Association of Major Disease Outbreaks With Adolescent and Youth Mental Health in Low- and Middle-Income Countries: A Systematic Scoping Review

Key Points

Question  Are major disease outbreaks from 2009 to 2021 associated with the mental health of adolescents and youth in low- and middle-income countries?

Findings  In this systematic scoping review of 6 databases, a total of 57 studies of the influenza A (H1N1) and SARS-CoV-2 infections revealed that these outbreaks were associated with adolescent and youth mental health. Results suggest high rates of anxiety and depressive symptoms, in addition to posttraumatic stress disorder, general stress, and health-related anxiety among adolescents.

Meaning  Findings suggest that the H1N1 and SARS-CoV-2 outbreaks were associated with adolescent and youth mental health; future studies with improved measurement tools and the inclusion of a wider range of mental disorders and risk factors will help ascertain how epidemics affect adolescent mental health in low- and middle-income countries.

Importance  Adolescents and young people have been historically understudied populations, and previous studies indicate that during epidemics, these populations, especially in low- and middle-income countries (LMICs), are at high risk of developing mental disturbances.

Objective  To identify the existing evidence regarding the association of mental health with outbreaks of the influenza A (H1N1), Zika, Ebola, and SARS-CoV-2 virus in exposed youth and adolescents in LMICs.

Evidence Review  Across 6 databases (Embase, Cochrane Library, PubMed, PsycINFO, Scopus, and Web of Science), the mental health outcomes of adolescents and youth (aged 10-24 years) associated with 4 major pandemic outbreaks from January 2009 to January 2021 in LMICs were reviewed. A group of 3 authors at each stage carried out the screening, selection, and quality assessment using Joanna Briggs Institute checklists. The social determinants of adolescent well-being framework was used as a guide to organizing the review.

Findings  A total of 57 studies fulfilled the search criteria, 55 related to the SARS-CoV-2 (COVID-19) pandemic and 2 covered the H1N1 influenza epidemics. There were no studies associated with Zika or Ebola outbreaks that met screening criteria. The studies reported high rates of anxiety and depressive symptoms among adolescents, including posttraumatic stress disorder, general stress, and health-related anxiety. Potential risk factors associated with poor mental health outcomes included female sex; home residence in areas with strict lockdown limitations on social and physical movement; reduced physical activity; poor parental, family, or social support; previous exposure to COVID-19 infection; or being part of an already vulnerable group (eg, previous psychiatric conditions, childhood trauma, or HIV infection).

Conclusions and Relevance  Results of this systematic scoping review suggest that the COVID-19 pandemic and H1N1 epidemic were associated with adverse mental health among adolescents and youth from LMICs. Vulnerable youth and adolescents may be at higher risk of developing mental health–related complications, requiring more responsive interventions and further research. Geographically localized disease outbreaks such as Ebola, Zika, and H1N1 influenza are highly understudied and warrant future investigation.

Individual, familial, and socioeconomic vulnerabilities and attitudes toward quarantine and other public health measures influence adolescents’ mental health during disease outbreaks.¹ Many risk-taking behaviors, such as heavy drinking, substance use, or sexual risk-taking, begin during adolescence.^2,3 Economic instability and changes in routine and recreational activities have severely affected family functioning and have led to increased high-risk behaviors.^1,4

Recent studies have revealed that, for adolescents, the consequences of the COVID-19 lockdown and protracted quarantine and closures could be severe anxiety and depression, acute stress disorders, and posttraumatic stress disorder (PTSD).^5-8 In low- and middle-income countries (LMICs), these disturbances co-occur with multilevel risk factors and abject material conditions, known as social determinants of adolescent well-being, that affect life outcomes (Figure 1).

This systematic scoping review attempted to evaluate the associations of 4 major disease outbreaks with adolescent mental health in LMICs, considering all types of relevant study designs. The influenza A (H1N1) outbreak of 2009,⁹ popularly known as swine flu, the Zika virus pandemic of 2015,¹⁰ the Ebola hemorrhagic fever of 2013 to 2019,¹¹ and the most recent and ongoing COVID-19 pandemic, which started in late 2019,¹ are studied. We also evaluated the potential psychological and mental health associations of vulnerable adolescent and youth populations with these outbreaks (eAppendix in the Supplement).

Based on our inclusion criteria (eTable 1 in the Supplement), a comprehensive search strategy was used to identify potential studies from 6 electronic databases: Embase, Cochrane Library, PubMed, PsycINFO, Scopus, and Web of Science. H.A. and M.K. designed the search strategy, and H.A., with M.F.K. and M.H.T., searched databases. Where the database allowed, restrictions were added to ensure that only human studies were searched and published after 2009, when the first outbreak occurred. The search strategy for PubMed is available in eTable 2 in the Supplement, and the rest are available in eTables 3 to 7 in the Supplement. The search was carried out between January 15 to 17, 2021. The articles were imported, and duplicates were removed. This study followed the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) Extension for Scoping Reviews statement.¹²

Two authors (P.S., S.K.) independently reviewed the abstracts and titles using the Rayyan software.¹³ There was substantial agreement between the 2 authors (κ of 0.77). If there were any doubts concerning the fulfillment of these criteria, it was resolved through discussion between M.K., R.S., P.S., S.K., and H.A. (eAppendix in the Supplement).

Next, full-text articles were assessed by H.A., M.H.T., and M.F. to review if the included studies met the inclusion criteria. We emailed the corresponding author twice for the requested article if an article was unavailable. No restrictions were applied to languages. After selecting the eligible articles, 3 authors (H.A., M.H.T., and M.F.) separately extracted the required items based on the extraction form designed for the process (eTable 8 in the Supplement). Any discrepancies or disagreements in data extraction were resolved through discussions between M.H.T., M.F., and H.A. until a consensus was reached. We tabulated the results based on the selected items (eTable 9 in the Supplement) to systematically summarize the results of every disorder for each outbreak studied. Based on a systematic scoping review’s remit,¹⁴ this study categorized the publications and the mentioned data items to map the current evidence.

Along with the JBI SUMARI tool (Joanna Briggs Institute),¹⁵ we applied the JBI critical appraisal checklists (eAppendix in the Supplement) to assess the methodological quality of the included studies. The responses selected for each item in the checklist were yes, no, unclear, or not applicable (eTables 10-13 in the Supplement). Notably, as this was a scoping review aiming to map the literature and indicate the gaps, all the studies were included. We performed the risk of bias assessment for only those studies that exhibited a vigorous methodology.

In the final step, we synthesized the data based on a conceptual framework that builds on articulating risk and protective factors associated with the social determinants of health approach. Viner et al¹⁶ have approached the intersection of multilevel factors for adolescent mental health well-being in a similar conceptual model. We have framed our results and discussion, keeping these broad domains in sight (Figure 1).

A total of 57 studies^17-73 were included in the analysis (Figure 2 and eTable 8 in the Supplement), of which 55^{17-55,57-61,63-73} focused on the SARS-CoV-2 virus (COVID-19 pandemic), and 2^23,40 focused on H1N1 (eTable 9 in the Supplement). None of the Zika or Ebola outbreak studies were included as they did not satisfy our inclusion criteria (Figure 2 and eTable 8 in the Supplement).

Most of the studies (35 [61%])^{17-19,23,27,30-33,35,38,40,42-45,47-49,53-61,65,66,69-73} reported findings from China. Other countries that were represented included Argentina (1),³⁹ Bangladesh (4),^24,25,28,62 Brazil (1),²² Ethiopia (1),⁴⁶ Egypt (1),²¹ Indonesia (1),⁴¹ India (6),^{34,37,63,64,67,68} Jordan (1),³⁶ Kenya (1),²⁰ Malaysia (1),²⁶ Nepal (2),^67,68 Nigeria (2),^67,68 and Turkey (3)²⁹ (Figures 3 and 4).

A total of 55 studies^{17-55,57-70,72,73} adopted a cross-sectional study design, and the majority of them used convenience or snowball sampling. One study⁷¹ used a longitudinal cohort, and 1 study⁵⁶ was a randomized clinical trial.

The age range of study participants varied, although the participants in most studies belonged to the age group of 17 to 24 years. Certain studies with mixed participants were only partially included.

Two H1N1 studies^23,40 were found and included in the review. These were mainly about the emotional challenges associated with misinformation or exposure to quarantine.

One study done by Gu et al²³ did not specifically assess the relationship between sex and age and the psychological effect of the H1N1 on youth and adolescents. Another study by Wang et al,⁴⁰ which focused on the psychological effects of quarantine during the H1N1 epidemic, revealed no difference in scores associated with the psychological association between men and women who were under quarantine and those who were not. The same results found that age may not be significantly associated with the negative effect of quarantine.

In China, a study²³ on H1N1 exposure in students of a university found that half of the included students had misconceptions about its transmission. The study found that among 10.7% of participants (88 of 825), high/very high fear of contracting H1N1 was associated with some form of mental distress, that is, panic, depression, or emotional disturbances (odds ratio [OR], 3.81; 95% CI, 1.95-7.44). Overall, 45% of participants (371 of 825) were worried that a family member would contract H1N1. In another study by Wang et al,⁴⁰ there were no significant differences in the immediate negative psychological outcomes on those quarantined with H1N1 and those not quarantined⁴⁰; however, dissatisfaction with control measures was a predictor of positive scores on the Impact of Event Scale-Revised (OR, 2.22; 95% CI, 1.37-3.60; P = .049) and Self-Report Questionnaire-20 (OR, 2.22; 95% CI, 1.28-3.85; P = .005).

The vast majority of studies took place from July through December 2020. Depressive and anxiety symptoms were the most commonly studied and reported.

Younger students (aged ≤15 years) reported less depressive and anxiety symptoms than older students across a few studies.^25,27,45 Being a more senior student (older than 15 years) was also associated with PTSD symptoms in China (211 of 2485 students; mean [SD] score on Posttraumatic Stress Disorder Checkist–Civilian Version trauma assessment, 23.4 [8.4]; F = 9.89; P < .001),³⁸ and similar findings were reported in Bangladesh (822 of 3122 students; age, 25-29 years; Depression, Anxiety, Stress Scale (DASS) mean [SD] score, 54.1 [29.5]; β = 0.03; P = .047),²⁵ China (OR, 2.89; 95% CI, 0.96-8.67),⁴² and Malaysia (age >25 years; anxiety: adjusted OR, 0.56; 95% CI, 0.35-0.89; P = .02),²⁶ although another study³² found that younger students were more likely to experience PTSD (OR, 8.71; 95% CI, 1.97-38.43). In Malaysia, the prevalence of psychological distress was 52.8% in 408 of 772 school-age children, with older students less likely to be anxious than younger students (adjusted OR, 0.56; 95% CI, 0.55-0.89; P = .02).²⁶ In Argentina, depressive symptoms were higher among youth aged 18 to 24 years than in the adult population, with 57.8% (5811 of 10 053; mean [SD] 9-item Patient Health Questionnaire score, 9.8 [6.2]) reporting being possibly depressed in comparison with 33.7% (3388 of 10 053) for the total sample.³⁹ In Brazil, anxiety symptoms were reported by 18% of youth (10 of 55) aged 10 to 11 years and 20% of youth (6 of 30) aged 12 years.²²

Women were more likely than men to report higher rates of depressive and anxiety symptoms (China: 57.4% of women [1201 of 7866] with anxiety symptoms),^18,29,36 and lower rates of knowledge about COVID-19 infection mitigation.^34,36,44 In China, a study⁴⁴ among seventh to 12th graders found that female students had significantly higher mood disturbance scores than their male counterparts (t [1,678] = −3.26; P < .01), and more senior students in high school had higher mood disturbance scores than those in middle school (F [5, 1674] = 6.82; P < .001). The only study on insomnia found that women reported higher rates of insomnia than men (OR, 1.38; 95% CI, 1.21-1.57).⁴⁹ Being female was a risk factor for depression and anxiety in another study from China (OR, 1.33; 95% CI, 1.19-1.50; P < .001).¹⁸ A study³⁶ from Jordan found that female adolescents had higher rates of anxiety symptoms than their male counterparts (mean anxiety scores: men, 7.7 vs women, 8.6; P = .002).

A study⁴¹ from Indonesia found that 10.6% of participants (12 of 113) were at risk of emotional problems, 15.0% (17 of 113) for conduct behavior, 38.1% (43 of 113) for peer-related problems, 8% (8 of 113) for hyperactivity, and 28.3% (32 of 113) for prosocial behavior problems. The subjective sense of the participant’s mental well-being amid the COVID-19 pandemic was not significantly correlated with emotional problems (OR, 0.20; 95% CI, 0.04-1.01). Meanwhile, the prosocial problem behaviors (OR, 0.14; 95% CI, 0.02-0.75) and parental support (OR, 0.09; 95% CI, 0.14-0.60) reduced total difficulties and were associated with a personal sense of mental well-being.⁴¹ Among college students in Indonesia, those with sleep issues and less income reported higher rates of probable depression, and in China, less than 6 hours of sleep was associated with depression (β =1.850; SE = 0.065; OR, 6.361; 95% CI, 5.60-7.23; P < .001).⁴⁹ In Bangladesh, students living with families were 1.8 times (95% CI, 1.02-3.31) more likely to have mild to severe anxiety symptoms.²⁴

Children with siblings were more likely than children without siblings to have anxiety and depression symptoms (OR, 1.16; 95% CI, 1.06-1.27; P = .001).¹⁷ The quality of the family environment and parent-child relationships were protective factors (OR, 0.62; 95% CI, 0.55-0.70; P < .001).¹⁷ In China, college students who had experienced sexual abuse could have a greater risk of anxiety (adjusted OR, 1.39; 95% CI, 1.20-1.60; P < .001) and depression (adjusted OR, 1.96; 95% CI, 1.37-2.80; P < .001) in early adulthood and more than double the risk of acute stress reactions (adjusted OR, 2.73; 95% CI, 1.47-5.05; P = .001) during this outbreak than those who did not experience sexual abuse. For participants with 4 areas of early adversity during their childhood, the risk of acute stress reactions reached 2.92 (95% CI, 1.82-10.38; P = .009).³⁰ Furthermore, they also reported higher rates of PTSD after being exposed to SARS-CoV-2 infection (95% CI, 1.82-10.38; P = .009).³⁰ In Bangladesh, students living with families reported higher depression (OR, 2.595; 95% CI, 1.42-4.75; P = .002).^24,28 Perceived overprotective parenting and less warmth from parents correlated with higher rates of anxiety among adolescents in China (overprotection: r = 0.06; P < .001; lack of emotional warmth; r = −0.17; P < .001).¹⁸ In Turkey, it was found that poor parental mental health due to the COVID-19 pandemic may be linked to poorer mental health outcomes in young adolescents ages 12 to 13 years. In Turkey, adolescents with previous psychiatric referrals showed higher scores on anxiety scales (OR, 4.39; 95% CI, 2.48-25.30; P = .01), along with those who had a family member with COVID-19 infection (OR, 3.81; 95% CI, 1.78-13.57; P = .02).²⁹ In Kenya, youth with vertically transmitted HIV status reported higher levels of depressive symptoms and were consistent in reporting older adolescents having higher depressive symptoms than their younger counterparts (21% for participants aged 20-24 years; 5% for 15-19 years; and 6% for 10-14 years; P < .001). This could be attributed to the lockdown, which was associated with decreased social contact and resources for support.²⁰ In Ethiopia, young adults experienced significant acute stress reaction symptoms (32 of 374 [29.4%]).⁴⁶ In Uttar Pradesh and Bihar in India, young women were more likely to report depressive symptoms than young men (β = 0.06; 95% CI, 0-0.11), and women with a history of experiencing violence were likely to be depressed (β = 0.30; 95% CI, 0.13-0.48).³⁴

Youth residing in areas with more restrictions experienced decreased recreational and physical activity compared with pre–COVID-19 pandemic times. Male students and older students were more likely to report depressive and anxiety symptoms in China (using DASS, depression: β = −1.26; t = −7.96, P < .001; anxiety: β = −0.70; t = −5.64; P < .001; stress: β = −1.01; t = −6.21; P < .001; total score: β = −2.97; t = −7.20; P < .001),¹⁹ and lower depressive scores were associated with regular exercise and an exercising schedule (depression: B = −1.26; t = −7.96; P < .001; anxiety: β = −0.70; t = −5.64; P < .001; stress: β = −1.01; t = −6.21; P < .001; total score: β = −2.97; t = −7.20; P < .001).⁴⁷ Students who exercised at a level of moderate and high according to scores on the International Physical Activity Questionnaire Short Form reported less depression, confusion, anger, and fatigue symptoms compared with those with a low level of activity (depression: β = −2.02; t = −9.17; P < .001; anxiety: β = −1.21; t = −6.99; P < .001; stress: β = −2.20; t = −9.79; P < .001; total score: β = −5.43; t = −9.49; P < .001).⁴⁷ Those with increased screen time and extreme worry, including thoughts of spreading COVID-19 infection to family members, seemed to experience more acute stress and depressive symptoms (in Pakistan: 54.8% of participants [217 of 396] aged 15-18 years experienced severe anxiety; OR, 1.76; 95% CI, 1.55-1.99; P < .001; participated in distant learning: OR, 0.71; 95% CI, 0.56-0.89; P = .004; were concerned about COVID-19: OR, 0.41; 95% CI, 0.21-0.81; P = .01; experienced sleep duration per day of <6 hours: OR, 2.60; 95% CI, 1.95-3.48; P < .001; experienced physical exercise duration per day of <30 minutes: OR, 1.64; 95% CI, 1.46-1.85; P < .001).^48,74 Increased screen time, including exposure to social media, was associated with a higher mood disturbance score, whereas less screen time and more physical activity were linked to fewer conflicts with parents (sleep duration per day <6 hours: OR, 2.60; 95% CI, 1.95-3.48; P < .001; physical exercise duration per day of <30 minutes: OR, = 1.64; 95% CI, 1.46-1.85; P < .001) and were independent predictors of depression.^31,33,48,74 Bangladeshi college students who did not exercise regularly reported higher rates of mild to severe anxiety symptoms (1727 of 3122 students; β = 0.09; P < .001).²⁵ Among female adolescents in China, those who had physical activity for less than 30 minutes per day had higher odds of experiencing depression than those who had physical activity for more than 30 minutes per day (1096 of 4805 students [46.4%]; OR, 1.64; 95% CI, 1.46-1.85; P < .001).⁴⁸

Poor parental and peer support were associated with a higher risk of depressive and anxiety symptoms in Indonesia.⁴¹ A study²⁶ conducted in Malaysia identified that students receiving social and governmental support and greater general support reported lower frequencies of depression (adjusted OR, 0.68; 95%, CI, 0.47-0.99; P = .04) and stress (adjusted OR, 0.53; 95%, CI, 0.35-0.80; P = .003) and improved mental well-being (adjusted OR, 1.54; 95%, CI, 1.06-2.22; P = .02). Furthermore, financial uncertainty (β = 3.7; 95% CI, 2.01-5.34), fear of infection (β = 1.20; 95% CI, –0.01 to 2.41), and inadequate food supply (β = 2.94; 95% CI, 0.76-5.13) were associated with higher rates of anxiety, depression, and stress symptoms.²⁸ In Jordan, those with a lower income were reported to have higher rates of anxiety symptoms than their counterparts (151 of 1540 adolescents [38.0%]; P < .001).³⁶ False news from social media was another predictor in higher anxiety reporting (mean anxiety score men vs women, 9.0 vs 7.7; P = .004), although prolonged quarantine was not a predictor. During quarantine, adolescents in India commonly reported worry (68.6%), helplessness (66.1%), and fear (62.0%); children who had experienced quarantine had higher rates of fear, nervousness, and annoyance.³⁷

Higher rates of COVID-19 infection and worries concerning the higher rates of infection in the community were associated with a higher reporting of depressive symptoms. Those with actual exposure and worries about exposure to COVID-19 infection faced a greater risk of anxiety symptoms in China (OR, 0.840; 95% CI, 0.739 to 0.956, P = .008).³⁵ Reports on excessive worries about COVID-19 infection and loneliness were associated with depression and anxiety symptoms among adolescents (OR, 3.81; 95% CI, 1.78-13.57 P = .02).²⁹ Being in an area with high rates of COVID-19 infection was also associated with PTSD (z = 2.27; 95% CI, 1.87-2.67) and depression (z = 1.23; 95% CI, 0.96-1.50; P = .001).³⁸

Certain studies included young health care workers and medical students, who are particularly prone to mental distress due to being on the front lines of the pandemic.⁷⁵ In Malaysia, senior clinical students (years 4-5) reported less anxiety and stress than junior students (anxiety: adjusted OR, 0.55; 95% CI, 0.41-0.74; P < .001; stress: adjusted OR, 0.69; 95% CI, 0.50-0.96; P = .03).²⁶ In Egypt, among nursing student interns, students with higher duration of internship and clinical experience (8 months vs 5 months) reported higher levels of psychological distress (OR, 3.78; 95% CI, 1.07-3.22; P = .04), and men reported much less psychological distress compared with women (OR, 0.17; 95% CI, 0.06-0.49; P < .001).²¹ Transmitting the illness to families was one of the concerns, a fear commonly identified in other articles (all adjusted ORs ≥1.21; P < .001).³¹ However, in China, being a graduate student (adjusted OR, 2.03; 95% CI, 1.18-3.49; P = .01), having negative thoughts or actions (adjusted OR, 1.55; 95% CI, 1.38-1.73; P < .001), and feeling depressed (adjusted OR, 6.84; 95% CI, 4.00-11.71; P < .001) were associated with a higher risk of anxiety. Among a sample of medical students in China, probable PTSD prevalence was 2.7% (67 of 2485) and probable depression prevalence was 9.0% (224 of 2485),⁴³ and those in Beijing reported less anxiety than those in the epicenter city of Wuhan (adjusted OR, 0.9; 95% CI, 0.82-1.00; P = .049). In addition, women reported more depressive symptoms than men (adjusted OR, 1.98; 95% CI, 1.19-3.29; P = .009).⁴³

In this systematic scoping review, the findings identified high rates of anxiety and depressive symptoms, with PTSD, general stress, and health-related anxiety in youth and adolescents. Almost all of the included studies were from the COVID-19 pandemic, and there were no studies assessing the effect of the Zika or Ebola outbreaks on adolescent and youth mental health. Due to the global nature of the COVID-19 infection and its high-reaching impact since being declared a global emergency by the WHO in March 2020, it is reasonable to see why there are vastly more studies of the COVID-19 pandemic than of the other outbreaks.

It is likely that the Zika and Ebola studies did not have such an extensive assessment of adolescent mental health compared with the others due to their regional nature. Both Zika and Ebola primarily affected Latin America and Africa. We included 2 H1N1 studies that came from China. There was certainly a paucity of literature evaluating the mental health consequences of these outbreaks in the countries where they were present. Evidence on the impact of loss and disruption on the psychological well-being of adolescents was even more limited. Of the studies included from the H1N1 outbreak, adolescents mostly feared contracting H1N1 and transmitting it to their family members. We found a focus on coping strategies, parental communication, relationships, and an appraisal of panic, health anxiety, hyperactivity, behavioral disorders, and perceived social support (eTable 9 in the Supplement). In the 2 studies,^23,40 psychological distress was not strongly associated with adoption of measures like quarantine, suggesting that these preventative measures may not fully mitigate distress.

Adolescent girls experienced worse mental health outcomes than boys. Sex-sensitive research focusing on addressing the impact of violence, abuse, and maltreatment seems to be critical.⁷⁶ The impact of school closure on high school students is thought to be particularly challenging. These students, closer to their graduation date, may lose more by prolonged school closure or limited access to educational opportunities, leading to academic and social deterioration. Schools are a vital resource for students, especially for the low-income families who depend on these resources for their adolescents’ well-being.⁷⁷ Students who felt more isolated had more associated depressive symptoms.³⁷ A significant economic downturn due to the lockdown could be linked to depression among many, and across countries, families faced more hardships, due to inadequate support and resources from government and organizations.²⁸ Measures used were the PTSD symptom scale, General Anxiety Disorder 7, State and Trait anxiety, Impact of Events Scale, and Patient Health Questionnaire-9. Although these tools cover common mental disorders, the study of social well-being, resilience, general functioning appropriate to age, and quality of life associated outcomes were not reported as frequently. It will be instructive to track multidimensional risk and protective factors in longitudinal and or multinational studies to better understand the complex health and societal impact of the pandemic⁷⁸ on youth.

Research capabilities are an essential component of informing policy and practice, and good research enables the development of evidence-based practice guidelines.⁷⁹ Recent studies have recommended that the clinical and research training-of-trainers model offers a low-cost, scalable strategy to develop technical skills and system-level capacity to carry out evidence-based research.⁸⁰ However, as we found in our review, the Ebola and Zika outbreaks did not receive attention, which may have been attributable to their occurrence in economically constrained and marginalized countries.

School-based mental health programs offered via radios and other digital means or through synchronous interfaces need to include education about coping mechanisms, risk, and protective factor information, and simple self-support strategies.⁸¹ Provision for the support of highly distressed or vulnerable pupils in the school context would add value. In addition, caregiver mental health support information and family interventions would be necessary (eTable 14 in the Supplement). Prioritizing selective and targeted prevention interventions where feasible and future preparedness for increased mental health problems after outbreaks are recommended.⁸²

This review was not without its limitations. The majority of the studies were from China, and hence it was not possible to generalize the findings most studies report on the COVID-19 pandemic, with few studies on other outbreaks. Almost all studies used a cross-sectional design with no longitudinal follow-up, limiting the causality between the exposure and outcomes. With these kinds of studies, we explicitly acknowledge that we cannot prove any type of causality. Moreover, the used sampling method was mostly convenient or snowball sampling with limited generalizability. Future studies should include representative sampling methods to enhance generalizability and confidence in findings.

Results of this systematic scoping review suggest that during the COVID-19 pandemic and H1N1 outbreak, higher risks of developing mental disturbances were associated with being female sex, being young (<18 years) with concurrent psychiatric conditions, having a history of childhood trauma or HIV, being a medical health care worker, living in areas with high COVID-19 infection spread, and having weak family or social support. LMICs are particularly vulnerable due to poor health systems, workforce, and lack of accessible, affordable, youth-responsive mental health services.

Accepted for Publication: August 9, 2022.

Published Online: October 12, 2022. doi:10.1001/jamapsychiatry.2022.3109

Corresponding Author: Manasi Kumar, PhD, Brain and Mind Institute, Aga Khan University, 3rd Parklands Ave, Nairobi, Kenya (manasi.kumar@aku.edu).

Author Contributions: Drs Kumar and Akbarialiabad had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Concept and design: Kumar, Akbarialiabad, Taghrir, R. Shidhaye.

Acquisition, analysis, or interpretation of data: Kumar, Akbarialiabad, Forjoud Kouhanjani, Kiburi, P. Shidhaye, Taghrir.

Drafting of the manuscript: Kumar, Akbarialiabad, Forjoud Kouhanjani, Kiburi, Taghrir.

Critical revision of the manuscript for important intellectual content: Kumar, Akbarialiabad, Kiburi, P. Shidhaye, Taghrir, R. Shidhaye.

Statistical analysis: Kumar, Akbarialiabad, Kiburi, Taghrir.

Obtained funding: Kumar.

Administrative, technical, or material support: Kumar, Akbarialiabad.

Supervision: Kumar, Akbarialiabad, R. Shidhaye.

Conflict of Interest Disclosures: Dr Shidhaye reported receiving grant support from DBT-Wellcome Trust India Alliance Fellowship in Clinical and Public Health Research. No other disclosures were reported.

Additional Contributions: We thank Muhammad Rahman, BSc (University of Washington) for providing assistance in the first draft of this article and Vincent Nyongesa, BSc (University of Nairobi) for help with amending referencing in the final version of the paper. No one was financially compensated for their contribution.