Numerous studies have reported elevated psychological distress during the coronavirus disease 2019 (COVID-19) pandemic in Germany (Bäuerle et al., Reference Bäuerle, Teufel, Musche, Weismüller, Kohler, Hetkamp and Skoda2020; Benke, Autenrieth, Asselmann, & Pané-Farré, Reference Benke, Autenrieth, Asselmann and Pané-Farré2020; Jungmann & Witthöft, Reference Jungmann and Witthöft2020) and other countries worldwide (Prati & Mancini, Reference Prati and Mancini2021). However, longitudinal or repeated cross-sectional studies investigating the role of lockdown restrictions on mental health are rare although such studies promise to be highly useful to identify persons at risk to develop increased anxiety and depression (Daly, Sutin, & Robinson, Reference Daly, Sutin and Robinson2020; Debowska, Horeczy, Boduszek, & Dolinski, Reference Debowska, Horeczy, Boduszek and Dolinski2020). This study aimed at examining whether symptoms of depression and anxiety were increased during v. before the first COVID-19 lockdown in Germany, whether symptoms remained elevated after lockdown restrictions were eased, and whether the observed effects were predicted by sociodemographic variables.
Three cross-sectional online surveys in Germany (Fig. 1) assessed anxiety (GAD-2; Spitzer, Kroenke, Williams, & Löwe, Reference Spitzer, Kroenke, Williams and Löwe2006), depression (PHQ-2; Kroenke, Spitzer, & Williams, Reference Kroenke, Spitzer and Williams2001), and sociodemographic factors during the early stage of the first COVID-19 wave, i.e. prior to the lockdown (10th March – 24th of March 2020; N = 11 220), at mid-stage, i.e. during peak of daily infections and lockdown-related restrictions (25th March – 13th April 2020, N = 4268), and during late stage, i.e. ease of lockdown and decline of daily infection rates (17th April–15th May 2020, N = 4335). All three samples were recruited via convenience sampling methods and differed regarding sex (early stage: 72.2% females; mid-stage: 79.2% females; late stage: 75.8% females; χ2(1) = 83.94, p < 0.001), age (early stage: 39.2% 16–34 years, 42.0% 35–54 years, 18.8% 55 and older; mid-stage: 65.4% 16–34 years, 26.9% 35–54 years, 7.7% 55 and older; late stage: 49.1% 16–34 years, 35.6% 35–54 years, 15.3% 55 and older; χ2(2) = 1095.38, p < 0.001), employment status (early stage: 74.4% employed, 9.9% unemployed, 13.0% students/in training, 1.6% retired; mid-stage: 52.6% employed, 5.6% unemployed, 34.9% students/in training, 3.1% retired; late stage: 69.6% employed, 6.6% unemployed, 17.3% students/in training, 6.2% retired; χ2(3) = 1342.63, p < 0.001), children (early stage: 28.3% having children; mid-stage: 29.5% having children; late stage: 49.1% having children; χ2(1) = 641.60, p < 0.001), and history of mental disorders (early stage: 12.9% with mental disorder; mid-stage: 13.4% with mental disorder; late stage: 38.0% with mental disorder; χ2(1) = 1410.18, p < 0.001). The samples did not differ regarding level of education (early stage: 4.5% low, 50.4% medium, 44.5% high; mid-stage: 4.5 low, 51.4% medium, 43.7% high; late stage: 4.2% low, 51.5% medium, 44.2% high; χ2(2) = 1.86, p < 0.762).
Stage (contrast coded: early v. mid- and late stage; mid- v. late stage) and sociodemographic variables served as predictors in the first step of hierarchical multiple regressions for anxiety and depression, respectively. In a second step, interactions of the contrast-coded variables of stage with sociodemographic variables were entered into the regression.
Overall and in line with our previous findings (Bäuerle et al., Reference Bäuerle, Teufel, Musche, Weismüller, Kohler, Hetkamp and Skoda2020; Benke et al., Reference Benke, Autenrieth, Asselmann and Pané-Farré2020; Jungmann & Witthöft, Reference Jungmann and Witthöft2020), women v. men, younger v. older individuals, individuals with lower v. higher educational level, individuals who were unemployed v. employed, and with v. without a previous history of mental disorders experienced higher anxiety and depressive symptoms during the first COVID-19 wave (see Table 1). Moreover, living without v. with children was associated with more depression but not anxiety.
Anxiety and depression during v. before lockdown:
Anxiety and depression were significantly higher during the lockdown than before. However, these symptom differences were greater in younger v. older individuals, in less v. more educated individuals and in individuals without v. with a previous history of mental disorders (see Table 1). Moreover, the symptom difference in anxiety (with higher anxiety during the lockdown v. before) was greater in men v. women and unemployed v. employed individuals.
a Reference category of dummy-coded variable.
b Contrast-coded stage of the first wave of COVID-19: early (−2), mid- (1), late (1).
c Early (0), mid- (1), late (−1).
Anxiety and depression during peak of v. easing of lockdown restrictions
Overall, depression and anxiety decreased during easing compared to peak of lockdown restrictions (see Table 1). However, anxiety was increased during the easing v. peak of restrictions in younger v. older individuals, in men v. women, in students/individuals in training v. employed individuals and in individuals without v. with children (see Table 1). Depression during easing v. peak of lockdown measures was higher in men v. women. The symptom difference in depression (with lower symptoms during easing v. peak of lockdown measures) was greater in individuals with v. without a history of mental disorders.
The results suggest that implementation of lockdown restrictions is associated with increased levels of depression and anxiety that even persist during ease of lockdown in specific groups. Men and individuals without a history of mental disorders were particularly affected by lockdown restrictions. The results are consistent with previous longitudinal studies from the UK, which demonstrated that younger individuals experienced higher levels of distress due to lockdown measures (Daly et al., Reference Daly, Sutin and Robinson2020; Pierce et al., Reference Pierce, Hope, Ford, Hatch, Hotopf, John and Abel2020). This study identifies vulnerable groups (Bäuerle et al., Reference Bäuerle, Teufel, Musche, Weismüller, Kohler, Hetkamp and Skoda2020; Benke et al., Reference Benke, Autenrieth, Asselmann and Pané-Farré2020; Jungmann & Witthöft, Reference Jungmann and Witthöft2020), which might need tailored support to avoid exacerbation or the development of manifest psychological disorders. Longitudinal studies based on representative community samples are needed to replicate our findings.
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Conflict of interest
The authors assert that all procedures contributing to this study comply with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2008.