Abstract:

The high prevalence of loneliness and various accompanying adverse consequences (e.g., depression, higher blood pressure, insomnia, immune stress responses and worse cognition) are becoming serious public concerns. To provide insights on prevention and intervention programs, this study examines the properties and development of loneliness behaviors in elderly people. Taking a longitudinal perspective enables researchers to understand who is expected to transition to a higher risk status in the future, which will help to predict symptoms so that tailored interventions can be designed to protect the elderly from loneliness.
Two-wave longitudinal data over 4 years were derived from the American Health and Retirement Study (HRS). A total of 3238 women and 2205 men in their 50s were recruited. Latent class and latent transition models were used to identify meaningful subgroups of the elderly with different symptoms (i.e., lack of companionship, being left out, isolation from others, lack of belonging, and withdrawn from society) to describe transitions between those classes over the study period and to examine the effects of covariates on the latent transition model. Six covariates, including gender, marital status, attitudes toward aging, life orientation, social support and daily activities, were examined as potential predictors of loneliness.
Four loneliness subgroups were identified: Mild Loneliness (Class 1), Social Loneliness (Class 2), Emotional Loneliness (Class 3), and Severe Loneliness (Class 4). Mild Loneliness status was the most prevalent, barely showing any lonely behavior, while relatively serious loneliness occurred in Social Loneliness, Emotional Loneliness and Severe Loneliness. Specifically, individuals in Social Loneliness lacked social intercourse relation, individuals in Emotional Loneliness lacked a close relationship, and individuals in Severe Loneliness had a high probability of exhibiting all the above mentioned symptoms. Elderly people in Mild Loneliness and Severe Loneliness were highly stable, while Social Loneliness and Emotional Loneliness tended to change to other statuses rather than remaining in the original status. Particularly, the participants in Social Loneliness demonstrated a prominent trend to transition to a less problematic status (Social Loneliness to Mild Loneliness), and the participants in Emotional Loneliness tended to change to a more problematic status (Emotional Loneliness to Severe Loneliness). All participants in the four subgroups showed a strong willingness to communicate with others. Results of multinomial logistic regression revealed that elderly males were more likely to be in the Emotional Loneliness group. In addition, more social support as well as a positive attitude toward aging and optimistic life orientation were more likely to keep elderly people from entering more severe loneliness statuses, indicating an ameliorated trend of loneliness that was expected.
This study demonstrated a transition pattern in elderly people loneliness with an individual-centered approach. Differential treatment effects were found across baseline loneliness classes, suggesting the benefits of tailoring intervention programs to yield good outcomes in elderly people.

Key words: elderly people, loneliness, latent transition analysis