Study: Chronic loneliness increases stroke risk in middle-aged, older adults

Key takeaways:

• Participants who were considered lonely at baseline had a 25% increased risk for stroke.
• This risk increased to 56% among those who consistently had loneliness over at least 4 years.

Loneliness was associated with an increased risk for stroke in middle-aged and older adults, and the risk was particularly high among those with chronic loneliness, a study in eClinicalMedicine showed.

“Loneliness has been identified as a modifiable risk factor for stroke, which is a leading cause of death and disability in the United States,” Yenee Soh, ScD, SM, a research associate at Harvard T.H. Chan School of Public Health, told Healio. “Most prior studies examine loneliness at a single time point, yet loneliness can be transient or chronic. To assess whether loneliness interventions may benefit stroke prevention, it is important to empirically assess how changes in loneliness may be associated with stroke risk.”

In the study, Soh and colleagues looked at 2006 to 2018 Health and Retirement Study (HRS) data to assess the link between stroke and loneliness at a singular point and multiple points in time.

From 2006 to 2008, which served as the study’s baseline period, 12,161 participants aged 50 years or older without a history of stroke answered questions on the Revised UCLA Loneliness Scale.

From 2010 to 2012, which served as the follow-up period, 8,936 participants who remained in the HRS answered the questions again.

The researchers categorized those participants into one of four groups based on their loneliness scores at the baseline and follow-up loneliness assessment periods:

• consistently low loneliness, composed of those who scored low at baseline and follow-up;
• remitting loneliness, composed of those who scored high at baseline and low at follow-up;
• recent-onset loneliness, composed of those who scored low at baseline and high at follow-up; and
• consistently high loneliness, composed of those who scored high at baseline and follow-up.

Overall, 1,237 incidences of stroke occurred from 2006 to 2018 among participants who answered questions at baseline, and 601 occurred among those who answered at both baseline and follow up.

Soh and colleagues’ baseline analysis showed that a one-unit increase in loneliness score was tied to a 5% increased risk for stroke (HR = 1.05; 95% CI, 1.01-1.08).

Individuals classified as lonely at baseline had a 25% higher risk for stroke vs. those who were not (adjusted HR = 1.25; 95% CI, 1.06-1.47).

When the researchers looked at changes in loneliness over the two time points, “those who had remitting and recent onset loneliness did not have a significantly higher risk than those who consistently reported not feeling lonely,” Soh noted.

Participants who were categorized as having consistently high loneliness had a 56% increased risk for stroke compared with those who reported consistently low loneliness (HR = 1.56; 95% CI, 1.11-2.18).

Soh pointed out that loneliness is a public health issue and has “significant implications for physical health.”

“It is important to routinely assess loneliness, as the consequences may be worse if ignored,” she said. “When considering interventions for loneliness, it is important to address the subjective feeling of loneliness and to not conflate it with social isolation (ie, lack of social contact with others).”

Future research examining loneliness patterns over shorter and longer periods of time “may help better understand how loneliness impacts stroke risk,” Soh added.

“Research is also needed to better understand the potential mechanisms underlying the loneliness-stroke association, and whether these findings apply to younger populations or non-U.S. contexts,” she said.

References:

• Chronic loneliness may increase stroke risk among older adults. https://www.hsph.harvard.edu/news/press-releases/chronic-loneliness-may-increase-stroke-risk-among-older-adults/. Published June 24, 2024. Accessed June 24, 2024.
• Soh Y, et al. EClinicalMedicine. 2024;doi:10.1016/j.eclinm.2024.102639.