This study aimed to examine loneliness among two birth cohorts, born 20 years apart, when they were 70 years of age, and to identify factors explaining loneliness.

The cohorts consisted of older home-dwelling residents of Turku, Finland, from the birth cohort 1920 in 1991 (N = 1530) and the birth cohort 1940 in 2011 (N = 1307). Suffering from loneliness was assessed with the question: ‘Do you suffer from loneliness?’ Cross-tabulations with chi-square test, general linear model (GLM) and multiple regression analysis were used in statistical testing and modeling.

In the 1940 cohort, around one-fifth (18%) of the respondents suffered from loneliness at least sometimes, while the corresponding figure in the 1920 cohort was around one-fourth (26%). Our analyses indicated that the effect of cohort was not a statistically significant explanatory factor of loneliness. Living status, self-rated health and memory compared to age peers were statistically significant explanatory factors for suffering from loneliness. When we controlled the effect of depressiveness on the experience of loneliness, it was shown that the effects of living status and self-rated health remained statistically significant, whereas memory compared to age peers did not. Depressiveness itself was highly important. The combined effect of living status and self-rated health emerged as the most significant explanatory factor for loneliness. Older people with poor self-rated health who lived alone were most likely to suffer from loneliness.

The findings give healthcare professionals an opportunity to plan for interventions aimed at combating loneliness among home-dwelling older people.