risk factor for dementia,” Ehrlich told MedPage Today. “We also know that a very large fraction of vision impairment — possibly in excess of 80% — is avoidable or has simply yet to be addressed.”

“We found it surprising that vision impairment had been ignored in key models of modifiable dementia risk factors that are used to shape health policy and resource allocation,” Ehrlich said.

“We sought to illustrate that vision impairment is just as influential as a number of other long accepted modifiable dementia risk factors,” he continued. “When we include vision alongside these factors, we’re able to account for an even greater proportion of dementia cases that, in theory, could be prevented.”

Sensory loss is emerging as an important dementia risk factor. A recent prospective study showed that cataract surgery was tied to a nearly 30% lower risk of dementia in older adults. In an autopsy study, hearing impairment was linked with neuropathological hallmarks of dementia.

When the Lancet Commission updated its assessment of 12 key modifiable risk factors that could put a dent in the global dementia burden, hearing loss was high on the list, but vision loss was not included.

Many of the pathways by which hearing loss may lead to dementia may hold for vision as well, “including increased cognitive load, changes to brain structure and function, and increased social isolation and loneliness,” noted Jennifer Deal, PhD, of the Cochlear Center for Hearing and Public Health in Baltimore, and Julio Rojas, MD, PhD, of the University of California San Francisco, in an accompanying editorial.

The vision-dementia link has not been as well studied, but vision is “a target for high-yield nonpharmacologic interventions, such as cataract surgery or prescription of eyeglasses” and could have a considerable effect on dementia prevalence, they pointed out.

In their analysis, Ehrlich and co-authors used data from the 2018 round of the Health and Retirement Study that included a probability sample of U.S. adults ages 50 and older. They evaluated 16,690 participants whose weighted demographic characteristics were 54% female, 52% age 65 or older, 10.6% Black, and 80% white.

The model included 12 dementia risk factors. The researchers calculated the population attributable fraction based on the prevalence and relative risk of dementia for each risk factor.

Hypertension had the highest weighted population attributable fraction for dementia (12.4%), followed by obesity (9.2%), depression (9.1%), hearing loss (7.0%), traumatic brain injury (6.1%), diabetes (5.1%), smoking (3.2%), physical inactivity (3.1%), less education (3.1%), social isolation (1.9%), vision loss (1.8%), and excessive alcohol consumption (0.3%).

“These results suggest that vision impairment should be included in the Lancet Commission’s life-course model of potentially modifiable dementia risk factors,” Ehrlich and colleagues wrote.

The findings come with caveats, the editorialists noted.

“Although a critical measure for public health planning, the population attributable fraction requires many strong assumptions, including that the vision loss-dementia association has been estimated without error (there is always error in observational studies), and that vision loss is a cause of dementia,” Deal and Rojas wrote.

“At this time, although there is observational research to support an association, it is not yet known whether vision loss causes dementia,” they added.