Physical attractiveness has been linked to better economic, dyadic, and health outcomes but is understudied. We focus here on the gendered implications of attractiveness for one component of social well-being, access to intimate partnership and sexuality, among older adults. And we examine the role of body size as measured and as rated by an observer in evaluations of attractiveness and the diverging consequences for women and men. We use data from Rounds 1 (2005/06) and 2 (2010/11) of the National Social Life, Health, and Aging Project (N=2144) to explore the association of two measures of body size, weight relative to height (BMI), and interviewer assessments of body size, with sexual behavior that requires a partner and with sexual behavior that does not. We find that at larger body sizes as reflected in both BMI and rated body shape, women, but not men, face a lower probability of having a partner and engaging in partnered sex, and a lower frequency of vaginal intercourse and receiving sexual touch. These associations are mediated by physical functioning for BMI and by attractiveness as rated by the interviewer for rated body shape. We also find that women, but not men, are more likely to report finding sex not pleasurable at a higher BMI, which partly operates through the mechanism of functional limitations. We suggest that these findings reflect different attractiveness standards for men and women, which reduce women's access to partners and partnered sex but not solitary sex, such as masturbation.

Despite extensive research demonstrating how marriage affects health, less is known about how health changes impact marriages, particularly when one spouse's health declines. This study extends the stress process model to dyadic contexts to ask whether and how one spouse's cognitive decline is associated with the other spouse's perceived marital strain, with attention to potential differences by gender. It also examines the influence of social support from family and friends, as well as social engagement and community participation on the dyadic dynamic. We estimated Actor-Partner Interdependence models with data (N = 620 couples) from Waves 2 and 3 of the National Social Life, Health, and Aging Project. Our results highlight gender asymmetries in dyadic stress processes. Wives reported higher marital strain when their husbands had dementia, while husbands reported lower marital strain when their wives experienced mild cognitive impairment and dementia. Friend support is associated with lower marital strain for wives whose husbands experience cognitive impairment, while socialization is related to higher marital strain for men whose wives experience cognitive impairment.

Life course theory describes each person's trajectory as intertwined with others, with the family serving as a key conduit through which resources and experiences, including those in childhood, are exchanged. We apply the framework of “linked long arms” that extends the linked lives perspective within intimate dyads, allowing for the early-life experiences and resources of one partner to affect both their own experiences and resources and those of their partner later in life. Using dyadic models with couple-level data from the National Social Life, Health, and Aging Project (N=1,214 couples), we estimate how childhood disadvantage is associated with one's own and one's partner's later-life social connectedness. Results show that people carry the influence of childhood socioeconomic and family disadvantages, but not that of violence, into their family social lives many decades later. Older women who experienced disadvantage in childhood, particularly low parental education and family unhappiness, report having fewer friends and fewer confidants, less frequent socializing, and lower participation in community organizations than others. Husbands of women whose families were unhappy report fewer friends than others, but we see no other spillovers from the wife's early disadvantage to the husband's later social life. In contrast, the wives of men who experienced disadvantage in childhood, especially lower parental education, report fewer friends and less frequent formal community participation. Older adults who grew up in families that were socially, economically, or emotionally disadvantaged show diminished late-life social well-being, and often so do their spouses, with women particularly sensitive.

Siblings are the most common household companions in childhood, yet demographers know little about how sibship configurations reverberate across the life course to shape population-level cognitive health at older ages. The author assesses whether coresiding siblings in childhood are linked to late-life cognitive trajectories and asks how associations differ by sibship size, age spacing, and sex composition. This study uses 12 biennial waves of the Health and Retirement Study (1998-2020; 42,530 person-wave observations) linked to respondents' records in the 1940 full-count US census (N = 6,187). The author estimates late-life cognitive decline trajectories by childhood coresiding sibship structure, adjusting for childhood socioeconomic, demographic, geographic, and household attributes alongside time-varying adulthood health covariates. Compared with only children, individuals who coreside with more than one sibling experience faster annual declines in cognition, net of covariates. Decline steepens monotonically with each additional sibling and is faster when at least one sibling is closely spaced. Men raised with only brothers display the lowest baseline scores, whereas women, especially those with brothers, show the most rapid deterioration. Findings are robust to birth order controls and alternative model specifications. Resource dilution in large or closely spaced sibships erodes cognitive reserve over the life course, consistent with gendered sibling caregiving norms that magnify women's later-life risks. By merging historical census data with longitudinal survey data, this study provides evidence that childhood sibship structures have enduring, gender-contingent associations with cognitive aging, highlighting siblings as an overlooked demographic determinant of population health.

Social contagion is a key mechanism that shapes health behaviors, but few studies have applied this approach at the regional level to examine how vaccination beliefs and rates vary and diffuse across geographic areas. Through modifying the traditional SIR model, this paper addresses this gap by applying social network theory to a new compartmental model to simulate regional contagion in COVID-19 vaccination rates in England, using panel data of new and accumulated vaccination numbers from December 2020 to June 2022. The model estimates each region's initial and changing vaccination beliefs and their mutual influence on each other. The results reveal that Southeastern regions in England had higher initial vaccination beliefs and stronger spillover effects on other regions than northwestern regions. The paper suggests that policies to increase vaccination rates should consider the heterogeneity and peer effects among regions and other factors that may affect vaccination beliefs. The paper also discusses the limitations of the network model and directions for future research.