Social Relationships, Living Arrangements and Loneliness.

Background Oldest-old people are expected to be particularly likely to experience loneliness due to the loss of their intimate partner or of same-aged social network members. It is assumed that individuals in different living arrangements maintain different kinds of social networks because they adjust their networks to their specific needs. However, not much is known about the variation in the social networks of the oldest-old depending on their living arrangements and how this variation is related to loneliness. This is the first study that seeks to fill this research gap by examining how the composition and the size of a social network varies among the oldest-old depending on their living arrangements with a partner (coresidential partnership, living apart together (LAT) partnership, no partnership), and how this variation contributes to explain loneliness among the oldest-old. Methods We used cross-sectional data from the representative survey NRW80+ (Quality of Life and Well-Being of the Very Old in North-Rhine Westphalia). The sample of analysis used in this study consists of 1860 respondents from the German state of North-Rhine Westphalia aged 80 years and older. Associations between social network characteristics and living arrangements were tested using χ 2 -tests and one-way ANOVA. Ordered logit models were used to explain loneliness. Results Respondents in a coresidential partnership maintained larger social networks than those in an LAT partnership and those with no intimate partner. Furthermore, the respondents with no partner maintained more diverse social networks. Compared to those in the other living arrangements, the respondents in an LAT partnership maintained the smallest and least diverse social networks. Being in a coresidential partnership and the social network size were found to be negatively associated with loneliness. Conclusion First, the results indicate that respondents who do not have a partner adjusted their social networks to meet their needs in the absence of this relationship. Second, we conclude that being in a coresidential partnership and having a large social network protects the oldest-old against loneliness.