Depression and loneliness of older adults in Europe and Israel after the first wave of covid-19

Abstract

Epidemic control measures that aim to introduce social distancing help to decelerate the spread of the COVID-19 pandemic. However, their consequences in terms of mental well-being might be negative, especially for older adults. While existing studies mainly focus on the time during the first lockdown, we look at the weeks afterward in order to measure the medium-term consequences of the first wave of the pandemic. Using data from the SHARE Corona Survey, we include retired respondents aged 60 and above from 25 European countries plus Israel. Combining SHARE data with macro-data from the Oxford COVID-19 Government Response Tracker allows us to include macro-indicators at the country level, namely the number of deaths per 100,000 and the number of days with stringent epidemic control measures, in addition to individual characteristics. The findings show that both macro-indicators are influential for increased feelings of sadness/depression, but that individual factors are crucial for explaining increased feelings of loneliness in the time after the first lockdown. Models with interaction terms reveal that the included macro-indicators have negative well-being consequences, particularly for the oldest survey participants. Additionally, the results reveal that especially those living alone had a higher risk for increased loneliness in the time after the first COVID-19 wave.

Supplementary information: The online version contains supplementary material available at 10.1007/s10433-021-00640-8.

Keywords: Covid-19; Cross-country comparison; Loneliness; Multilevel analysis; SHARE; Sadness/depression.

Fig. 1

Country differences in the prevalence of more sadness/depression and more loneliness, Data: Preliminary SHARE Wave 8 Release 0 (n = 27,889)

Fig. 2

Association between country differences in the prevalence of more sadness/depression and loneliness with deaths per 100,000, Data: Preliminary SHARE Wave 8 Release 0 (n = 27,889)

Fig. 3

Association between country differences in the prevalence of more sadness/depression and loneliness with days with stringency index above 60, Data: Preliminary SHARE Wave 8 Release 0 (n = 27,889)

Fig. 4

Average marginal effects of main and interaction effects for feeling sadder/more depressive and deaths per 100,000 for each age group (n = 27,889). Note. Model 1 = Main effect of deaths per 100,000, Model 2 = interaction with number of deaths and age. AMEs were calculated for the model with fixed predictors at individual and macro-level with random intercepts, Data: Preliminary SHARE Wave 8 Release 0

Fig. 5

Average marginal effects of main and interaction effects for feeling sadder/more depressiv and number of days with stringency measures above 60 for each age group (n = 27,889). Model 1 = Main effect of number of days with stringency above 60 0, Model 2 = interaction with number of days with stringency above 60 and age. AMEs were calculated for the model with fixed predictors at individual and macro-level with random intercepts, Data: Preliminary SHARE Wave 8 Release 0

Fig. 6

Average marginal effects of main and interaction effects for feeling lonelier and number of deaths for living alone (n = 27,889). Model 1 = Main effect of deaths per 100,000, Model 2 = interaction with number of deaths and living alone. AMEs were calculated for the model with fixed predictors at individual and macro-level with random intercepts, Data: Preliminary SHARE Wave 8 Release 0

Fig. 7

Average marginal effects of main and interaction effects for feeling lonelier and number of deaths for personal contacts (n = 27,889). Model 1 = Main effect of deaths per 100,000, Model 2 = interaction with number of deaths and personal contacts. AMEs were calculated for the model with fixed predictors at individual and macro-level with random intercepts, Data: Preliminary SHARE Wave 8 Release 0