to compare the relationship between personality traits and the loneliness rate of change between the UK and the US.

We used data from 7932 older adults 52 years and older from the English Longitudinal Study of Ageing (ELSA) and 7,979 older adults 50 years and older from the Health and Retirement Study (HRS). In ELSA, we considered wave 5 (2010/2011) as our baseline and wave 6 (2012/2013) to wave 9 (2018/2019) for loneliness follow-up. In HRS, we used wave 10 (2010) as baseline and wave 11 (2012) to wave 14 (2018) as follow-up. Loneliness was measured using the three-item R-UCLA, and personality traits were measured using the 25 items from MIDUS 1 and 2 Studies. We used the rate of change to analyse the changes in loneliness over time and multilevel mixed-effects linear regression to analyse the relationship between personality traits and loneliness rate of change. We adjusted the models by social isolation, sociodemographic, economic and health outcomes.

ELSA and HRS participants were similar except for their age (67 years in ELSA and 73 in HRS) and ethnicity (98% white in ELSA and 79% white in HRS). Over eight years, we observed a decrease in loneliness in the participants of both cohorts. The overall rate of decline was -3.93 in ELSA and -2.38 in HRS. Among the ELSA participants, in the fully adjusted models, extroversion (ß= 0.012, 95% CI: 0.004-0.021) and neuroticism (ß= - 0.010, 95% CI: -0.018-0.002) were the only personality traits associated with the loneliness rate of decline, while among the HRS participants, extroversion (ß=0.084, 95% CI: 0.070- 0.098), neuroticism (ß=-0.095, 95% CI: -0.109-0.082), agreeableness (ß=0.055, 95% CI: 0.039-0.071), conscientiousness (ß=0.045, 95% CI:0.029-0.061) and openness to experience (ß=0.031, 95% CI: 0.019-0.044) were associated with the loneliness rate of decline in the fully adjusted models.

There are important country differences in the relationship between personality traits and loneliness rate of decline. We hypothesised that in the absence of a social protection system and universal health care, the role of psychological factors might become even more relevant to predict loneliness. Cultural factors might also be playing a role.