Social Isolation Linked to Being Older Than Biological Age, Greater Mortality

Published on: 

In a new study, investigators evaluated if social contact influenced the age gap between biological age, estimated with artificial intelligence (AI), and chronological age.

Social isolation increases biological aging—making people biologically older than their age—and increases the risk of all-cause mortality, a new study found.1

“This study highlights the critical interplay between social isolation, health and aging,” said investigator Amir Lerman, MD, a cardiologist at Mayo Clinic, in a press release.2

Investigators aimed to evaluate the relationship between social contact and biological aging. To do so, they compared the Social Network Index with the age gap created by the chronological age and the biological age. An AI-enabled electrocardiogram (AI-ECG) determined the predicted biological age.

The study included 280,000 adults aged ≥ 18 years who received outpatient care between June 2019 and March 2022. Most participants were non-Hispanic white (86.3%), half were female (50.9%), and the mean chronological age was 59.8 ± 16.4 years. Participants had AI-ECG records filed within a year of completing a social determinants of health questionnaire and had AI-ECG records filed. The filing of the AI-ECG was unrelated to the study.

Mayo Clinic developed its own AI-ECG model, using 12-lead ECG from 7,647,830 adult subjects aged ≥ 18 years. The team used the AI-ECG model to compare estimated biological age with chronological age. Previous studies found the AI-ECG predicted age represents the heart’s biological age—a positive age gap signals increased biological aging and a negative age gap signals slower biological aging.

The Social Network Index evaluated social isolation, presenting people with 6 multiple-choice questions regarding belonging to any social club or organization, how often they participate in social activities per year, how often they talk on the phone with family and friends per week, how often they attend religious services per year, how often they get together with friends or family in person per week, and marital status or if they live with a partner. Each question received a score of 0 or 1. A score of 0 – 4 represented various degrees of social isolation.

When looking at medical history, the team found 33% of participants had hypertension, 18.8% had hyperlipidemia, and 11.8% had diabetes. A low Social Network Index score was linked to a greater prevalence of comorbidities.

The mean age gap between chronological and biological age was −0.2 ± 9.16 years. An univariable linear regression model showed a significant link between social isolation and age gap (-1.9; 95% CI, -20 to -1.7; P < .001). After adjusting for demographic and comorbidities, the team observed participants with a greater Social Network Index score—and thus better social contact—had a smaller difference between chronological and AI-ECG determined age (−0.11; 95% confidence interval [CI], −0.22 to −0.01; P < .001). This was the case for all genders and age groups.

The team noticed racial health disparities; non-white participants had on average a greater difference between chronological and biological age than white participants.

Not only is social isolation linked to increased biological aging, but it is also linked to a significant all-cause mortality risk (hazard ratio [HR], 0.47; 95% CI, 0.43 to 0.5; P < .001). Participants who died were significantly older (70 ± 14.2 years vs 59.3 ± 16.3 years), and common comorbidities were hypertension (48.6%), hyperlipidemia (33.9%), and chronic kidney disease (24.8%).

During the 2-year follow-up, 4.9% of the participants died, and 42% were women. Participants were more likely to die if they had a low social index score of ≤ 1. Likewise, a greater Social Network Index was linked with a lower mortality risk (P < .001).

Investigators said the results were limited by not including an equal amount of all racial groups and not having a complete representation of the general population, areas of improvement for the AI-ECG algorithm, not adjusting the P values for comparisons, potential unmeasured confounders, and possible selection bias due to the fact the sample was selected from medical service visits.

“Social isolation combined with demographic and medical conditions appears to be a significant risk factor for accelerated aging,” Lerman said.2 “But we also know that people can change their behavior — have more social interaction, exercise regularly, eat a healthy diet, stop smoking, get adequate sleep, etc. Making and sustaining these changes may go a long way toward improving overall health."


  1. Rajai, N, Medina-Inojosa, J, Lewis, B. et al. Association Between Social Isolation With Age-Gap Determined by Artificial Intelligence-Enabled Electrocardiography. JACC Adv. null2024, 0 (0) .
  2. Social Isolation Linked to Biological Age Gap, Higher Mortality Rate. EurekAlert! March 20, 2024. Accessed March 26, 2024.