Longitudinal changes in loneliness in old age: Trends and predictors

Past research has been inconsistent concerning longitudinal changes in loneliness in old age and factors that correlate with loneliness in older persons. Two theories relating to loneliness in old age, the Socioemotional Selectivity Theory and the Model of Depression and Loneliness (MODEL), differ in their predictions of the course of loneliness in old age. Therefore this study sets out to examine: (1) whether loneliness increases with age, (2) characteristics of those who are lonely in the older population; and (3) predictors of becoming lonely in old age. The analysis used data from the Cross-Sectional and Longitudinal Aging Study (CALAS), which studied health, mental, and social status of people aged 75 years and older living in Israel. Analysis of the data showed an increase in loneliness with age among older persons. The strongest predictor of loneliness was marital status, where married persons were less lonely than the unmarried. Among the unmarried, poor health, worse financial resources, and a lack of social network and emotional support independently predicted loneliness. Among the married, lower cognitive vitality and more past trauma independently predicted loneliness. Among those not lonely in wave 1 of the study, those who became lonely were more likely to be female, have insufficient financial resources, and have poorer health and functional status. These findings reveal important implications for development of preventive programs, including the importance of considering demographic and health factors and previous trauma, the possibility of developing different models for married and unmarried persons, and the importance of targeting women.