Unmasking the mental health scars of COVID-19: A longitudinal investigation of children and adolescents in post-lockdown China

The long-term mental health consequences of COVID-19 in children and adolescents remain unclear. We investigated the impact of COVID-19 infection on mental health after China ’ s zero-COVID policy relaxation, focusing on symptom-specific and social-family risk factors for mental health issues in children and adolescents. In a longitudinal study, 8348 youths (aged 10 – 18) were assessed twice (T1: September to October 2022 and T2: April to May 2023). Mental health changes ( Δ = T1-T2) were compared between COVID-19-infected (COVID + , n = 4108) and non-infected (COVID-, n = 4240). After balancing social – family confounding factors at T1 with propensity score-based inverse probability weights, multivariable logistic regression was employed to assess associations between COVID-19 infection and the onset/worsening of mental health symptoms. Multivariable logistic regression was conducted to explore specific acute COVID-19 symptoms and social – family risk factors associated with the onset/worsening of mental health symptoms in COVID + group. Compared to COVID-group, COVID + group exhibited lower overall mental health improvement ( Δ ). COVID + group was associated with increased risks of depression worsening (OR 1.20, 95 % CI 1.04 – 1.39), anxiety worsening (OR 1.30, 95 % CI 1.15 – 1.47), stress worsening (OR 1.23, 95 % CI 1.03 – 1.46), insomnia worsening (OR 1.21, 95 % CI 1.05 – 1.39), and emotional symptoms worsening (OR 1.72, 95 % CI 1.27 – 2.33). Moderate-to-severe difficulty thinking, breathlessness, and gastrointestinal symptoms were specific COVID-19 symptoms associated with worsening of various mental health outcomes. Furthermore, academic difficulties, economic disadvantages, family conflicts, food addiction, and alcohol consumption were identified as social-family risk factors for worsening mental health symptoms in COVID + youths. COVID-19 infection leaves lasting mental health scars in youths, extending beyond the acute phase. Specific symptoms, particularly cognitive dysfunction and respiratory/gastrointestinal distress play a significant role in this vulnerability. Social-family factors further modulate these effects, high-lighting the need for comprehensive interventions that address both biological and psychosocial aspects. This study provides valuable insights for tailoring mental health support to youths navigating the consequences of the COVID-19 pandemic.


Introduction
On December 7, 2022, the Chinese government made the pivotal decision to lift its stringent prevention measures for coronavirus disease-2019 .In the following month, over half of China's population developed COVID-19, and this proportion grew to an estimated 82⋅4% within the next 3 months (Fu et al., 2023).This level of concurrent infection within a vast population is a historical rarity that provides an important window into the effects of COVID-19 on population health.COVID-19 results from a "cytokine storm" that cascades into systemic inflammation, including neuroinflammation, and can lead to persistent physical and neurologic sequelae colloquially known as "Long COVID" (Hingorani et al., 2022;Monje and Iwasaki, 2022;Ramasamy and Subbian, 2021).Mild COVID-19 infection has been associated with year-long symptoms such as dizziness, cognitive impairment, sleep disturbances, depression, and anxiety (Aiyegbusi et al., 2021;Borel et al., 2022;Ceban et al., 2022;Mizrahi et al., 2023).In youths, neurologic and psychiatric symptoms, including cognitive impairment and insomnia, have persisted up to 6 months after mild COVID-19 infection (Taquet et al., 2022).Furthermore, neuropsychiatric sequelae predominantly occurred among acute symptomatic patients with COVID-19 compared to asymptomatic COVID-19 patients (Penninx et al., 2022).At the same time, the greater the number of acute COVID-19 symptoms, the higher the severity of depression and anxiety observed three months later (Tracy et al., 2024).These findings imply that acute-phase symptoms significantly impact the mental health sequelae of COVID-19.
Beyond its physiologic effects, COVID-19 has also impacted population mental health via the drastic social changes as the pandemic evolved.The complex totality of COVID-19′s effects is particularly apparent in youths worldwide.Childhood and adolescence are critical periods for physical, emotional, social, and academic development Y. Zhu et al. influenced by various factors such as family environment, stress exposure, and healthcare access.COVID-19 disrupted learning, social networks, and healthcare access, as well as intensified socioeconomic differences and existing social problems, leading to increased depression, anxiety, and psychological distress during the pandemic (Barbosa-Camacho et al., 2022;Borel et al., 2022;Sardar et al., 2020).In China, the rigor and duration of COVID-19 precautions, including strict confinement and social distancing, were added challenges for youths.Notably, youths exposed to prolonged periods of confinement are more prone to serious mental health issues than adults (Hawes et al., 2022;Murata et al., 2021).Thus, COVID-19 has brought child and adolescent mental health to the forefront of global public health priorities.Longitudinal population-based studies are urgently needed to further understand the long-term effects of COVID-19 in youths and identify effective interventions for the current mental health crisis among the world's youths.
In this study, we examined the mental health effects of COVID-19 infection and associated risk and protective factors among school-aged youths in Jiangsu, China from T1 (September to October 2022) to T2 (April 2023 to May 2023), before and after the lifting of COVID-19 restrictions.The high concurrence of COVID-19 infection within the population between T1 and T2 minimized potential confounding effects such as timing between COVID-19 infection and major societal shifts.The study was embedded within a five-year comprehensive assessment of youth health in China, School-based Evaluation Advancing Response to Child Health (SEARCH) (Zhang et al., 2023).Thus, this work is poised for continued longitudinal monitoring of COVID-19 mental health effects among school-aged youths.

Study design
As above, this study was embedded in SEARCH, a longitudinal population-based assessment of mental well-being and related factors in 4th to 12th grade students, ages 10 to 18 years, geographically distributed across Jiangsu, China, conducted in collaboration with the Institute of Child and Adolescent Health Promotion, Jiangsu Provincial Center for Disease Control and Prevention (CDC).All participants were enrolled in grade-based school cohorts within the SEARCH study.The participants sampling methodology proceeded as follows: initially, invitations were sent to the CDC departments of three surveyed cities across different regions of Jiangsu province (north, central, and south).Subsequently, these departments circulated detailed information about the SEARCH program to all elementary and secondary schools within their jurisdiction.Ultimately, 11 schools participated in the cohort study.Participants completed questionnaires using the 'SEARCH' website designed by our research team on computer in the school's computer classroom.More details can be found in the SEARCH protocol (Zhang et al., 2023).The present study analyzed data from the first two SEARCH survey waves, T1 (September − October 2022) and T2 (April − May 2023) and included 8348 youths [4500 males and 3848 females; mean (SD) age of 14⋅07 (2⋅26) years] (Fig. 1).
All participants and their legal guardians provided informed consent and assent after detailed study description.The study was approved by the ethics committee of the Affiliated Brain Hospital of Nanjing Medical University (2022-KY095-02).

Study measures
Prior COVID-19 infection was confirmed and characterized using self-report questionnaires.Questions included: 1) Was this the first-time COVID-19 infection confirmed through the polymerase chain reaction (PCR), antibody, or lateral flow test, or the presence of COVID-19 positive symptoms after December 7, 2022? 2) Did you experience any of the acute-phase symptoms of COVID-19 infection and their severity?3) What is the overall severity of acute COVID-19 infection symptoms?4) What is the cumulative extent of psychological distress experienced by you during the COVID-19 epidemic?5) What is the level of concern about the virus before COVID-19 infection?(appendix p1).We assessed for 13 acute-phase symptoms (fatigue, headache, sleep disturbances, decreased attention/memory, difficulty thinking, pharyngeal symptoms, nasal congestion, breathlessness, fever, cough/phlegm, olfactory impairment, gastrointestinal symptoms, fear of COVID-19 virus).
Symptom severity was assessed using the 21-item Depression, Anxiety, and Stress Scale (DASS), Insomnia Severity Index (ISI), and the Strengths and Difficulties Questionnaire (SDQ).DASS measures depressive and anxiety symptoms and stress over the past week [depression subscore (>9 as presence of depressive symptoms), anxiety subscore (>7 as presence of anxiety symptoms), and stress subscore (>15 as presence of stress symptoms)].ISI measures sleep problems over the past 2 weeks, and total score > 7 defined as presence of insomnia.SDQ measures internalizing symptoms [i.e., emotional symptom subscore (>6 as presence of emotional symptoms) and peer problem subscore (>5 as presence of peer problems)], externalizing symptoms [i.e., conduct problem subscore (>4 as presence of conduct problems) and hyperactivity-inattention subscore (>6 as presence of hyperactivity-inattention problems)], and prosocial behavior subscore (<5 as abnormal of prosocial behavior) over the past 6 months.In total, nine symptom measures (DASS depression, anxiety, and stress subscores; ISI total score; SDQ emotional symptom, peer problem, conduct problem, hyperactivity-inattention, and prosocial behavior subscores) were used in subsequent analyses.Please see appendix p2 for further details.
School-related, family, and lifestyle factors were assessed at T1 1) to control for baseline differences in analysis of mental health effects of COVID-19 infection and 2) to determine risk and protective factors for changes in mental health status in youths with COVID-19 infection.School-related factors consisted of school level [primary-school level (10-12 years old), middle-school level (12-15 years old), or high-school level (15-18 years old)]; leadership status in school (yes/n0); academic ranking (top 25 %, 26-50 %, 51-75 %, or bottom 25 %); and change in academic performance from T1 to T2 (progress, stabilization, or regression).Family factors consisted of economic resource level as determined by place of residence (affluence, moderate, or poverty); only child status (yes/no); and frequency of family conflicts (never, sometimes, or always).Lifestyle factors consisted of food addiction over the past year as assessed by Modified Yale Food Addiction Scale (yes/no); cigarette use over the week (yes/no); alcohol use over the past week (yes/no), and regular physical exercise before the outbreak of the pandemic [frequency (infrequent, 1-3 days per week, or 4-7 days per week); intensity (no exercise, mild (e.g., walking), moderate (e.g., cycling, tai chi, table tennis, and badminton), or vigorous (e.g.,s running, swimming, soccer, basketball, and jumping rope); duration (<30 min, 30-60 min, or > 60 min)].

Statistical analysis
All statistical analyses were performed using R Studio software (4.3.1).All statistical significances were determined using a two-sided Pvalue threshold of < 0⋅05.

Effects of COVID-19 infection on mental health status changes using risk stratification
Participants were stratified into three groups based on their mental health status changes from T1 to T2 for each of the nine symptom measures: healthy controls (HC; asymptomatic at T1 and T2), symptom Y. Zhu et al. onset (ONS; asymptomatic at T1 and symptomatic at T2), and symptom worsening (WORS; symptomatic at T1 and T2, increased symptom scores from T1 to T2).The HC group consisted of those who were asymptomatic at T1 and T2 for all nine symptom measures.Sample sizes for the HC, ONS, and WORS groups for each symptom measure are listed in appendix p3.Participants who were symptomatic at T1 and asymptomatic at T2 were excluded.These participants were excluded as their symptom changes were not associated with COVID-19 infection (appendix p4), not aligning with the primary aim of this study.
Propensity score-based weighting was used to control for baseline (T1) school-related, family, and lifestyle confounding factors between COVID + and COVID-groups.For each symptom measure, we constructed separate logistic regression models to estimate the propensity scores for each participants after controlling for T1 school-related, family, and lifestyle factors.Then we calculated propensity scorebased inverse probability weights (IPW) for each participant, and employed multivariate logistic regression models for each symptom measure with COVID-19 infection status (COVID + and COVID-) as independent variable and mental health status (HC, ONS, and WORS) as dependent variable, in terms of Odds Ratio (OR) and 95 % confidence interval (CI).

Risk and protective factors for mental health status changes in COVID + group
We employed lasso regression to address collinearity and overfitting issues of 13 acute COVID-19 symptoms.Multivariate logistic regression models were used to evaluate the effects of acute COVID-19 symptoms, and school-related, family, and lifestyle factors on mental health status changes (HC, ONS, and WORS) in terms of OR and 95 % CI.

Results
The study flowchart is shown in Fig. 1.

Changes in mental health status associated with COVID-19 infection
After inverse probability weighting, the confounding factors were well balanced between COVID + and COVID-groups (appendix p5-9).The T test and X 2 test were preformed between youths without (COVID-) and with (COVID + ) COVID-19 infection.

Effects of acute COVID-19 symptoms, social and family factors on mental health status in COVID + group
Severity levels for all 13 acute COVID-19 are presented in appendix p10.After conducting lasso regressions, decreased attention/memory, difficulty thinking, breathlessness, cough/phlegm, gastrointestinal symptoms and feeling fear of COVID-19 virus (appendix p11) were included for subsequent multivariate logistic regression analyses assessing the effect of acute COVID-19 symptoms on mental health status.

Moderate to sever
Exc.

Level of concern about the virus before COVID-19 infection
Family-related variables Economic status at the place of residence Moderate 4⋅38 2⋅65-7⋅22

Daily life variables Food addiction
Yes 3⋅61 1⋅99-6⋅56 were positively associated with conduct problems ONS.Moderate-to-severe decreased attention/memory, difficulty thinking, breathlessness, and gastrointestinal symptoms showed a higher risk of depression, anxiety, stress, insomnia, emotional symptoms, and conduct problems (Table 2a).However, overall severity of acute COVID-19 symptoms was not significantly associated with mental health symptoms.Moderate-to-severe fear of COVID-19 virus during the acute phase was positively associated with ONS and WORS status (Table 2).However, the level of concern about the virus before COVID-19 infection was not significantly associated with mental health status.Interestingly, before COVID-19 infection, the cumulative extent of psychological distress induced by pandemic restrictions was identified as a risk factor for ONS and WORS status (Tables 2 and 3).
Being ranked in the bottom 25 % of academic performance, and academic performance regression increased risk for depression worsening, as well as for anxiety worsening (Table 2).Living in poverty or moderate economic status, frequent family conflicts, having food addiction, and alcohol consumption were positively associated with ONS and WORS status (Table 2).Engaging in regular moderate-intensity physical exercise was negatively associated with ONS and WORS status (Table 2).

Discussion
We believe this is the first population-based longitudinal study examining the effects of COVID-19 infection on youth mental health.The study capitalized on the coincidental lifting of COVID-19 restrictions in China from T1 and T2 and was able to examine a large cohort experiencing COVID infection within the same time period.Thus confounding effects related to major societal shifts were minimized.
Of the 8348 youth participants, 49⋅2% self-reported first-time COVID-19 infections.We found significant differences in depression, anxiety symptoms, internalizing symptoms, and externalizing symptoms between COVID + group and COVID-group.Specifically, the mental health of the COVID + group worsened, while that of the COVID-group showed improvement.Even after controlling for baseline school-related, family and lifestyle factors, COVID-19 infection is a risk factor for the onset and worsening of emotional and sleep problems.This study concurrently utilized the DASS and SDQ to assess emotional symptoms because they assess symptoms over different time frames.The DASS evaluates short-term symptoms, particularly within the past week, which aligns well with the timing of the two surveys conducted during academic terms.Thus, the DASS reflects emotional differences before and after infection within the school context.On the other hand, the SDQ provides a broader assessment by capturing long-term symptoms over the past six months.This allows for the assessment of emotional symptom severity across the period from T1 to T2.The combined use of these scales confirms COVID-19 infection as a risk factor for both shortterm and long-term emotional symptom worsening, thereby bolstering the robustness and credibility of the study's findings.
Mental health symptoms lasting longer than one month are considered a manifestation of long-term COVID, which is attributed to the persistent multisystem inflammatory syndrome in the nonacute phase of COVID-19 (Jiang et al., 2020).The unresolved systemic inflammation affects immune cell transmigration, blood-brain barrier permeability, and neurotransmission mechanisms, potentially leading to long-lasting neuroinflammation and providing a basis for persistent psychiatric manifestations (Mazza et al., 2020).Besides, evidence suggests that immune-inflammatory dysregulation may attenuate the efficacy of antidepressant treatments (Liu et al., 2020).These findings suggest that addressing the systemic immune response triggered by COVID-19 is a crucial factor for prevention and treatment of mental health symptoms.
The prevailing acute symptoms of COVID-19 included fatigue (95 %) and headache (82 %) (Bliddal et al., 2021), while psychiatric symptoms in the acute phase are relatively mild but long-lasting.The invasion of different organ tissues by SARS-CoV-2, the virus that causes COVID-19,

Table 2b
Risk and protective factors of ONS and WORS in COVID + group (SDQ) − continued.
ONS: symptom onset; WORS: symptom worsening; OR: Odds Ratio; CI: confidence interval.The healthy controls as reference.Acute-phase symptoms of COVID-19 infection: no acute symptom as reference; gender: male as reference; academic levels: primary level as reference; rank of academic performance: top 25% as reference; changes in academic performance: progress as reference; economic status at the place of residence: affluence as reference; frequency of family conflicts: never as triggers various acute symptoms, which are related to the subsequent occurrence of mental symptoms.For instance, SARS-CoV-2 can persistently survive for 30 days in the upper respiratory and gastrointestinal epithelium (Wu et al., 2020), likely resulting in brain-gut axis dysfunction, that as a possible reason for mental health symptoms of long COVID (Blackett et al., 2022).Our founding simultaneously confirm that gastrointestinal symptoms are risk factors for psychiatric symptoms onset and worsening.Besides, moderate-to-severe difficulty thinking, and breathlessness also found as risk factors for the worsening of mental health issues post-infection.Systemic inflammation after COVID-19 infection is known to promote cognitive decline (Heneka et al., 2020), while poor cognition is often accompanied by emotional symptoms (Wei et al., 2023;Zhu et al., 2019).Worsham et al. found that the deterioration of breathlessness can be a psychologically traumatizing sensation (Worsham et al., 2021).The respiratory system's role in modulating the autonomic nervous system, respiratory techniques have widely employed in clinical psychotherapy to enhance mental well-being (Balban et al., 2023;Wu et al., 2019).Implementing breathing training during COVID-19 recovery may prevent mental health symptoms for those who experienced breathlessness in the acute phase.In light of our findings,the overall severity of acute COVID-19 symptoms was not significantly associated with mental health symptoms, which indicats that it is importance of focusing on specific symptoms in the acute phase for youths.Another interesting finding is that, during the acute phase of infection, a heightened fear of the virus emerged as a risk factor for mental health isusses.However, before COVID-19 infection, concerns about the virus itself did not impact mental health status post-infection.Conversely, psychological distress due to pandemic policy measures is a known risk factor for the onset of depression and anxiety symptoms.Combined with our previous finding that mental health symptoms improved in COVID + group after the comprehensive lifting of epidemic restrictions.The containment policies did indeed burden the mental health of children and adolescents (Collaborators, 2021), but this is expected to be alleviated by the lifting of restrictions.In addition to acutephase fear of the virus and psychological distress due to pandemic policy measures, academic difficulties, economic disadvantage, family conflict, and unhealthy dietary habits were identified as risk factors for mental health status post-infection.Our findings emphasized that, regardless of COVID-19 infection, older students facing higher academic pressure, ranking in the bottom 25 %, and experiencing academic regression encountered challenges to their mental well-being.Previous study has also found that declined academic performance, due to COVID-19 was a risk factor for depression and anxiety (Song et al., 2023).All that suggests that fostering healthy academic values in students and adopting constructive attitudes in parents and teachers when evaluating students' performance can significantly safeguard the mental well-being of children and adolescents in the post-pandemic era.COVID-reduced financial challenges, and family conflicts were associated with poor mental health (Temple et al., 2022).Creating favorable living environments holds particular significance for youths post-COVID infection.
Encouragingly, maintaining a light-to-moderate exercise at least once a week demonstrated protective effects against post-infection mental health issues.Previous research confirming that regular exercise is associated with a lower risk of depression, anxiety, and insomnia (Yuan et al., 2022).Individuals engaging in regular exercise can lead to a reduction in inflammation, potentially providing relief from mental health symptoms (Paolucci et al., 2018).That suggests promoting healthy lifestyles to alleviate the post-COVID infection mental health problems.This study has the following limitations.We used selfreporting for COVID-19 infection and lacked validation with laboratory results.We also did not evaluate other sequelae of COVID such as physical pain.This could be an avenue for future research on the comprehensive impact of long COVID on individuals' mental well-being.
In conclusion, after balancing the family-social factors, our findings indicate that COVID-19 infection is a risk factor for mental health issues in children and adolescents.Acute symptoms of difficulty thinking, shortness of breath, and gastrointestinal symptoms are also risk factors for long COVID-related mental health symptoms.Notably, concerns about the virus before infection do not impact mental health, but fear of the virus during the acute phase of infection emerged as a risk factor for post-infection mental health issues.Given the long-lasting and recurring infections of COVID-19 in China and worldwide, these findings have important public health implications and highlight the need for greater attention to acute symptoms to alleviate post-infection mental health symptoms.Furthermore, academic difficulties, high family environmental stress, and unhealthy lifestyle habits contribute to the mental burden on children and adolescents infected with COVID-19.Schools and families are expected to create a favorable living environment while promoting healthy lifestyles to alleviate mental health problems associated with long COVID.

Fig. 1 .
Fig. 1.Flowchart of children and adolescents invited to participate in the study.
exercise(continued on next page) Y.Zhu et al.
P value OR 95 %CI P value OR 95 %CI P value OR 95 %CI P value OR 95 %CI P value OR 95 %CI P value OR 95 %CI P value OR 95 %CI P value 30-60 min

Table 1
The characteristics of participants.

Table 2a
Risk and protective factors of ONS and WORS in COVID + group (DASS and ISI).