Does the Internet Protect the Elderly From Cognitive Decline?

The results of a large longitudinal study spanning several years support a decrease in the risk for dementia among older adults who regularly use the internet for < 2 hours per day. 

Several cross-sectional and longitudinal studies (though with relatively short follow-up periods) suggest that regular internet use helps maintain cognitive reserve, although some observers have voiced skepticism. This hypothesis is particularly relevant for older patients facing the potentially detrimental effects of brain aging. According to some studies, memory, cognitive performance, and verbal reasoning tend to be better preserved among internet users.

Several factors come into play, including socioeconomic disparities, socio-educational level, and generational differences, since internet usage varies qualitatively and quantitatively with age. Older patients theoretically have more limited internet usage. Under these conditions, the effect on cognitive functions would likely be modest compared with generations who were immersed in digital technology early on and tend to overuse it. After a certain age, accelerated brain aging would weigh much more heavily than any potential positive effects of the internet. It is worth noting that the negative effects of internet use have mainly been studied in young subjects, thus there is a lack of data concerning older patients.

Nearly 20,000 Participants

These considerations highlight the significance of a longitudinal cohort study that included 18,154 adults aged 50-64.9 years who were free from any dementia at baseline. These adults were participating in the Health and Retirement Study. The median follow-up period was 7.9 years, and follow-up extended to 17.1 years in some cases. Given that adults with better cognitive health are likely to self-select as regular users, the propensity score method was employed to control for this nonrandom factor using inverse probability weighting.

The risk for dementia based on initial internet use was estimated using the Cox proportional hazards model, incorporating potentially late entry into the workforce and several covariables. Interactions with education level, gender, generation, and ethnic origin were also considered. Cumulative internet exposure in terms of regular periodic use throughout life was included in the statistical analysis, as well as the hours spent on this activity each day. The analyses were conducted from September 2021 to November 2022.

Risk Nearly Halved

Regular internet use was associated with a reduced risk for dementia, compared with irregular use. The hazard ratio (HR) for dementia was estimated at 0.57. After adjustment for the nonrandom factor of self-selection, this association persisted, and the HR decreased to 0.54. Accounting for baseline cognitive decline did not substantially change these results and yielded an HR of 0.62. The difference in risk between regular and irregular users was not altered by considering potential confounding factors such as education level, ethnic origin, gender, or generation. The longer the cumulative exposure over life, the lower the risk for dementia during follow-up.

The relationship between dementia risk and daily internet usage hours seems to follow a U-shaped curve, with the lowest risk observed for durations between 0.1 and 2 hours. However, these estimates did not reach statistical significance because of the small sample size analyzed.

The risk for dementia appears to be approximately twice as low among regular internet users compared with nonusers. This hypothesis deserves serious consideration because of the large sample size and long follow-up duration, as well as careful consideration of as many potential confounding factors as possible. Potential negative effects remain to be clarified as the study was not designed to detect them. The results of previous studies suggest that internet usage should be moderate for optimal benefit, with approximately 2 hours per day being the most suitable duration, regardless of age, until proven otherwise.

This story was translated from JIM, which is part of the Medscape professional network, using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.