Time trends in emotional well-being and self-esteem in children and adolescents during the COVID-19 pandemic

The COVID-19 pandemic has presented unprecedented challenges to the mental health of children and adolescents. Recent studies have revealed that mental illnesses among children and adolescents increased during the pandemic [1‐3]. For instance, Newlove-Delgado and colleagues showed that in July 2020, 5- to 16-year-olds in England experienced increased mental health problems compared to before the pandemic [2]. While such studies provide valuable evidence amidst an unprecedented crisis, most studies were conducted during the initial stages of the pandemic (i.e., up to early 2021) when many countries were under repeated strict social distances measures and when the information on the new virus was limited. It is likely that the mental health of children and adolescents changed with time.

In adults, studies on time trends in mental health during the pandemic have generally found that longer into the pandemic, the better their mental health, representing the psychological adaptation to the COVID-19 crisis [4‐6]. In addition, the stringency of social distancing policies has been reported to show a strong negative association with mental well-being [7, 8]. However, this may not be generalizable to children and adolescents. Whereas many adults continued to work during lockdowns, albeit remotely, many youth experienced school closures, during which their lives changed completely (e.g., having no means of socializing with peers, being unable to participate in extracurricular activities, and spending much less time outside) [1]. In Japan, though national-level school closures were from April to May 2020, this was preceded by spring break in March 2020 and followed by summer vacation from July to August 2020, meaning that youth spent very little time in school during the first several months of the pandemic. Even as the school closures ended, many children and adolescents were asked to refrain from various activities that involved gathering and socializing due to social distancing policies in schools and restrictions in extracurricular activities. For instance, Japan’s Ministry of Education, Culture, Sports, Science and Technology has recommended limiting extracurricular activities to a minimum when the infection rates are high [9]. As extracurricular programs have been associated with improved mental well-being according to a previous study in pre-pandemic era [10], limiting school activities under social distancing policy may have been detrimental to the mental health of youth. Despite this, evidence on the associations between stringency of social distancing policy and mental well-being in children and adolescents is scarce.

Given their unique pandemic experience, the evaluation of the mental health of children and adolescents beyond the initial stages of the pandemic and a better understanding of its association with the stringency of the social distancing policies in place will have significant policy and clinical implications. Therefore, the present study aimed to describe long-term trends in emotional well-being and self-esteem among children and adolescents in Japan during the COVID-19 pandemic and its association with stringency of the social distancing policies in place. We also aimed to examine whether trends in these mental health outcomes differed by individual characteristics such as gender and age, in order to identify groups who are at highest risk of impaired mental health amidst the pandemic.

The number of respondents aged 6–17 years old were 2430, 981, 2111, 501, and 487 in the first, second, third, fifth and seventh waves, respectively (Additional file 1: Figure S1). Table 1 reports the distributions of child age and gender for each wave. The 6- to 11-year-olds had a higher participation rate in waves 2 and 3 than in waves 1, 5, and 7. More females participated than males in all surveys ranging from 60.4% for wave 1–50.7% for wave 7. Other sociodemographic characteristics are shown in Additional file 1: Table S1.

Table 2 shows the weighted adjusted trends in emotional well-being and self-esteem over time among youth aged 6–17 years. Participants in May 2020 had the lowest self-esteem (adjusted mean, 52.5; 95% CI 51.3–53.7) compared to participants of other waves. Emotional well-being transiently increased from May 2020 to September/October 2020, but decreased into 2021; participants in December 2021 had the lowest emotional well-being (adjusted mean, 70.3; 95% CI 68.3–72.2). The results remained consistent after removing participants from three municipalities in the third wave and after removing individuals who participated in multiple surveys (Additional file 1: Tables S2 and S3). The increases in the weighted adjusted estimates of emotional well-being and self-esteem came after the school closures early in the pandemic in 2020. Although there were no school closures after this period, there were multiple declarations of emergencies over the course of the study period, in response to the several spikes in COVID-19 cases (Fig. 1).

Participants aged 12–17 years had lower emotional well-being and self-esteem compared to 6- to 11-year-olds in all waves (Fig. 2A and B, Additional file 1: Table S4). Females had lower emotional well-being than males in May 2020 (70.5 for females vs 72.7 for males; p = 0.006; Fig. 2C, Additional file 1: Table S5) and lower self-esteem than males in May 2020 (50.9 for females vs 54.0 for males; p = 0.003; Fig. 2D, Additional file 1: Table S5) and September/October 2020 (51.6 for females vs 54.9 for males; p = 0.001; Fig. 2D, Additional file 1: Table S5).

In general, there was a negative association with stringency of social distancing policies for both emotional well-being (coefficient, −1.30; 95% CI −2.55 to −0.05; Fig. 3A, Additional file 1: Table S6) and self-esteem (coefficient, −1.29; 95% CI −3.32 to 0.74; Fig. 3B, Additional file 1: Table S6) in absolute scales for a 5-unit increase in the Stringency Index, though the estimate for self-esteem was imprecise with a large CI. As the observed Stringency Indices ranged from 25.9 to 50.9, the confidence intervals of some of the estimates are large and the estimates of the outcomes for Stringency Indices outside the observed range are pure predictions. We did not find evidence that the association between stringency of social distancing policies and emotional well-being differed by age group (Fig. 3C, Additional file 1: Table S6), but children aged 6–11 had a more negative association between Stringency Index and self-esteem than those aged 12–17 (coefficient for interaction term with 12- to 17-year-olds as reference, −2.44; 95% CI −4.49 to −0.39; Fig. 3D, Additional file 1: Table S6). There was a negative association between Stringency Index and emotional well-being for females (coefficient, −1.67; 95% CI −3.11 to −0.24; Fig. 3E, Additional file 1: Table S6), but this was not statistically significant for males. We did not find a statistically significant association between Stringency Index and self-esteem for either gender (Fig. 3F, Additional file 1: Table S6).

We found that emotional well-being and self-esteem of children and adolescents in Japan improved transiently in the few months from the start of the pandemic to Summer to Autumn 2020, followed by a slight worsening trend into the second year of the pandemic. While emotional well-being stayed lower compared to initial levels nearly 2 years into the pandemic, self-esteem began to improve by late 2021. Our subgroup analyses revealed that emotional well-being and self-esteem were lower for adolescents compared to younger children aged 6–11 years throughout the course of the pandemic, but gender differences were small. Children and adolescents exposed to more stringent social distancing measures tended to have lower emotional well-being and self-esteem in general. However, the association for stringency and self-esteem differed by age group: children aged 6–11 years tended to have lower self-esteem with more stringent measures, while those aged 12–17 years did not display this trend. Furthermore, emotional well-being among females tended to be particularly lower with more stringent measures, although there was no statistically significant gender difference.

The overall trends in emotional well-being and self-esteem mimic that of mental health after disasters: the affected people’s mental health is known to improve transiently immediately after a disaster, then follows a worsening pattern. This is termed the “honeymoon effect,” in which factors like the collective feelings of concern and altruism are thought to prevent the worsening of mental health in the initial phases after an event, and demoralization, anxiety, and lifestyle changes set in in later phases [20]. For instance, after the Great Eastern Japan Earthquake in 2011, suicide death rates in Fukushima, Japan declined in the first 2 years after the disaster but surpassed pre-disaster levels in the third year [21]. While it is typical that this initial improving-worsening pattern occurs over a span of several years, the unprecedented and ongoing nature of the COVID-19 pandemic may have caused it to happen over a span of months. This is further corroborated by the increase in suicide rates in Japan following an initial decline during the COVID-19 pandemic [22], which mirrors the trends in self-esteem in our study sample. Interestingly, in contrast to a previous study that suggested that the detrimental effects of war on the emotional development of young children worsened with chronic exposure [23], the self-esteem of children in December 2021 showed an improving pattern in our study. This is likely multifactorial, but it may be that children and the society started to get used to the pandemic and the social distancing measures, 2 years into the pandemic (whereas it is much more difficult to rebuild society when it is chronically vulnerable to the destructions of war). Furthermore, we found that self-esteem began to improve later in 2021, while emotional well-being did not. A previous study has suggested the mediating role of self-esteem in the relationship between stress and psychological well-being [24], and our findings may reflect this; with improvement in self-esteem, well-being may subsequently improve. To get a clearer picture of the long-term consequences of the pandemic, future studies should track the longitudinal changes in the self-esteem and emotional well-being of children affected by the pandemic over a longer period of time.

Generally, we found that the stringency of social distancing policies was negatively associated with children and adolescents’ emotional well-being and self-esteem, which is in line with previous studies that showed a negative association between the Stringency Index and mental health among adults [7, 8] and young children aged 3–8 years [25]. Our study expands the findings from these previous studies by showing that a negative association exists between the stringency of social distancing policies and self-reported emotional well-being and self-esteem of both children and adolescents. In line with the argument by many experts that school closures and social distancing policies during the pandemic put massive strain on children and adolescents [26, 27], our study suggests that stringent social distancing policies should be implemented with caution to protect the mental health of children and adolescents. Indeed, as evidence suggests that the effects of some social distancing measures (e.g., school closures) in containing COVID-19 may be limited in some settings (although at the time of the study, there were extremely few pediatric COVID-19 cases, and as the authors suggest, it is possible that school closures could have a discernible effect on COVID-19 infections once the reproduction number within a school environment is greater than a certain threshold) [28], policymakers should consider the risks and benefits of implementing such policies, especially from the perspective of children and adolescents. In the context of youth in Japan, the high stringency index could reflect school closures and restrictions in various school-related activities. For instance, children and adolescents spent a significant amount of time away from school early in the pandemic, with spring break in March 2020, national-level school closures from April to May 2020, and summer vacation from July to August 2020. Later in the pandemic, students were still asked to refrain from gathering and socializing at schools and extracurricular activities [9]. Possibly reflecting this, suicides attributed to social concerns increased in Japan from summer 2020, underscoring the importance of regular social interactions [29]. To protect the mental well-being of youth during a pandemic, opportunities for regular social interactions may be beneficial.

The age group and gender differences in emotional well-being and self-esteem found in our study agree with previous studies conducted before the pandemic, which suggest that subjective well-being decreases with age during adolescence [30] and that females tend to report lower subjective well-being than males [31]. The finding that younger and female children may have been especially vulnerable to stringent social distancing policy is, to our knowledge, novel. Younger children may have been especially affected by social distancing policy since they had fewer means to communicate with their peers during school closures and lockdowns: younger children spend less time on social media [32] and fewer of them own mobile phones [33], meaning they are less likely to have access to the primary means of communicating with peers during lockdowns. Furthermore, our finding that females appeared to be more vulnerable to the stringency of closure policy is not unexpected. A previous study has shown that during the COVID-19 pandemic, women reported lower self-esteem while social distancing compared to men [34]. The authors suggested that this relationship may be mediated by the higher need to belong, which may explain our findings too. Analyses of factors mediating the relationship between the stringency of social distancing policies and the changes in mental health are warranted to explore this further.

During the COVID-19 pandemic, the emotional well-being and self-esteem of children and adolescents in Japan improved transiently from the start of the pandemic to Summer to Autumn 2020, followed by a slight worsening trend into the second year of the pandemic. While emotional well-being stayed lower compared to initial levels nearly 2 years into the pandemic, self-esteem began to improve by late 2021. Our finding that more stringent social distancing measures was associated with lower emotional well-being and self-esteem, especially 6–11 year-olds’ self-esteem and females’ emotional well-being, highlights the need to further examine the impact of social distancing measures on children and adolescents’ mental health, with a focus on vulnerable populations.