Exploring social network typologies and their impact on health and mental well-being in older adults: Evidence from JAGES.

The types of social networks, their prevalence, and their relationship to health outcomes in older age have been different across countries and cultures. Most of the literature has focused on USA or in European countries and little is known about the social network typologies among older adults from Japan. This study aimed to identify these patterns of social network typologies and examine the differences in sociodemographic and related to health variables. 23894 participants from the JAGES project (2019), aged 65 or older (M = 74.74, DT = 6.39) from Japan. Statistical analyses included Latent Profile Analysis (LPA) followed by ANOVAs, Chi square and multinomial logistic regressions tests to compare the profiles. Four profiles were identified: family (66.9%), spouse (16.6%), diverse (14.5%), and neighbor/others (1.9%). The profiles differ statistically (p < 0.001) in all sociodemographic characteristics and in the means of depression, loneliness, self-perceived health, and happiness. Compared with the "family" network, younger men, with fewer chronic illnesses but higher levels of depression and loneliness were more likely to be in the "spouse" profile, older women with lower socioeconomic status, but less lonely and happier in the "diverse" profile and adults who still working, have lower socioeconomic status and are less happy into the "neighbors/others" group. We discuss the differences between the profiles found, the potential differences with previous studies and the specific cultural Japanese nuances that may explain the characteristics of the network types founded. • Familial, spouse, diverse and neighbor/others social networks profiles were identified. • Differ statistically in depression, loneliness, self-perceived health, and happiness. • The diverse group presents the best health outcomes, followed by "family". • Spouse group has protective sociodemographic factors but poor health outcomes. [ABSTRACT FROM AUTHOR]