Does Social Isolation in Older Adults Increase Dementia Risk?

Changes in social isolation in older adults has an impact on health outcomes such as disability, stroke, and dementia, according to study results published in JAMA Network Open.

A majority of research on social isolation and health outcomes come from cross-sectional studies, and not much is known about the direct association between social isolation changes and health outcomes. To gain better insight into the effects of social isolation, researchers estimated associations of social isolation with disability, dementia, cardiovascular disease (CVD), and stroke.

The researchers analyzed data from 6 birth cohorts of Health and Retirement Study (HRS) participants from 2006 to 2020. Leave-Behind Questionnaires (LBQ) were included in HRS’ biennial core survey from 2006-2008, which included the Steptoe 5-item Social Isolation Index (SII) to assess participant’s social isolation status. Half of HRS respondents were chosen for LBQ, resulting in an SII assessment every 4 years.

With SSI, participants were asked to respond if they:

1. Were unmarried/living alone
2. Had less than monthly contact with children (includes face to face meet up, phone calls, writing or email)
3. Had less than monthly contact with other family members
4. Had less than monthly contact with friends
5. Did not participate in groups, clubs, or social organizations

The final score of the SSI was the sum of these 5 items, ranging from 0-5.  The higher the score, the greater the social isolation.

The researchers defined the baseline assessment as the year of participants’ first SII measurement, and the second SII measurement occurring 4 years after baseline. The researchers then searched for a decrease, increase, or no change in the participant’s score.

For participants categorized as initially isolated at baseline, the researchers defined change groups as shifting from isolation to non-isolation vs remaining isolated.

Meanwhile, for participants categorized as initially non-isolated at baseline, change groups were defined as shifting from non-isolation to isolation vs remaining non-isolated. The researchers also investigated the incident outcomes of mortality, disability, dementia, CVD, and stroke from participants’ Medicare records, the Medicare Chronic Conditions Data Warehouse (CCW) and the Social Security Death Index (SSDI).

Among these participants, 9093 individuals were not socially isolated at baseline and 4556 were socially isolated. Of the baseline non-isolated participants, 1055 experienced a decrease in isolation, 4553 had no change, and 3485 experienced an increase in isolation at their second SII measurement.

The SII changed by a mean of −1.0 points for the decreased isolation, 0 points for no change, and 1.3 points for the increased isolation groups over 4 years. Among the 4556 baseline isolated participants, 2067 experienced a decrease in isolation, 1796 had no change, and 693 experienced an increase in isolation at the second SII measurement. The SII changed by a mean of −1.37 points for the decreased isolation, 0 points for no change, and 1.41 points for the increased isolation groups.

The increased isolation group tended to be older, women, married, and non-Hispanic White with lower levels of education. At baseline, they had poor physical and cognitive function and an increased prevalence of comorbid conditions (hypertension, CVD, arthritis).

Among the 13,694 participants (mean age at baseline, 65.3; women, 8011) isolated at baseline, the researchers found that those with increased isolation had higher mortality (incidence rate [IR], 68.19 per 1000 person-years; 95% CI, 60.89-76.36) than those who were stable (IR, 44.02; 95% CI, 40.47-47.88) or had decreased isolation (IR, 37.77; 95% CI, 34.73-41.09).

In addition, there was an association between increased isolation and a higher risk for mortality (adjusted hazard ratio [aHR], 1.29; 95% CI, 1.09-1.51), disability (aHR, 1.35; 95% CI, 1.09-1.67), and dementia (aHR, 1.4; 95% CI, 1.02-1.93)

A decrease in isolation was not associated with lower risk for any of the health outcomes except for mortality (aHR, 0.73; 95% CI, 0.61-0.87) in participants who were not socially isolated at baseline.

These findings indicate that increased isolation is associated with increased risk for mortality, disability, and dementia, regardless of baseline isolation status. These results reinforce the importance of social isolation interventions to minimize its effects on mortality, physical, and cognitive impairment.

Study limitations included the HRS as a cohort study, which cannot define casualty. Change in isolation was only measured in a 4-year span. The SII questionnaire was based on 5 questions, which could have been too narrow of a range. Additionally, Medicare may not fully define CVD, Stroke, and dementia diagnosis. Lastly, social isolation is difficult to scope due to personal life events, such as a death in the family, leading to increased social isolation which cannot be prevented.

“Addressing the inevitability of certain life events and incorporating this reality into intervention strategies is crucial for reducing social isolation and improving health outcomes,” the researchers concluded.