The Relation of Social Network Types to Physical Health and Mental Well-Being in Older Asian Americans

Background and Purpose:

Previous research has identified several types of social networks common across cultures, including “diverse” (maintaining diverse social ties), “family” (focusing on family relationships), “friends” (focusing on relationships with friends), and “restricted” (lacking social ties generally). However, few studies have assessed network types in older Asian Americans and few network studies have included more subjective criterion variables such as perception about the strength of family relationship. The purpose of this study was to: (1) develop an empirical typology of the social networks in older Asian Americans using both structural and subjective criterion variables; and (2) examine the relation of network types on physical health and mental well-being.  We hypothesized that there would be both common and unique network types in older Asian Americans and that these types would differ in their implications for health and well-being.

Methods:

The participants included 533 older Asian Americans (Chinese, Asian Indian, Korean, Vietnamese, and other Asians) who participated in the 2015 Asian American Quality of Life Survey in Central Texas (M_age = 69.4, SD = 6.88).  The 10-page survey was developed in English and translated into six Asian languages, and participants completed the questionnaire in English or in their native language.  Latent profile analysis (LPA) was conducted using seven social network-related variables: marital status, size of family network, size of friend network, family solidarity, congregational attendance, importance of religion, and community cohesion.  The identified typologies were then regressed on the indicators of health and well-being (poor rating of health, probable mental distress, and dissatisfaction with life). 

Results:

The LPA identified the model with four network types as being most optimal (BIC = 14693.76, Entropy = .86, LMR-LRT = p < .01).  The groups were named “diverse” (enriched networks with strong endorsement with diverse relationships), “moderately diverse” (networks with moderate endorsement with various social ties), “marginally restricted” (marginally engaged with family, friends, and congregation), and “restricted/congregational” (structurally disengaged and having limited relationships with family and friends, but congregational).  The results from logistic regression models show that compared to the “diverse” group, the “moderately diverse” and “marginally restricted” groups had greater odds of falling into the probable mental distress category, and the “marginally restricted” and “restricted/congregational” groups had greater odds of not being satisfied with life.

Conclusions and Implications:

Unlike previous network studies, there were no distinct family or friends network types among these older Asian Americans.  Rather network types were shaped in the continuum of different social ties and subjective evaluation on strength of social ties. Certain ethnic groups were over-represented in specific network types.  For example, more than half of Chinese belonged to the marginally restricted type.  Such findings suggest the importance of understanding ethnic group variations in network vulnerabilities when considering interventions.