Association of Social Support With Brain Volume and Cognition

Key Points Question What is the association of different forms of social support with an early neuroanatomical marker of Alzheimer disease vulnerability and cognitive function? Findings In this cross-sectional study, high (vs low) availability of supportive listening was associated with cognitive resilience, which indicated better global cognitive function than expected for lower cerebral volume. This association was absent for other forms of social support. Meaning In psychosocial interventions and related public health strategies to promote neurocognitive health, precise targeting of specific forms of social support, such as supportive listening, may be warranted.


Image Analysis
MRI scan digital information was transferred after acquisition to the central laboratory directed by a co-author (CD) for processing and analysis. Analyses were all conducted blind to participant identifying information. Images were evaluated with semiautomatic segmentation analyses using operator-guided removal of non-brain elements by operator-guided tracing of the dura matter within the cranial vault. This included the middle cranial fossa and was above the posterior fossa and cerebellum. The cranial vault measure derived was defined as total cranial volume and was used as a head size estimate to correct for established sex differences in head size.
Total cerebral brain volume was quantified using a multi-step process starting with image segmentation to define brain matter from cerebral spinal fluid (CSF). Subtraction of the second echo image from the first echo image yielded a difference image. Following image segmentation into brain matter and CSF, the operator returned to the image to measure lobar brain volumes. To preserve measurement precision, segmented brain-CSF images were rotated separately from the original image.
The image was transformed into anatomic standard space; the operator then returned to the image and identified brain lobar and regional CSF measures. Volumes were all calculated as the sum of the pixels within the identified region of interest multiplied by the pixel volume in milliliters. Repeat analysis of intra-and inter-rater reliabilities were consistently above 0.90. 36 The measured total cerebral volume was corrected for head size using the ratio of total cerebral volume over total cranial volume, multiplied by 100 c Global cognitive score calculated by summing the products of the standardizing formulas and the component loadings for each cognitive task.

eTable 2. Sample Distribution of Five Social Support Domains, Stratified By Level
a For each type of social support, participants responded to the following items on the Berkman-Syme Social Network Index: listening, "Can you count on anyone to listen to you when you need to talk?"; advice, "Is there someone available to give you good advice about a problem?"; love-affection, "Is there someone available to you who shows you love and affection?"; emotional support, "Can you count on anyone to provide you with emotional support?"; and, sufficient contact, "Do you have as much contact as you would like with someone you feel close to, someone in whom you can trust and confide?". b Each type of social support domain was included as a predictor in separate models above and as a two-level variable, high versus low. A high level was defined as responding "most of the time or all of the time" versus "some, little, or none of the time" for the respective item: Listener, "Can you count on anyone to listen to you when you need to talk?" Advice, "Is there someone available to give you good advice about a problem?" Love-affection, "Is there someone available to you who shows you love and affection?" Emotional support, "Can you count on anyone to provide you with emotional support?" Sufficient contact, "Do you have as much contact as you would like with someone you feel close to, someone in whom you can trust and confide?"

eFigure 1. Predicted Association Between Cerebral Volume and Global Cognition By Availability of Supportive Listening: All Participants
To account for covariates, models are based on the residuals of total cerebral volume and global cognitive scores when regressed onto the primary set of covariates: age, age 2 , sex, education, and time interval from social support assessment to visit when MRI and neuropsychological testing were obtained. Blue line=High listener availability (n=1898). Red line=Low listener availability (n=279). Bands indicate 95% confidence intervals.

eFigure 1. Predicted Association Between Cerebral Volume and Global Cognition By Availability of Supportive Listening: All Participants
High listener availability Low listener availability =0.08 =0.20

Availability of Supportive Listening: Participants Age ≥65
To account for covariates, models are based on the residuals of total cerebral volume and global cognitive scores when regressed onto the primary set of covariates: age, age 2 , sex, education, and time interval from social support assessment to visit when MRI and neuropsychological testing were obtained. Blue line=High listener availability (n=786). Red line=Low listener availability (n=110). Bands indicate 95% confidence intervals. =0.17

=0.22
High listener availability Low listener availability