Specific Aims

The Specific Aims page from the funded application, as submitted.

Social Network Cognitive Buffers and AD/ADRD: Learning from COVID-19 Disruptions

F31 PI: Chris Soria, UC Berkeley

Individuals with decreased social support, lower social interaction, and higher perceived loneliness tend to exhibit worse cognitive functioning as they age, suggesting a causal link between a robust social network and Alzheimer’s Disease and Alzheimer’s Disease Related Dementias (AD/ADRD). Social Network Cognitive Buffers (SNCBs) are social network characteristics, such as contact frequency and network size, that could directly reduce the rate of cognitive aging by causing the individual to engage in the cognitive complexity of social interaction. SNCBs could also indirectly improve cognitive aging by decreasing depressive symptoms.

Prior work has made a distinction between the quantity versus the quality of SNCBs. While some studies have found cognitive associations primarily with quantitative measures such as social network size and contact frequency, other studies indicate the potential importance of quality as measured e.g. by relationship satisfaction and resulting loneliness. Further research is necessary to explore the independent effects of each and gain a better understanding of their relative importance, particularly using causal estimation approaches.

Additionally, while prior work has investigated whether social isolation and loneliness diminish cognitive functioning at older ages, little work has explored whether reducing social isolation and loneliness can have a restorative effect at older ages. It is plausible that replacing an individual’s social network deficiencies could not only reduce the rate of cognitive decline but could even improve cognitive functioning. The overall objective of this application is to assess the impact of COVID-19-induced social isolation and loneliness on cognitive aging in older adults and to pinpoint which social interaction elements most effectively mitigate cognitive decline.

The Health and Retirement Study (HRS) is ideal for this analysis due to its longitudinal design, allowing for causal interpretation. With data collected since 1992, the HRS surveyed respondents before the pandemic (2018), during (2020, plus a special COVID-19 module), and in the period emerging from the pandemic (2022, released this year). In each main wave, the HRS collects data on cognitive functioning through the Telephone Interviews for Cognitive Status (TICS) for measuring “global cognitive functioning”, alongside SNCB and related questions about family structure, household composition, marital status, loneliness and social contact, and information on engagement with their community in several dimensions.

Aim 1. To assess declines in cognitive aging because of the loss of SNCBs during COVID-19 and any potential improvement due to the restoration of SNCBs in people aged 65+.

Hypothesis 1a. Cognitive performance declined more rapidly among older adults who experienced greater loss of SNCBs during COVID-19.

Hypothesis 1b. “Restoring” SNCBs will decrease, or even reverse, the rate of cognitive decline.

Aim 2. Evaluate the disparate effects on individuals in order to approach core mechanisms.

Hypothesis 2a. Loneliness and reduced social contact will independently affect cognitive performance, with distinct impacts observed from either condition on its own and in interaction with each other.

Hypothesis 2b. Married individuals limiting their external social interactions will exhibit larger cognitive effects compared to their counterparts who maintain diverse social engagements, controlling for bereavement.

This study will employ causal inference methodologies, including fixed effects and mediation analysis techniques (with appropriate controls for confounders), studying the unique backdrop of the COVID-19 pandemic to assess the impacts of social isolation and the ensuing post-pandemic recovery effects to investigate the importance of Social Network Cognitive Buffers (SNCBs) in AD/ADRD. This study will distinguish between older adults with diverse social networks and those with limited diversity, while also differentiating feelings of loneliness from objective measures of social isolation. This approach allows for a precise assessment of their respective impacts on cognitive aging, which will in turn help inform public health policy and intervention. Overall, conducting this study, aided by this fellowship, will propel my expertise in social networks and AD/ADRD, paving the way for my prospective professorship in the social demography of aging.