Journal of Child Psychology and Psychiatry **:* (2023), pp **–** doi:10.1111/jcpp.

13864

Annual Research Review: Sex, gender, and

internalizing conditions among adolescents in the
21st century – trends, causes, consequences
Katherine M. Keyes,1 and Jonathan M. Platt2
1
Department of Epidemiology, Mailman School of Public Health, Columbia University, New York, NY, USA; 2College of
Public Health, University of Iowa, Iowa City, IA, USA

Internalizing conditions of psychopathology include depressive and anxiety disorders; they most often onset in
adolescence, are relatively common, and contribute to significant population morbidity and mortality. In this
research review, we present the evidence that internalizing conditions, including depression and anxiety, as well as
psychological distress, suicidal thoughts and self-harm, and fatal suicide, are considerably increasing in adolescent
populations across many countries. Evidence indicates that increases are currently greatest in female adolescents.
We present an epidemiological framework for evaluating the causes of these increases, and synthesize research on
whether several established risk factors (e.g., age of pubertal transition and stressful life events) and novel risk
factors (e.g., digital technology and social media) meet conditions necessary to be plausible causes of increases in
adolescent internalizing conditions. We conclude that there are a multitude of potential causes of increases in
adolescent internalizing conditions, outline evidence gaps including the lack of research on nonbinary and gender
nonconforming populations, and recommend necessary prevention and intervention foci from a clinical and public
health perspective. Keywords: Adolescence; internalizing; depression; anxiety; gender difference.

et al., 2019). As such, a diverse set of mechanisms

Introduction
have been proposed as causes of gender differences
Internalizing conditions are mental health symptoms
in internalizing conditions.
that individuals express inwardly, such as anxiety,
In recent years, significant increases in depres-
sadness, worthlessness, and withdrawal. At the
sion, psychological distress, anxiety, and suicidal
threshold of clinical impairment, internalizing symp-
behaviors among adolescents have been reported.
toms can lead to anxiety disorders and mood
However, findings vary by country and sample
disorders such as depression. Anxiety is one of the
composition, measures, time period, and other
most common disorders in young people, affecting
factors (Herrman et al., 2022). In addition to
4%–20% of children and adolescents, while depres-
documenting emerging trends, explanations of why
sion affects 2%–8% of children and adolescents
gender differences in internalizing conditions may be
(Abbafati et al., 2020) and is a leading cause of
changing have proliferated, mostly speculatively.
disability globally (WHO, 2021a, 2021b). Related
Understanding how gender differences in internaliz-
constructs such as psychological distress also
ing conditions have been changing over time is
manifest as internalizing symptoms. Internalizing
necessary to identify emerging high-risk groups
problems often begin early in childhood and can
and key drivers of mental health challenges and
negatively impact an individual’s relationships,
group disparities.
development, and daily functioning (Ghandour
The aims of this study are to (a) review and
et al., 2019), and increase the risk of suicide and
summarize the current literature on gender differ-
substance use disorders (Berny & Tanner-
ences in internalizing conditions and suicidal behav-
Smith, 2022). As a result, internalizing disorders
iors to identify changes over time and any potential
carry a significant cost for health and social well-
factors to explain inconsistent findings across stud-
being.
ies; (b) use an epidemiological framework to evaluate
Across clinical and community settings and mea-
the evidence for several proposed causes of temporal
sured with various instruments, girls and women are
changes in gender differences in internalizing con-
more likely than boys and men to experience the
ditions and suicidal behaviors; and (c) propose
range of internalizing conditions and symptoms,
future research and clinical priority areas. In
including depression, psychological distress (Bor,
addressing these aims, we build on previous related
Dean, Najman, & Hayatbakhsh, 2014; Salk, Hyde, &
research (e.g., Sweeting, West, Young, & Der, 2010)
Abramson, 2017), anxiety (Asher, Asnaani, &
by providing an epidemiological framework through
Aderka, 2017; Sevar, Vythilingum, & Castle, 2015),
which any putative cause of interest may be evalu-
and suicidal ideation (Miranda-Mendizabal
ated as a plausible causal explanation for changing
trends. We emphasize that this review is not
Conflict of interest statement: No conflicts declared. intended to be exhaustive; however, we hope that

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
2 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

future investigations will be strengthened by the use 2018 across 36 countries in Europe, and North
of this framework in considering additional causes. America (A. Cosma et al., 2020). The authors utilized
Finally, we note that more expansive measures of data from the World Health Organization’s Health
gender expression than binary classifications is a Behaviour in School-aged Children (HBSC) survey,
critical lens for mental health differences (Abramo- which is a repeated cross-sectional study of 11–
vich & Cleverley, 2018). However, the vast majority 15 year olds, completed every four years with the
of research on gender differences in internalizing same study measures and design, designed to
conditions to date has relied on binary categories. As estimate time trends. After accounting for cross-
a result, we focus on studies that have defined national differences, the study authors reported that
gender using a binary male/female classification. In psychosomatic complaints increased overall, but at
addition, health differences are often discussed a significantly greater rate for girls than boys.
without differentiation between sex assigned at birth It should be noted that trend data, for both
and gender expression. In this review, we will utilize internalizing symptoms and depression and anxiety,
the term ‘gender’ differences generally for consis- are based on self-report; the extent to which changes
tency, acknowledging that studies likely involve over time in willingness to report mental health
misclassification when respondents are only pro- symptoms may inflate true prevalence remains
vided with binary options, and erasure of nonbinary unknown (Foulkes & Andrews, 2023).
gender expressions. When discussing the contribu-
tion of biological or genetic factors, we will use sex
Trends in depression
differences.
Depression risk has consistently been higher among
girls and women, and recent evidence suggests that
gender differences may be increasing in recent years.
Recent trends in gender differences in
One review of these trends estimated rates among
internalizing conditions and suicidal behaviors
adolescent girls and boys in the United States (ages
among adolescents
10–19), from 1989 to 2017 (Platt, Bates, Jager,
Trends in internalizing symptoms
McLaughlin, & Keyes, 2021). Twenty-seven diagnos-
In this section, we summarize the recent research to tic and 33 symptom-level depression studies were
estimate temporal variation in gender differences in identified, mostly cross-sectional or repeated cross-
symptoms and diagnoses of internalizing disorders. sectional design, from community-based samples.
Throughout, we note how trends were measured Based on meta-regression methods to examine
(e.g., diagnostic scales and symptom checklists), as changes in model estimates of the association of
trends have been shown to differ according to study gender with depression, depression differences
design-based characteristics (Baxter et al., 2014). increased overall for both symptom and diagnostic
We focused primarily on systematic reviews and measures. Findings were robust to the period of
meta-analyses of community-based samples pub- symptom recall and measure of depression. Much
lished since 2010, and those that report changes in like the studies of internalizing symptoms, this study
the magnitude of gender differences across the year found that the gap between girls and boys in
of data collection. Community-based samples were depression has been increasing in recent decades.
prioritized to minimize variation due to selection into Figure 1 illustrates this growing gap in adolescent
clinical or other specific settings. The majority of depression plotting model estimates of the magni-
studies focused on high-income countries in the tude of gender differences in diagnostic depression
global North. over the year the study was completed. Study
A review of gender differences in internalizing estimates for other groups throughout adulthood
symptoms included 19 epidemiological studies from and also included, to contrast these trends. In all
12 countries from 1983 to 2011 (Bor et al., 2014). other age groups, depression differences have nar-
While these data are now up to 40 years old, the rowed over the same time period.
large sample and wide time interval make this study
an important review of internalizing symptom
Trends in anxiety
trends. Included studies were from middle- and
high-income countries (e.g., the United Kingdom, Anxiety disorders comprise conditions characterized
Finland, the United States, Australia, and China). by excessive fear or apprehension about real or
The authors reported increasing symptoms among perceived threats, including phobias, panic, and
adolescents ages 11–18. Increases were greater generalized anxiety disorders (Klein, 1980). Anxiety
among adolescent girls than boys, found in five of disorders are more common in girls and women
eight studies in girls, and three studies in boys. (Ruscio et al., 2017); however, to date, there is
Patterns did not systematically vary by country, limited research assessing changes in adolescent
sample, and method of outcome assessment. In a anxiety symptoms and disorders over time. Between
more recent study, Cosma et al estimated changes in 2016 and 2020, there is evidence that anxiety
adolescent psychosomatic complaints from 2002 to diagnoses increased among children (ages birth to

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 3

1.5

1.0 Age
10−19
log OR

20−39
40−59
60+

0.5

0.0

1980 1985 1990 1995 2000 2005 2010 2015 2020

Study Year

Figure 1 Meta-regression estimates of the depression gap between adolescent girls and boys over time. Source: Platt et al. (2021)

17) in the United States, with one study reporting (e.g., suicidal ideation and intentional self-injury)
prevalence change from 7.1% to 9.2%(Lebrun- are a significant cause of morbidity. Suicidal behav-
Harris, Ghandour, Kogan, & Warren, 2022). Among iors are highly comorbid with internalizing condi-
young adults ages 18–25, rates are increasing tions and also exhibit consistent gender patterning.
consistently across gender (Goodwin, Weinberger, While girls have higher rates of suicidal thoughts
Kim, Wu, & Galea, 2020). and attempts, higher rates of completed suicide are
While there is evidence of increasing anxiety in the consistently observed in boys across many countries
United States, studies of gender differences in time (Abraham & Sher, 2019). There is also evidence that
trends are limited. One recent study examined suicide rates have been changing in recent years,
changes in gender differences in anxiety from 2012 though trends vary across countries. In the United
to 2018 in Wisconsin among adolescents ages 14–18 States, there is evidence that rates of suicide death
(Parodi et al., 2022). A validated symptom screening among adolescents (ages 15–19) has increased
instrument was used to determine rates of anxiety among boys and girls to their highest levels since
that were indicative of probable disorder. Among 2000 (Miron, Yu, Wilf-Miron, & Kohane, 2019). In
girls, anxiety increased from 42% to 56% over the other high-income countries, suicide rates are
study period, while anxiety increased from 26% to decreasing for boys and girls on average. Using
31% among boys. Most recently, a meta-analysis WHO and OECD data of 29 high-income countries
reviewed studies that measured adolescent anxiety from 1995 to 2014, Roh et al compared suicide rates
during the first six months (February–July 2020) of from 1995 to 2005 with rates from 2006 to 2014
the COVID-19 pandemic (Racine et al., 2021). stratified by gender (Roh, Jung, & Hong, 2018).
Twenty-five studies were identified, from countries Among those ages 15–19, boys were consistently
in North and South America, Europe, and China. more likely to die by suicide than girls in both time
Over that period, the authors reported an increase in periods, and the average suicide rate decreased
anxiety symptoms that was significantly higher for across all countries from 11.02 to 9.38 among boys
adolescent girls than boys. While this study captured and 3.72 to 3.69 among girls. However, the gender
a relatively narrow and unique period in time, it gap in suicide decreased in 23 of 29 countries, either
signals a need to monitor youth anxiety in the longer because rates among girls decreased less or
term. Overall, more evidence is needed, but results increased more than rates among boys.
suggest that gender differences in clinically mean- Suicidal ideation and suicide attempts are highly
ingful anxiety symptoms are widening due to greater prevalent and are among the strongest predictors of
increases in girls over time. completed suicide. In the United States, there is
consistent evidence that suicidal behavior is increas-
ing among adolescents, with psychiatric emergency
Trends in suicidal behavior
department visits increasing 28% from 2011 to 2015
Suicide is a leading cause of death among adoles- (Kalb et al., 2019), and a doubling of pediatric
cents worldwide, and nonfatal suicidal behaviors emergency visits for intentional-self harm from

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
4 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

2007 to 2015 (Moylan et al., 2019). US survey data (WHO, 2021a, 2021b). Taken together, the findings
indicate that by 2021, nearly 1 in 4 adolescents girls emphasize the global scope of adolescent mental
seriously considered suicide in the past year, a health problems, and the need for more attention to
significant increase from previous yearly trends identifying high-risk groups and determinants of
and a greater increase among girls compared with mental health disparities across a range of social,
boys (CDC, 2023a, 2023b). Cross-national data cultural, and health-related contexts.
suggest global variation. A pooled analysis of data
from 2003 to 2017 across 90 countries estimated
variation in suicidal ideation and attempts among Interrogating causes of the increase in
adolescent girls and boys ages 13–17 (Campisi internalizing conditions: An epidemiological
et al., 2020). Suicidal ideation prevalence was framework
significantly greater among girls than boys (range: Identifying causes of symptom increases is critical to
0.8%–36.2% among girls and 0.7%–37.1% among develop and disseminate prevention and interven-
boys). By contrast, there were no gender differences tion initiatives. Causes of internalizing symptoms
in the rate of suicide attempts, which ranged from among youth are multilevel and multifactorial; they
3.6% to 53.7% among girls and 4.2% to 67.2% are distributed across societal and community
among boys. Trends likely varied widely across causes, family and school environments, comorbid-
regions and countries, but overall results suggested ity and genetic factors. Numerous theoretical frame-
that cross-national patterns are not changing appre- works have been developed to illustrate the
ciably over time. multileveled and lifecourse unfolding of mental
health risk (Bronfenbrenner, 1994; Krieger, 2001;
Lynch & Smith, 2005; Rutter & Sroufe, 2000), all of
Limited data to estimate trends in LMICs
which highlight the need to consider a broad array of
Research on gender and mental health in low- and both environmental and biological interplay. Yet,
middle-income countries (LMICs) is limited and across theoretical frameworks in selecting potential
generally insufficient for an examination of time causes to interrogate for inquiry, an epidemiological
trends. A recent study of cross-national gender framework can be useful in cross-cutting the theo-
differences in mental health in 2018 found nearly retical frameworks themselves. By epidemiological
ubiquitous gender differences in internalizing con- framework, we follow the tradition of our field in
ditions (e.g., psychological distress, life satisfaction, outlining and explicating the level and type of
eudaemonia, and hedonia), across 73 countries evidence (Bradford, Cbe, Frcp, & Frs, 1965; Hern
an
including LMICs in Central/South America, Central & Robins, 2010; Morris, 1975), triangulated across
Asia, and Southeast Asia, but no African countries sources (Lawlor, Tilling, & Smith, 2016), that are
(Campbell, Bann, & Patalay, 2021), with variation necessary to come to causal conclusions. Such
across countries and conditions. A systematic nar- frameworks have been developed and elaborated in
rative review of 37 studies of adolescent mental our field for the last century. For this study, we
health problems in 16 Sub-Saharan African nations identify three conditions that are necessary but not
found generally consistent evidence of greater levels sufficient for a hypothesized cause to have a
of anxiety, depression, and suicidality among girls sufficient evidence base to explain increases in
than boys (Jorns-Presentati et al., 2021). Another internalizing conditions that are concentrated in
review of global gender differences in adolescent girls.
depression identified 72 study estimates from data First, and perhaps most fundamentally, the
collected between 2001 and 2020 (Shorey, Ng, & hypothesized causes need to be associated with the
Wong, 2022). The prevalence of elevated depressive outcomes of interest, including internalizing condi-
symptoms increased 14% over time, rising from 24% tions, distress, and/or suicidal behavior. Indeed, if
in 2001–2010 to 38% in 2011–2020. The prevalence there is no signal of an association, then potential
of elevated depressive symptoms among was high risk factors cannot explain recent trends in adoles-
among female adolescents (32% vs. 24% among cent internalizing conditions. This framing does not
boys), and risks were greatest for female adolescents, specify that potential risk factors need to be either
particularly from Middle Eastern, African, and Asian necessary or sufficient; internalizing conditions have
nations. This review did not assess gender differ- multifactorial causes that may be associated or
ences in variation over time, however. interact with each other (Rothman, 1976;
Understanding the global scope of suicide is VanderWeele & Richardson, 2012). Evidence is
similarly limited by incomplete data. Many countries stronger still if available studies sufficiently address
do not have reliable vital statistics data systems, and issues such as measurement error (Ogburn &
thus do not fully capture or misclassify deaths due to VanderWeele, 2012), confounding (Greenland &
suicide. An estimated 77% of global suicides occur in Robins, 1986), selection bias (Munaf o, Tilling, Tay-
low- and middle-income countries in 2019, lor, Evans, & Smith, 2018), as well as aligned
highlighting the urgent need for public health concepts such as positivity and consistency (Hernan
surveillance in countries where data are most limited & Robins, 2010; VanderWeele, 2009). In psychiatric

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 5

epidemiology, which largely relies on observational been used in epidemiology, for example, to describe
studies rather than experimental, causal inference how to evaluate racial disparities in health (Ward
remains a challenge, and triangulation of evidence et al., 2019).
across multiple studies conducted in different popu- We have summarized the criteria that we will use
lations, with different measurement techniques and to evaluate several potential causes of the increase in
causal identification strategies, improves our ability internalizing conditions, and the increased concen-
to draw conclusions about whether associations are tration among female adolescents, in Figure 2 as a
sufficiently strong and robust (Lawlor et al., 2016). guide for evaluating the literature in this review. In
Furthermore, because the evidence indicates that Parts III and IV, we evaluate the strength of evidence
internalizing trends have impacted female more than to satisfy the criteria important driver of current
male adolescents, evidence is stronger still for a trends.
putative risk factor to explain observed trends if
there is an interaction by gender, with a stronger
association for females compared with male Part III: Changing prevalence of several known
adolescents. risk factors: the evidence
Second, we need to not only identify factors that In this section, we demonstrate how the epidemio-
cause internalizing conditions, but also identify logical framework can be used to evaluate the
factors that cause trends in internalizing conditions, strength of evidence for selected risk factors that
especially those that differ by gender. The distinction have a long history and evidence base in psychiatric
between individual-level associations and causes of epidemiology as causes of internalizing symptoms.
incidence has a long history in epidemiology, as We did not conduct a systematic review of all
classically exemplified in Geoffrey Rose’s seminal potential risk factors, and those selected here are
paper “Sick individuals and sick populations” not exhaustive of all causes of internalizing symp-
(Rose, 2001). Rose outlined that multiple popula- toms. Indeed, others may have chosen different risk
tions (or in our case, multiple time periods) may have factors from all levels of organization across societal
very difference incidence and prevalence of a health changes across the last century to molecular path-
outcome, while within these populations, the deter- ways, or focused on other components of the risk
minants of why some individuals are sick and other factors that we chose. Our goal was to demonstrate
are not may be similar. In other words, the distribu- how the epidemiological framework presented above
tion of risk factors determine why one population could be a guide for researchers aiming to evaluate
has greater burden of sickness than another popu- evidence across the field.
lation. For example, it is clear that depression has
genomic underpinnings (Sullivan, Neale, & Kend-
Stressful life events
ler, 2000). However, genetic risk factors do not
explain why adolescents today have more depression In 1967, Holmes and Rahe developed a measure of
than adolescents 10 years ago, because germline 43 stressful life events (SLEs), which they defined as
and other genomic mutations largely do not change ‘occurrences that were likely to bring about
at a population level over that time frame. Risk readjustment-requiring changes in people’s usual
factors that explain increased prevalence over time activities’ (Holmes & Rahe, 1967); these events were
must themselves either be changing in prevalence, or reliably (although with substantial measurement
changing in virulence or magnitude of effect, critique) (Dohrenwend, 2006; Rabkin & Struen-
over time. ing, 1976) associated with the onset, severity, and
Finally, because the evidence indicates that trends persistence of internalizing disorders (Holmes &
in adolescent internalizing conditions are stronger Rahe, 1967), providing the scaffolding for an epide-
among girls compared with boys, potential risk miologic approach to stress and mental health
factors that are increasing over time should be either research. Since then, studies of SLEs and health
increasing more among girls, or alternatively, the have grown exponentially (Dohrenwend, 2006). The
strength of the association over time should be risks from SLEs on internalizing conditions has been
increasing among girls compared with boys. For well-established (March-Llanes, Marqu
es-Feixa,
example, risk factors in this category could poten- Mezquita, Fa~ nan
as, & Moya-Higueras, 2017). How-
tially contribute to increased incidence of anxiety if ever, there is less consensus about the role of SLEs
the relationship between gender and anxiety inci- in gender differences in internalizing conditions.
dence is growing over time, even if the prevalence of This is partially due to the substantial variation in
the hypothesized risk factor in the population is not. the definition of SLE and the number and type of
Thus, a third criterion for a hypothesized risk factor SLEs included in individual studies (Dohren-
to explain trends in internalizing conditions in the wend, 2006). Important stressors are historically
population is for that risk factor prevalence (or contingent and instruments are sometimes updated
strength of effect) to vary by gender. This framing of to include novel stressors (e.g., cyber-bullying).
hypotheses in terms of prevalence and interaction as Some studies consider only interpersonal experi-
explanations for demographic differences has long ences, while others consider experiences at more

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
6 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

Figure 2 Criteria for a hypothesized risk factor to be a cause of the gendered increase in adolescent depression

macro levels. In addition, some studies in this (Rutter, 1981). While boys’ responses may be
literature also consider traumas and adversities, protective against internalizing psychopathology, it
without explicitly differentiating these terms from has also been shown to increase the risk of
SLEs. When attempting to summarize gender pat- externalizing and substance use problems (Wills &
terning across these studies, accounting for a Filer, 1996).
comprehensive list of SLEs becomes a missing data
problem. Is the prevalence increasing over the relevant time
While evidence for gender patterning in SLEs is period?. Understanding whether the prevalence of
inconsistent in terms of the number and severity of SLEs is increasing over time is complicated by the
events (Bebbington, 1996), specific stressors are widespread variability in the particular items cap-
patterned by gender. Girls are more likely to tured within any SLE inventory. Perhaps for that
experience stressors such as sexual harassment reason, research to understand whether SLEs are
and abuse, are victims of partner violence more increasing over time is very limited. One recent
frequently (Chan, 2011), and may be more suscep- review focused on trends in childhood adversities
tible to the effects of interpersonal stressors or to in the United States (Finkelhor, 2020), defined as
stress experienced among family relationships major stressors or life events that overwhelm chil-
(Kessler & McLeod, 1984). Conversely, boys are more dren’s normal capacity for coping. Time trends
likely to directly experience physical violence, per- differed by type of adversity, with the prevalence of
sonal accidents, and other injuries, as well as to some events increasing and others decreasing.
witness the death or injury of someone else (Tolin & Declines were found in rates of: parental death/
Foa, 2006). illness, sibling death, family poverty, exposure to
Stressors also account for more variance in female domestic violence, parental divorce, serious child-
(vs. male) internalizing conditions scores across the hood illness, physical abuse, sexual abuse, physical
lifecourse. This finding captures both the differences and emotional bullying, and exposure to community
in exposure to SLEs, but also the stress responses violence. Increases were found in rates of: parental
(Nolen-Hoeksema, Larson, & Grayson, 1999) and drug and alcohol abuse and parental incarceration.
coping. Girls are more likely to report fear, irritabil- Overall, the authors conclude that there appear to
ity, confusion, rumination, and other negative have been more historical and recent improvements
thoughts in reaction to a stressful event (Risman & in ACEs than deteriorations.
Davis, 2013), which may increase their risk of Other efforts to understand trends in stressful
internalizing problems, whereas boys are often events have relied on more general measures of
socialized to cope with negative thoughts in antiso- stress or negative experiences. One ongoing Gallup
cial, impulsive, and other externalizing ways study (Gallup, 2022) tracks positive and negative

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 7

experiences across the globe, as an index of self- that students who experience cyber-bullying are also
reported experiences of physical pain, worry, sad- more likely experience face-to-face bullying and vice
ness, stress, and anger. Overall, negative experi- versa (Kreski, Chen, Olfson, Cerd
a, Martins,
ences have been steadily increasing since 2014 and et al., 2022; Odgers & Jensen, 2020a); however,
are at the highest level since the study began in even small increases in both types of bullying
2006. Trends were driven most by high levels of experienced by girls at young ages suggests that
worry (42%) and stress (41%), although such mea- bullying may be an important driver of growing
sures may be capturing the underlying construct of gender differences in internalizing conditions.
internalizing symptoms, the may be indicative of
increases in internalizing problems rather than an
Gender inequality in education
explanation for them.
The impact of education on mental health is well-
Is the association or prevalence change differential established (Children, Youth, Families and Socio-
by gender?. To our knowledge, a comprehensive economic Status, 2010). Educational attainment
investigation of gender differences in SLE trends over increases skills, self-efficacy, and future socioeco-
time has not been conducted. Acknowledging this nomic opportunities (DiPrete & Buchmann, 2006),
limitation, we focused on two events more common and broadens social networks, all of which may act
among girls and women, sexual abuse and bullying. to decrease the depression gap through social stress
There are well-established gendered patterns in pathways (Ross & Mirowsky, 2006). Traditionally,
childhood sexual abuse (Barth, Bermetz, Heim, girls and women have had lower education than men
Trelle, & Tonia, 2013). Youth trends also appear to (Connell, 1987).
be decreasing overall (Finkelhor, Saito, &
Jones, 2018), though global gender differences have Is the prevalence increasing over the relevant time
not been systematically studied. One study using US period?. Overall, global patterns have trended
population representative samples from 2003 to toward greater gender equity in SES throughout
2008 found similar decreases overall, with no trend the 20th and 21st Century. Gender gaps in educa-
differences by gender (Finkelhor, Turner, Ormrod, & tional attainment narrowed and disappeared in
Hamby, 2010). While this limited evidence suggests recent generations. Globally, girls are now as likely
that trends in sexual abuse are not a likely driver of to finish primary and secondary school as boys, and
internalizing patterns, more research could more likely to receive postsecondary education
strengthen this conclusion. Among US adolescent (Global Gender Gap Report, 2019). While trends
girls, evidence is emerging of increases in reports of vary across regions, overall, the prevalence of gender
sexual violence (CDC, 2023a, 2023b), however, gaps in SES has narrowed over a relevant time
suggesting that continued surveillance is necessary. period.
Gendered trends in face-to-face and cyber-bullying
have also been evaluated. An analysis of adolescent Is the association stronger for girls
(ages 11, 13, and 15) bullying pooled data from 22 than boys?. Based on general trends, the decrease
countries to examine gender ratios of bullying in education disparities by gender would be expected
victimization (Smith, L
opez-Castro, Robinson, & to decrease girls’ risk of internalizing conditions.
G€orzig, 2019). From 1993 to 2014, boys were However, in addition to the many benefits of
consistently more likely to experience bullying than education for health and well-being, these trends
girls at all ages. However, girls were more likely to may have unintended negative mental health conse-
report experiencing online bullying. However, quences. As education levels rise, and populations
another large cross-national study found that gender shift to knowledge-based economies, education
differences in bullying perpetration and victimization becomes more influential for future mobility, and
differed by national level of gender inequality (Cosma more stressful in turn (H€ ogberg, 2021). Women are
et al., 2022). A meta-analysis of US studies (Ken- relatively more dependent on education for future
nedy, 2019) estimated changes in the prevalence of employment opportunities (Dougherty, 2005), put-
bullying involvement from 1998 to 2017. Authors ting them at higher risk. This risk appears to be
reported decreases in face-to-face bullying among positively related with community norms for educa-
youth overall; however, among girls, there was a tional and economic achievement (Luthar, Kumar, &
small but statistically significant increase in bullying Zillmer, 2020). In a recent review of 11 qualitative
prevalence, especially for those in eight grade or studies in the United States and Europe, Stentiford
below. Trends were similar but larger in magnitude et al. identified four achievement-related themes that
for studies focusing on cyber-bullying. Cyber- may be related to adolescent mental health: fears for
bullying is a relatively novel exposure and its effects the future, family-related pressure, competitive
on mental health is an area of active study. Recent school culture, and gendered expectations around
research has suggested that cyber-bullying alone is school performance. An imbalance of any of these
less impactful than face-to-face bullying (Hase, experiences increased girls’ risk of mental health
Goldberg, Smith, Stuck, & Campain, 2015), and problems. Another large study (Cosma et al., 2020)

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
8 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

of European and North American High School physical development can outpace slower progres-
students was able to quantitatively estimate the sing executive cognitive systems and create discor-
effect of school-related pressures on mental health. dance between an adolescent’s perceived age and
They reported that self-reported schoolwork pres- their level of cognitive development (Ge & Nat-
sure in partially accounted for increasing levels of suaki, 2009). This may increase exposure to new
psychosomatic health complaints, especially in ado- social norms, environments, and expectations from
lescent girls, suggesting that this aspect of changing peers and adults regarding their maturity. The
SES may be an important part of increasing inter- consequences of these mismatched experiences
nalizing conditions among adolescent girls. Overall, and cognitive abilities may increase levels of psy-
these studies suggests that the changing gender chosocial stress (Ge, Conger, & Elder, 1996) and
patterning in education may have a complex rela- trauma (Vicary, Klingaman, & Harkness, 1995), and
tionship with mental health. However, whether other mental health problems (Ge & Natsuaki, 2009;
greater gender equity in education is responsible Stice, Presnell, & Bearman, 2001). This so-called
for increasing or decreasing mental health problems ‘developmental mismatch’ is hypothesized to be
in adolescents (and whether those patterns are further exacerbated by exposure to novel stimuli
consistent throughout the world) remains equivocal. (NPR, 2022), though more data to test hypotheses
Regardless of overall effects, it is important to ensure are needed.
supportive schooling expectations and systems for
girls and boys, in order to maximize the positive Is the prevalence increasing over the relevant time
aspects of education in adolescence and beyond. period?. While the literature on the role of hor-
mones in gender differences in internalizing condi-
tions remains equivocal, they are an important
Puberty and hormones
component of the broader process of puberty, which
The emergence of gender differences in internalizing coincides with a substantial increase in the rates of
conditions is approximately correlated with puberty internalizing conditions in girls (Ullsperger & Niko-
(Kessler et al., 2010), which has led many to study las, 2017). Trends in the age of puberty onset in
hormonal mechanisms as a cause. Some studies recent decades suggest modest deceases, although
have found that pubertal stage is a better predictor of much of the evidence is from nonrepresentative and
internalizing condition risk than calendar age among small samples, with imperfect measures of puberty
adolescent girls (Angold, Costello, & Worth- status (Sørensen et al., 2012). Compared with other
man, 1998). As a risk factor for internalizing historic trends in age of puberty onset, these recent
conditions, researchers have tested the impact of trends point to a distinct etiology, including obesity
female sex hormones through both biological and (Ong, Ahmed, & Dunger, 2006), fetal nutrition,
social mechanisms. Ovarian hormones, especially childhood dietary habits, physical activity, and
estrogens, testosterone, and progesterone, influence exposure to endocrine disrupting chemicals (Parent
neurotransmitters such as dopamine and serotonin et al., 2003).
production, which influence the quality of sleep,
concentration, appetite and mood (Rubinow & Is the prevalence changing more for girls
Schmidt, 1996). Other research has implicated these than boys?. In examining the impact of pubertal
hormones in biological differences in stress response development, the most extensive evidence has
(Young & Korszun, 2009). Still other studies have focused on pubertal timing (Graber, 2013). The
interrogated the effects of individual hormones evidence for decreasing puberty age is strongest for
(Nottelmann, Inoff-Germain, Susman, & Chrou- girls, although there is limited evidence of similar
sos, 1990) or the relative composition of multiple trends in boys. Since the 1970s, the average age of
hormones (Susman et al., 1985). Overall, while thelarche (i.e., Tanner stage 2 of glandular breast
hormones may partially explain interindividual risk, development) has decreased by 0.24 years per
they do not appear to be a direct cause of the decade, to around 10 years old (Eckert-Lind
between-group differences in internalizing problems. et al., 2020). The age at menarche (i.e., onset of
The association between puberty onset and mental menstruation) remained generally stable during the
health, then, has been hypothesized to be primarily latter half of the 20th Century (after substantially
due to the psychological stress of changes in body declining from age 17 to 13 in the prior century). In
morphology, including secondary sex characteris- the 21st Century, declines of 2.5–4 months per
tics, and also increased appearance of acne, weight decade have been reported. Among boys, the most
gain, and voice changes (Graber, 2013). These valid method of measuring puberty onset is via
increased risks are especially high when individuals orchidometry to measure testicular volume. This
experience puberty earlier than their peers, for method requires testicular palpation, which has
example, leading to perceptions that an adolescent seldom been performed in population-based studies
girl is older (physically or emotionally) than her (Sørensen et al., 2012). One of the biggest studies to
chronological age (Castellanos-Ryan, Parent, Vitaro, perform orchidometry and visual examination was
Tremblay, & S
eguin, 2013). In early puberty, The Copenhagen Puberty Study, which measured

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 9

puberty onset annually from 1991 to 2008 via visual emotions, including anorexia nervosa, binge eating,
genital staging and orchidometry, and reported and bulimia nervosa. Substance use includes the
documented a decline of 3 months in age at onset consumption of alcohol, cannabis, and other illicit
of puberty (Sørensen, Aksglaede, Petersen, & drugs, which at any level is hazardous for adoles-
Juul, 2010). An alternative marker of puberty onset cents and can have long-term health consequences.
is peak height velocity (PHV) (i.e., timing of the Both eating disorders and substance use also exhibit
pubertal growth spurt), which is a more generaliz- strong gender patterning. Therefore, understanding
able albeit less specific benchmark. A Swedish study trends in substance use and eating disorders may
of cohort effects in boys’ PHV reported small but uncover important mechanisms of growing internal-
statistically significant decreasing trends from 1947 izing disparities.
to 1991 (Ohlsson et al., 2019). Overall, more
research with valid puberty measures in represen- Is the prevalence increasing over the relevant time
tative samples are needed to confirm patterns in the period?. Eating disorders typically emerge in ado-
progression of puberty among boys and girls. lescence alongside other internalizing conditions,
and limited evidence suggests temporal variation in
the 2010s. Research to estimate the prevalence of
Genetic factors
eating disorders is limited by nonrepresentative
One of the strongest and most consistent risk factors populations, small sample sizes, and nonstandard
for internalizing conditions is a family history, and questionnaires, making it difficult to understand
genetic factors underlying depression and internal- time trends. This concern is further heightened by
izing symptoms continue to be identified (Levey the global COVID-19 pandemic, which led to a
et al., 2021; Wray et al., 2018). Putative genetic reported considerable increase in treatment for
mechanisms may influence serotonin balance and eating disorders (Hartman-Munick et al., 2022; Lin
inflammatory processes (Cosma et al., 2021), affect- et al., 2021).
ing sleep quality, appetite, and mood downstream.
Results from studies of candidate genes, genome- Is the prevalence changing more for girls
wide association studies, and gene–environment than boys?. A complete understanding of the
interactions have offered inconclusive explanations epidemiology of eating disorders is further limited
for gender differences (Zhao et al., 2020). Several by the fact that research is underdeveloped for
large systematic analyses of genome-wide associa- assessing eating disorders in boys (Murray
tion studies for depression found no SNP achieved et al., 2017). This lack of attention to eating
genome-wide significance in analysis of all samples disorders in boys is best illustrated by the inclusion
or in sex-stratified analyses (Neumann et al., 2022; of amenorrhea as a diagnostic criterion for eating
Sullivan et al., 2013). Overall, heritability estimates disorders until the DSM-5 (in 2018). Boys’ risk is
differ by study sample and design characteristics now a recognized concern and research is ongoing to
(e.g., age ranges and measurement methods), which characterize gender differences in symptoms and
limits the comparability between studies and the prevalence of eating disorders. A recent review found
validity of summary estimates (Patterson that anorexia nervosa may be increasing in girls, and
et al., 2018). rates of probable eating disorders among boys also
appear to be increasing. The authors also
Is the prevalence increasing over the relevant time highlighted the growing public health concern
period?. Additionally, the temporal variation in around eating disorders worldwide for women,
gender differences has been occurring on a timescale men, and those who identify as having non-binary
that is too short to be compatible with genetic gender identities (Sil

en & Keski-Rahkonen, 2022).
explanations. To the extent that there is a genetic For now, the evidence is inconclusive to link eating
component to gender differences in internalizing disorder trends with internalizing differences
conditions, it is likely part of a complex relationship between boys and girls (Sil
en & Keski-
between multiple polymorphisms and social–envi- Rahkonen, 2022).
ronmental factors. Overall, there is little support for Converse to internalizing disorders, substance use
genetic mechanisms as primary causes of temporal is historically more prevalent in boys than girls,
and gender variation in internalizing conditions. although there is wide cross-national variation in the
magnitude of gender differences (Seedat et al., 2009).
In recent years, significant shifts in these gender
Comorbid eating and substance use disorders
differences across many countries have resulted in
Internalizing disorders are highly comorbid with the convergences of gender differences in alcohol and
other mental and behavioral health conditions, cannabis disorders (Chapman et al., 2017; Slade
including eating disorders and substance use disor- et al., 2016), and in the United States, tobacco use
ders. Eating disorders are conditions characterized (vaping in particular) and illicit drug use (Brener
by severe and persistent disturbance in eating et al., 2022; Miech et al., 2019.). While gender
behaviors and associated distressing thoughts and differences have narrowed over time, they have done

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
10 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

so through decreasing rates for both boys and girls, is a paucity of evidence in Africa and Asia. However,
which if it were a direct cause of internalizing data are not consistent across studies (Golberstein,
conditions, would likely predict the opposite trends. Gonzales, & Meara, 2019), and no evidence suggests
that the impacts of global economic activity are
worse for girls compared with boys. Furthermore,
Macro-economic conditions
economic conditions impact adolescent mental
Throughout the course of the 21st century, there has health through both material and relative depriva-
been an increased focus on widening economic tion pathways; indeed, there is a robust evidence
inequality globally, with a shrinking middle class in base for ‘double disadvantage’, with children who
high-income countries, changes in global labor experience poverty at higher risk of mental health
markets, knowledge economies and competition for problems especially in school and community con-
jobs (Hope & Martelli, 2019), increases in the texts with higher income inequality (Odgers, 2015),
financial impact of debt, and other economic condi- and that adolescent mental health improves when
tions that are associated with worse mental health income inequality eases (Kim, Kim, & Chung, 2020).
within and across populations. Indeed, among
adults, macro-economic factors are hypothesized to Is the prevalence increasing over the relevant time
underlie recent increases in suicide and other injury period?. Recent shocks to the global market are
deaths (Case & Deaton, 2015). The impact of these particularly immediately preceded the time period of
global changes in the ability of families to maintain increasing adolescent internalizing conditions, sig-
financial stability, grow wealth, and move out of poor naling it as a potential contributor. Increasing trends
and middle-class market positions on adolescent occurred following the decline of global market and
mental health is understudied, but remains a economic activity during the Great Recession. As
plausible additional mechanism. such, a plausible hypotheses for a cause of these
trends is the macro-economic effects of global
Are macro-economic conditions associated with recession. In the United States, unemployment
adolescent mental health?. Children and adoles- during the Great Recession increased from 5.3
cents in poverty, defined variously across the percentage points in the starting 2007 to a height
literature by family income, relative poverty level of 10.0 in 2009 (Cunningham, 2018), and over six
compared an average, parental education, or paren- million American were impacted by home foreclo-
tal wealth, are at increased risk of internalizing sures (Starting Over, 2021). Effects of the Great
symptoms and a broad range of other mental health Recession were felt across the world as well, for
problems (Reiss, 2013). Yet, poverty is itself variable example, in Europe, regional unemployment was
by broader economic conditions in ways that are rising (Passari, 2020) banking systems and govern-
dynamic by time and place. Recessions and global ments were collapsing (Britanica, 2009) and debt
economic conditions affect the level of poverty in a crises were rising (Zlan & Debeuf, 2020). Recovery
given country or community, and is thus worth from the recession lasted years; unemployment in
considering in the rise of adolescent internalizing the United States did not rebound to pre-recession
symptoms. There are multiple mechanisms through levels until 2014, and housing markets remained
which recessions and other global economic condi- precarious even later. Families in already precarious
tions may impact adolescent mental health, includ- financial and labor positions experienced worse
ing increased poverty, parental stress, and housing consequences of the recession (Nobari, Whaley,
instability. Indeed, studies of recessions and adoles- Crespi, Prelip, & Wang, 2018).
cent mental health have documented associations
across a number of psychological domains (Cotti & Is the association stronger for girls
Simon, 2018; Kim & Hagquist, 2018; Rathmann than boys?. Several studies have documented
et al., 2016), as well as with suicidal behavior stronger effects of economic recessions among girls
(Gassman-Pines, Gibson-Davis, & Ananat, 2015). compared with boys. For example, Bubonya et al
In more recent data, a review of 40 studies assessing (Bubonya, Cobb-Clark, Christensen, Johnson, &
the impact of the late 2000s global economic shock Zubrick, 2019) reported that the impact of the Great
(termed ‘the Great Recession’ lasting from approxi- Recession in Australia was associated with higher
mately 2007 to 2009) on child and adolescent mental mental health problems among girls (but not boys),
health and substance use documented substantial and Cotti (Cotti & Simon, 2018) reported that
heterogeneity in the impact of the recession on child changes in the stock index during the Great Reces-
and adolescent mental health across countries sion was associated emotional difficulties among US
(Hiilamo, Hiilamo, Ristikari, & Virtanen, 2021), but girls (but not boys). Other studies have also found a
that studies generally report that families most stronger effect of economic changes among girls
financially affected by the recession evidence the across several countries (Gassman-Pines, Oltmans
highest increases in child mental health problems. Ananat, & Gibson-Davis, 2014), but other studies
Data have been drawn from many countries across have found no differences (Golberstein et al., 2019).
Europe, the United States, and Australia, but there Thus, a differential effect of the economic recession

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 11

and its ensuing changes is plausibly varying by

Not enough evidence

research is needed
No for most, yes for
Plausible cause of
gender, but more data are certainly needed to

Plausible, more
confirm.

internalizing

to evaluate
conditions?
increase in

bullying
Artifactual explanations

Mixed

Mixed
No

No
In addition to the potentially causal factors dis-
cussed above, there is the possibility that the

Yes, if considering all aspects

increases observed are artifacts of measurement,

conditions getting stronger?

such as willingness to report symptoms rather than

Association between risk

factor and internalizing

(biological and social)

of pubertal transition
true changes in symptom occurrence. Scholars have
tested whether internalizing measures (Baxter
et al., 2014) and changes in mental health stigma
(Sweeting, Young, & West, 2009) and awareness

Not assessed

Not assessed
(Ortiz & Khin Khin, 2018) explain trends, but data

Unknown
are not sufficient to conclude that observed
increases are artifactual. For example, to test the

No

No

No
impact of reporting bias, authors have compared
trends in mental health conditions with other

boys, strong evidence

Yes, decreasing less in

compare time trends
disorders in boys to
Prevalence changing
stigmatized outcomes (Pickett et al., 2013) and found
Table 1 Summary of evidence for several known risk factors for internalizing conditions as explanations for increasing rates

more for girls than

Evidence of eating
Yes, in potentially

Mixed evidence in
that changing trends were specific to internalizing

bullying in girls
Yes, increasing
conditions, supporting the conclusion that changes

mixed ways
are not solely artifactual.

in girls
boys?

girls
No

No
Summary
Decreasing for

disorders are
changing over

The search to explain gender differences in internal-

Prevalence is

precocious
most SLEs

decreasing

decreasing
increasing

increasing
puberty is

Yes, eating
izing conditions has included many potential causal
factors. Leveraging changing mental health trends
time?

among adolescents allows us to critically review

Yes,

Yes,

Yes,

Yes
No
existing studies to evaluate whether several of these
known causes of internalizing conditions could
Types and responses differ, but

mechanisms Strong support

Mixed support for Biological

plausibly be responsible for growing gender differ-

Association stronger among

ences. Based on this review, we have ruled out

for social mechanisms
not overall prevalence

several factors, while also identifying several poten-

tial causes that may explain part of the growing
disparities (Table 1). Trends in cyber-bullying,
Not assessed
education-related pressures, and an earlier emer-
gence of puberty all appear to be small but poten-
tially important factors in the worsening mental
girls?

health of adolescent girls. Furthermore, these factors

Yes

Yes
No

No

are all largely modifiable either directly or indirectly

precocious puberty
Yes, during puberty,

(i.e., social stress from early puberty). In addition,

there is some evidence that macro-economic factors
Associated with

may have gendered effects. More research is needed,

internalizing
conditions?

especially
adolescent

inclusive of contexts outside the global North, in

order to fully evaluate the generalizability of causal
mechanisms. However, this evaluation may provide
Yes

Yes

Yes
Yes

Yes

Yes

a helpful focus on key areas to target in order to

attenuate growing disparities. However, we note that
Gender inequality in education

our review is not exhaustive of all potential risk

Macro-economic conditions
Stressful life events (SLEs)

Comorbid substance use

factors for internalizing symptoms; rather, we hope

that what we have illustrated in this section is a
Comorbid psychiatric
Hormonal changes

rigorous approach to evaluating evidence for causes

of the increase in internalizing conditions. As other
Genetic factors

researchers develop hypotheses about other causes

Risk factor

disorders

of the increases, the framework and decision points

that we have provided in this section may be useful
in guiding evaluation.

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
12 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

(social media, TV, video games, and Internet stream-

Part IV: Several novel risk factors: the evidence
ing), and use relatively crude measures of amount
In this section, we review the strength of the evidence
and time of use, limiting ability to draw more
for the impact of several novel risk factors on
nuanced inference. Carefully done longitudinal
increases in adolescent internalizing conditions.
studies that separate between- versus within-person
Two central hypotheses have emerged to explain
variance in psychological distress and perceived
the increases in depressive symptoms among youth:
impairment due to digital technology use find that
digital technology use, and within that the more
associations are weak, and reflect broader environ-
specific category of social media use. As in our
mental vulnerability to some degree, suggesting that
section on known risk factors, the selected risk
identifying adolescents at risk of mental health
factors reviewed here are not exhaustive of all
problems should be placed within existing public
potential novel risk factors. Our selection of digital
health frameworks of risk identification (Burnell &
technology and social media is based on the amount
Odgers, 2023). Other studies using intensive longi-
of attention they have received as potential contrib-
tudinal data find generally positive effects of sus-
utors to the rise in internalizing symptoms in youth,
tained digital technology use (Marciano, Driver,
but others may have chosen different novel risk
Schulz, & Camerini, 2022). Indeed, selection effects
factors to review. Nonetheless, we posit that the
remain inadequately accounted for in most studies;
epidemiological framework that we have used here is
a plurality of studies remain cross-sectional, and
a useful guide for any hypothesis about a risk factor
even longitudinal studies do not have sufficient
that may be contributing to the increase in adoles-
information on mental health over time to document
cent internalizing symptoms.
the extent to which students with mental health
problems select into particular social media pat-
Screen time and digital technology use terns. Indeed, digital technology use does not occur
in a vacuum; students who experience online
Perhaps the most visible change, and therefore
bullying are more likely to experience offline bullying
among the most predominant hypotheses for the
as well, for example, and bullying experiences are
increase in adolescent internalizing conditions, is
concentrated among youth already at higher risk
digital technology use, including multiple modalities
(e.g., LGBT, racial/ethnic minority, low-income stu-
of screen time such as smartphones and tablets,
dents) (Earnshaw, Bogart, Poteat, Reisner, & Schus-
laptops, and digital streaming, as well as use of
ter, 2016; Jackman, Kreuze, Caceres, &
social media.
Schnall, 2020; Odgers & Jensen, 2020a). A recent
study of high school students in high-achieving
Is digital technology use associated with adoles-
schools documented that time spent on digital media
cent internalizing conditions?. There has been an
(including texting, social media, YouTube, and other
extraordinary amount of past-decade research on
platforms) was associated with higher adolescent
digital technology use and mental health, both in
internalizing symptoms through social comparison,
adolescents and more broadly defined age catego-
negative feedback from peers and risk self-
ries. The research is voluminous enough to be
presentation (e.g., posting pictures they would not
synthesized across multiple narrative reviews (Abi-
want their parents to see) (Moroney, Lee, Ebbert, &
Jaoude, Naylor, & Pignatiello, 2020; Odgers &
Luthar, 2022), suggesting again that the quality of
Jensen, 2020b), systematic reviews (Tang, Werner-
interactions within digital platforms may be more
Seidler, Torok, Mackinnon, & Christensen, 2021),
salient than amount of exposure.
narrative reviews of systematic reviews
Video game use has received perhaps substantial
(Orben, 2020), systematic reviews of systematic
attention in the literature, especially the extent to
reviews (Dickson et al., 2019), and multiple meta-
which video games that depict violent content, or
analyses (Ferguson et al., 2021; Sohn, Rees, Wild-
allow the user to engage in violence (e.g., first-person
ridge, Kalk, & Carter, 2019), which account for
shooter games) are causally associated with
hundreds of individual studies.
increases in aggression and violence in children, or
A consistent conclusion is that measures such as
more broadly with psychological distress and other
time spent using digital technology are weakly
mental health problems. Available reviews and meta-
negatively related to adolescent internalizing condi-
analyses actually come to conflicting conclusions,
tions, including depression, anxiety, distress, and
with one read of the evidence suggesting small but
aligned measures. Predominant hypothesized mech-
present associations (Anderson et al., 2010; Fergu-
anisms in the general category of increased time
son, 2015; Prescott, Sargent, & Hull, 2018), and
using digital technology include less face-to-face
others suggesting null effects (Ferguson, 2007).
interaction with peers and activities, more sedentary
Certainly, the plurality of evidence suggesting that
behavior, and sleep disruption, as well as exposure
children and adolescents with underlying risk fac-
to distressing content (other mechanisms specific to
tors for negative mental health consequences of
social media are reviewed below). However, studies
exposure to violent material may be most impacted
often combine heterogenous types of digital media
(Ferguson, 2015; Mathur & VanderWeele, 2019). To

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 13

the extent to that video games are increasing in and available evidence does not indicate gender
violence material, and are increasingly being con- differences in general reports of amount of time
sumed among children and adolescent, the using digital technology or ownership of smart
population-level impact of small effect sizes may phones and other devices used for steaming content.
accumulate to be substantial. Yet, conclusions There is emerging evidence of gender differences in
should be balanced with the additional growing the types of platforms more frequently accessed,
evidence base for the efficacy of using video games although the rapid speed of changing online prefer-
for treatment of internalizing conditions (Pine, Flem- ences make any conclusions relatively ephemeral. A
ing, McCallum, & Sutcliffe, 2020). recent national survey of US adolescents noted that
Furthermore, the extent to which small but while all adolescents frequently use YouTube, girls
negative associations may translate to population- report using platforms such as TikTok and Insta-
level or public health impacts remain debated. Small gram more recently than boys, and conversely, boys
magnitude associations can have moderate-to-large report more Twitch and Reddit use (Vogels
public health impacts as prevalence increases. Yet, et al., 2022). The extent to which these gender
methodological issues with the available literature differences in the types and platforms of digital
simply remain too dominant to generate conclusions technologies used have distinct relationships to
for public health. Additionally, many scholars have mental health, however, have yet to be established.
noted the potential mental health benefits of digital
technology (Allen, Ryan, Gray, McInerney, &
Social media
Waters, 2020; Haidt & Allen, 2020; Odgers &
Jensen, 2020a), for both building and maintaining Among the sources of digital technology that has
social connections especially for marginalized youth, prompted the most concern is social media, that is,
and for delivering interventions for depression, platforms that allow for users to share personal
suicidal crises, and psychological distress (McGorry content, ideas and opinions, sometimes through
et al., 2022). Thus, balancing the potential risks of known social networks and often with persons
‘too much’ digital technology use, a broad construct unknown to the user. Early research focused on
with limited utility, and the potential benefits of Facebook, yet usership has declined in recent years
harnessing new technologies to better reach youth in among adolescents, with platforms such as Twitch,
need of support, remains critical. TikTok, Twitter, Instagram, Snapchat, and others
comprising popular platforms for youth. Much like
Is the prevalence increasing over the relevant time digital technology research in general, the status of
period?. Data are clear that digital technologies our knowledge of the effects of social media on youth
have rapidly increased in the past 10 years. The is rapidly accumulating and evolving. The potential
proliferation of tablets, smart phones, watches, are mechanisms through which social media is hypoth-
just some of the devices that now are at the hands of esized to increase internalizing symptoms include
almost all adolescents in the US and elsewhere, and direct mechanisms, such an increase in social
the availability of content through YouTube, stream- comparison leading to low self-esteem, increases in
ing services, TikTok, and other platforms ensure that cyber bullying. Indirect mechanisms include those
distraction is always available. Indeed, the profit that are also hypothesized to mediate potential
directives of these devices and platforms directly effects of general amount of time spent using digital
hinge on increasing engagement among users, often technology and mental health (e.g., sleep and less
youth (Anderson & Jiang, 2018). Available evidence physical activity).
indicates that adolescents in 2016 spent twice the
amount of time using digital technology than they Is social media use associated with adolescent
did in 2006 (Twenge, Martin, & Spitzberg, 2018), for mental health?. Research on the relationship
example, and survey data indicate that use has between social media exposure and internalizing
continued to accelerate since that time in a relatively problems is decidedly mixed. There have been at
linear fashion (Anderson & Jiang, 2018). 95% of least 14 scoping and narrative reviews (Bozzola
nearly all US teens (ages 13–17) say they have access et al., 2022; McCrory, Best, & Maddock, 2020;
to a smartphone and 97% report accessing the Verduyn, Ybarra, R
esibois, Jonides, & Kross, 2017;
Internet daily. In 2022, 95% of teens in the US Vidal, Lhaksampa, Miller, & Platt, 2020), systematic
report ever using YouTube, 67% report ever using reviews (Damodar et al., 2021; Keles, McCrae, &
TikTok, 62% report ever using Instagram, 59% Grealish, 2019; McCrae, Gettings, & Purssell, 2017;
report ever using snapchat, 32% Facebook, and Memon, Sharma, Mohite, & Jain, 2018; Radtke,
23% Twitter (Vogels, Gelles-Watnick, & Apel, Schenkel, Keller, & von Lindern, 2022; Sedg-
Massarat, 2022). wick, Epstein, Dutta, & Ougrin, 2019), reviews of
reviews (Valkenburg, Meier, & Beyens, 2022), meta-
Is the association or prevalence change differential analyses (Hancock, Liu, Luo, & Mieczkowski, 2022;
by gender?. Digital technology use in general is Ivie, Pettitt, Moses, & Allen, 2020), meta-analyses of
ubiquitous among both adolescent boys and girls, reviews (Appel, Marker, & Gnambs, 2020; Meier &

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
14 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

Reinecke, 2020) on the relationships between social on body image and disordered eating indicates a
media use and mental health both among adoles- stronger signal for a causal effect than research that
cents and other age groups, with an array of more generally measures social media use and
conclusions emerging. First, numerous studies find mental health (Engeln, Loach, Imundo, & Zola, 2020;
a J-shaped curve in the association between time Robinson et al., 2019; Vandenbosch, Fardouly, &
spent on social media and internalizing symptoms Tiggemann, 2022). Finally, consensus on the nature
(Przybylski & Weinstein, 2017). That is, symptoms and magnitude of the association between digital
are elevated among student who use no or low technology use and internalizing conditions is chal-
amounts of social media, compared to those who lenging as the exposure is rapidly changing year to
use moderate amounts, and then escalate as year. Understanding whether social media is a risk
amounts of daily social media exposure increase. factor requires a stable construct to repeatedly
Yet, other studies do not find the J-shaped curve, measure across time and study population, yet the
and conclusions about causality remain uncertain, rapidly changing landscape necessarily means that
given that selection into high levels of social media research is ever-evolving as well. Available evidence
use based on mental health symptoms remain a indicates little support for change in the association
threat to validity. Second, experimental studies in between social media use and mental health among
which participants (often adults) use social media in adolescents (Vuorre, Orben, & Przybylski, 2021),
laboratory environments and report mood suggest and the magnitude of the change in the association
negative self-reported mood after use, as well as across time differs across measures of exposure and
improvements in mood and anxiety when random- outcome, making for a slippery evidence-base from
ized to social media breaks or ‘digital detox’, which to draw conclusions.
although the extent to which those experiments It is worth noting that a burgeoning literature has
generalize to internalizing symptoms or conditions, attempted to construct measures of social media
adolescents, or to real-world patterns of social media ‘addiction’ or dependence, but the documented
use, remain equivocal (Radtke et al., 2022). Third, validity of such constructs remains controversial.
the effect size of social media associations with Scales such as the Bergen Facebook Addiction Scale
internalizing symptoms is small, with correlations (Andreassen, TorbjØrn, Brunborg, & Pallesen, 2012)
generally ranging from about 0.1 to 0.15 across (and modified Bergen Social Media Addiction Scale)
adolescent populations. Some have argued that the measure canonical phenotypes of dependence such
effects of other known risk factors for internalizing as impaired control over use, urgency to use,
conditions are much stronger (Orben & Przy- unsuccessful quit attempts, and negative impacts
bylski, 2019), suggesting a mismatch between the in role function and sleep. However, critics of the
amount of attention on social media as a causal addiction framework for social media use include
factor, and the strength of the evidence for its role in that many canonical concepts in addiction are
adolescent internalizing conditions. insufficiently characterized (e.g., tolerance and with-
Furthermore, a central theme that is emerging drawal), suggesting that pathologizing patterns of
from this literature is that it is critical to bring more use may be premature (Panova & Carbonell, 2018).
depth into the measurement of social media use. Further, social media addiction or ‘problematic use’
Measures of ‘time spent’ using social media are research draws from a range of social theories about
insufficiently crude to capture unique risk and the emergence and risk for addictive behaviors (Sun
protective factors (Steers, 2016). In addition to & Zhang, 2021), and the rapidly evolving landscape
studies that have shown that self-reports of time on of social media use among adolescents suggest that
social media have substantial measurement error, research into disordered use phenotypes too will
adolescents’ online lives should be considered an need to rapidly evolve with the media through which
additional environment in which both harms and adolescents are engaging. Meta-analyses of preva-
benefits may accrue, much like home or school. If we lence estimates across the world show substantial
measured ‘time spent’ at school or at home as a risk heterogeneity and concentrated prevalence among
factor for mental health problems, it would be an adolescents and young adults. Depending on the
absurdly crude measure of potential risk factors in scale cut-off specified, pooled prevalence estimates
the school and home environment, and our resulting range from 5% for ‘severe’ to 25% for more lenient
research would likely be as inconsistent as the cut-offs for categorizing disorder (note that the meta-
current literature base for social media. Emerging analysis included adult samples) (Cheng, Lau, Chan,
taxonomies of social media and other technology & Luk, 2021). Evidence for validity of social media
measurement that focus on specific channels, inter- addiction generally derive from convergent validity,
actions, practices (passive versus active engage- with higher scores on scales correlating with other
ment), and motivations for use allow for a richer dimensions commonly associated with addictive
and more complex characterization of potential risks disorders (Andreassen, Pallesen, & Griffiths, 2017).
and benefits of social media use for youth (Meier & Finally, as noted above, there is growing consen-
Reinecke, 2020). For example, specific research on sus that the amount or type of social media exposure
Instagram and Snapchat use and negative impacts is likely to crude of a metric for assessing causal

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 15

effects of exposure, but rather, that the information compared with boys (Keles et al., 2019; Kelly,
within consumed media may be a more salient Zilanawala, Booker, & Sacker, 2018; Twenge, 2020),
direction for assessing impacts on youth. Adoles- although other reviews have found no gender
cents are exposed to a wide range of information, differences (Meier & Reinecke, 2020). Particular
from political news to thoughts and ideas, through aspects of social media, such as social comparison
technologies that are primed to keep adolescents (Nesi & Prinstein, 2015), and online harassment, are
watching, and serve them content that is aligned consistently stronger risk factors for internalizing
with their interests. Available evidence indicates that outcomes among girls compared with boys. Social
mental health problems have increased at a greater media as a mechanism to engage in discussion about
pace for adolescents who identify as political liberal self-harm and suicidal behavior is more frequently
compared with conservative (Gimbrone, Bates, utilized by girls (Memon et al., 2018). Furthermore,
Prins, & Keyes, 2022), which may be indicative of consistent evidence that social media prompts
reactions to increasing awareness of inequality, increases in body dissatisfaction and dysmorphia
unfair treatment of people with marginalized identi- as well as disordered eating more strongly among
ties and historical oppression such as immigrants girls compared with boys, which may in turn
and racial/ethnic minority populations, climate underlie increases in internalizing symptoms
anxiety, and other aspects of geopolitics that are (Hawes, Zimmer-Gembeck, & Campbell, 2020; Hol-
distressing when considering long-term future pros- land & Tiggemann, 2016; Saiphoo & Vahedi, 2019).
pects at an early age. The so-called ‘culture wars’ are These associations are amplified when considering
evident in every generation, but take new shape as antifat bias and weight stigma. While trends in
young people are activated and engaged in online pediatric and adolescent obesity have remained
content creation and consumption that reinforces fairly stable since 2010 (Sanyaolu, Okorie, Qi, Locke,
out-group and in-group thinking on all sides of the & Rehman, 2019), evidence indicates that weight
political spectrum. The increasing silos of informa- discrimination, antifat bias, and bullying have
tion by political affiliation and worldview, and the increased, especially on social media (Clark
reality that monetization of online content is opti- et al., 2021). Experiences of weight discrimination
mized by presenting the most extreme views on each are associated with psychiatric conditions, and
political side, may push adolescents towards dis- should be considered as a potential explanation
tress as they engage with content that further underlying associations when considering overall
polarizes thought and belief. Continued emphasis body image. However, as with other research on
on novel taxonomies of exposure, including polari- social media and mental health, limitations of the
zation of belief and ideology, will benefit the field as existing evidence base are numerous (Fardouly &
research moves forward. Vartanian, 2016), prompting the need for additional
research.
Is the prevalence increasing over the relevant time
period?. Similar to digital technology exposure in
Additional pathways
general, social media exposure has rapidly increased
in the past 10 years. From the early emergence of We have outlined the evidence for several novel
MySpace and Facebook, adolescent use of platforms causes of underlying increases in adolescent inter-
such as YouTube that allow for social interaction are nalizing conditions (Table 2). Yet, there remain an
now ubiquitous, and as time has progressed, addi- array of potential risk factors that are less well-
tional platforms have emerged. Data from the United documented, and less well-studied, that may be
States indicate that the majority of adolescents candidates for future research to integrate into the
report using social media platforms, and 15%–20% growing evidence base.
report using social media platforms ‘almost con- Additional pathways may be relevant within spe-
stantly’ (Vogels et al., 2022). While the specific sites cific context and countries, and responses to growing
popular with youth change frequently, there is internalizing symptoms may also be context and
substantial evidence to indicate that usage overall country specific as well. In the United States, for
has increased across many countries (Ruby, 2023). example, the escalating drug overdose (Jalal
et al., 2018) and gun violence (including school
Is the association stronger for girls shooting (Irwin, Wang, Cui, & Thompson, 2021))
than boys?. Evidence is relatively consistent that crises has affected millions within families. The role
at least some potential adverse effects of social media of school and family engagement and conflict,
are stronger among girls compared with boys, educational pressures and distress about college
consistent with social media being a plausible and university access and acceptance, declines in
explanation for why internalizing symptoms have physical activity (Davison, Schmalz, & Downs, 2010),
increased more for girls than for boys. Several sports participation, dating (Twenge & Park, 2019),
reviews have documented increases in the magni- climate anxiety (Hickman et al., 2021), religious
tude of the association between overall usage of service attendance and potential other protective
social media and mental well-being among girls factors such as community engagement (Kreski,

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
16 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

Table 2 Summary of evidence for several novel risk factors for internalizing conditions as explanations for increasing rates

Associated with Association Prevalence Association between risk Plausible cause of

adolescent stronger Prevalence changing more factor and internalizing increase in
internalizing among is changing for girls than conditions getting internalizing
Risk factor conditions? girls? over time? boys? stronger? conditions?

‘Screen time’ Mixed/small No Yes No No No

and digital associations
technology
use
Social media Mixed/moderate Yes Yes No Yes Plausible
associations

Chen, Olfson, Cerd
a, Hasin, et al., 2022; Wray-Lake, consistent that increases are concentrated in girls,
Schulenberg, Keyes, & Shubert, 2017), have all and thus we have reviewed the evidence for what
received additional attention (Sweeting causes gender differences in internalizing symp-
et al., 2010). Acute economic and political instability toms, as well as reviewed evidence for novel risk
are affecting populations across the globe, leading to factors that are changing over time and discussed
unprecedented displacement (Charlson et al., 2019; the extent to which their changes are gender-specific
UNHCR, 2021). in their effects.
Available data that have emerged since the onset of One central conclusion from this work is that there
the COVID-19 pandemic indicates that psychological are no simple stories for the increase in internalizing
distress may be accelerating at an even more rapid conditions in youth, and therefore it will require
pace in recent years (Harrison, Carducci, Klein, & substantial additional research and training to
Bhutta, 2022; Racine et al., 2021), although in continue to unpack the underlying reasons so that
promising news, suicide among young people has evidence-based solutions can be effectively imple-
stabilized or declined in most countries during the mented. Primary hypotheses for the reasons under-
pandemic (Pirkis et al., 2022). Furthermore, there has lying these increases include population level shifts
been substantial variability in the impact of the in puberty towards younger ages, digital technology
pandemic on youth mental health across countries and social media use, and economic factors, among
(Chavira, Ponting, & Ramos, 2022), and increases in other less studied mechanisms. As we have reviewed
psychological distress due to the pandemic among here, despite hundreds of studies on these hypoth-
youth are concentrated youth in families financially esized factors, the evidence base has significant
and otherwise adversely impacted by the pandemic limitations; investment in high-quality measure
(Hossain et al., 2022; Xiao et al., 2023). Continued and theory development that moves beyond basic
focus on understanding the potential short- and long- measures and incorporates online and offline risks
term consequences of pandemic-related illness and together as additional environments of both risk and
bereavement, disruptions in socialization and learn- protective factors are necessary. Our review high-
ing, and other effects of the pandemic will be a critical lights that there may be significant interactions
part of mental health surveillance for years to come. between the hypothesized causes of increases in
We propose that the epidemiological approach that adolescent internalizing symptoms, with earlier
we have described in this paper can be a model worth development intersecting with new online and offline
considering as these areas of research continue to risks that together underlie a pernicious space for
receive attention in coming years. For all considered the onset of a wide range of psychiatric risk.
pathways, assessing the strength of association, Furthermore, the science is woefully behind on
interactions with sex and gender, and changing moving beyond binary measures of gender. Not only
prevalence should be considered simultaneously to is binary gender inaccurate, but without assessing
enhance the rigor of research. internalizing symptoms across a broader spectrum
of gender, we miss important disparities. Indeed, the
available literature indicates that nonbinary, gender
Part V: Conclusion: levels of intervention and nonconforming, and transgender youth are at
pathways forward heightened risk for a range of psychiatric disorders
We have outlined the evidence that internalizing including mood and anxiety disorders, and for
conditions, including depression and anxiety, psy- suicide (Reisner et al., 2016; Wolford-Clevenger,
chological distress, and suicidal behaviors, are Frantell, Smith, Flores, & Stuart, 2018). Transgen-
increasing among adolescents across a number of der and gender nonconforming people worldwide
different countries. The evidence is accumulating face minority stress and discrimination (McCann,
that these increases are not just noise in the data, Donohue, & Brown, 2021), and difficulties accessing
but true increases affecting the lives of millions of health care (Koch, McLachlan, Victor, Westcott, &
adolescents and their families. Evidence is also Yager, 2019). In recent years, as the human rights of

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 17

transgender and gender nonconforming people has prevention, disorders are prevented prior to onset,
become more of a political spotlight, discrimination, rather than symptoms and crises treated after onset.
personal attacks, and restrictions on healthcare Numerous evidence-based programs are available for
access may further exacerbate mental health dis- primary prevention of internalizing disorders among
parities with cisgender people. Yet regular surveil- youth (Andrews et al., 2022; Godwin, 2020; Werner-
lance efforts often still focus on binary sex/gender, Seidler et al., 2021), and while they differ in their
or do not allow for disaggregation of trends by approach, some commonalities across programs
nonbinary gender categorization, which limits sci- provide evidence of core themes. Programs often
ence and intervention. include social–emotional learning, skills-based pro-
In many countries, mental health treatment for gramming to provide youth with tools to process
adolescents remains difficult to access and develop- negative emotions, and peer-support and peer-led
mentally inappropriate (McGorry et al., 2022). Many group work. A central core theme is also that mental
countries across the world, especially high-income well-being requires a sense of connection and
countries, are developing novel and scalable inter- meaning-making within the complex lives of adoles-
ventions and systems of care for youth, including by cents, the sense that adolescents have a central place
leveraging digital technology, that hold promise but in the world, robust social network to turn to when
require sustained investment and attention emotions are negative, and have a sense of purpose
(McGorry et al., 2022). Across many countries, the and future-oriented goals. Religious organizations
mental health workforce is simply not sufficient to historically were among the central hubs for this type
meet demand, especially in rural areas. Locating an of preventive work, yet religious affiliation and con-
appropriate provider who is sufficiently trained in nection has declined in recent decades among youth,
the complex array of and comorbidity within psychi- can be the cause of serious stressful life events, and
atric symptoms that are often present in adolescents youth identities (e.g., queer youth) that mismatch
is a serious challenge and often untenable to many with religious tenants can cause mental health
families. Primary care providers are often tasked problems through rejection and shame rather than
with treating conditions for which they have little ameliorate them. Developing a robust public health
specialized training. Thus, it is clear that while approach to primary prevention, drawing on decades
increased funding for treatment is a priority, espe- of empirical findings, and implementing in environ-
cially for adolescents who are currently experiencing ments where youth congregate such as the school
psychological distress, building the psychiatric setting, is among the most urgent and critical
workforce is a significant undertaking. necessary immediate steps to curtail growing inci-
Suicide prevention through crisis intervention is a dence and prevalence.
critical pillar to mitigate the mental health crisis
among youth. National efforts to scale crisis inter-
vention is expanding access in many countries Acknowledgments
(Hirschtritt, Howard, & Simon, 2023). Furthermore, The authors would like to thank Victoria Joseph, MPH,
proposals to expand nonpolice interventions for for editorial assistance. Sources of financial support:
people in crisis are expanding, especially important R01-MH121410; R01-DA048853.
for individuals racialized with minority status who
may underuse crisis intervention out of concern for
police intervention that may turn violent or fatal and Correspondence
may separate families and children (Zabelski, Katherine M. Keyes, Mailman School of Public Health,
Columbia University, 722 West 168th St., Room 724,
Kaniuka, Robertson, & Cramer, 2022).
New York, NY 10032, USA; Email: kmk2104@columbia.
While treatment and crisis intervention expansion
edu
are critically necessary, a focus on primary prevention
cannot be left to an afterthought. In primary

Key points
• Internalizing conditions including depression and anxiety, psychological stress, suicidal ideation and
self-harm, and fatal suicide are increasing in adolescent population in many countries, with increases
currently concentrated in female adolescents.
• Drivers of these increases are likely a multitude of interacting factors, and evidence supports the
plausibility of hormonal changes, and aspects of social media exposure, as having sufficient evidence
base for conclusion that they are contributing to increases.
• More research is needed on broader societal drivers of internalizing psychopathology among youth, as
well as nonbinary and gender nonconforming youth.

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
18 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

References Bebbington, P. (1996). The origins of sex differences in

Abbafati, C., Abbas, K.M., Abbasi-Kangevari, M., Abd-Allah, depressive disorder: Bridging the gap. International Review
F., Abdelalim, A., Abdollahi, M., . . . & Murray, C.J.L. (2020). of Psychiatry, 8, 295–332.
Global burden of 369 diseases and injuries in 204 countries Berny, L.M., & Tanner-Smith, E.E. (2022). Differential pre-
and territories, 1990–2019: A systematic analysis for the dictors of suicidal ideation and suicide attempts: Internal-
Global Burden of Disease Study 2019. The Lancet, 396, izing disorders and substance use in a clinical sample of
1204–1222. adolescents. Journal of Dual Diagnosis, 18, 59.
Abi-Jaoude, E., Naylor, K.T., & Pignatiello, A. (2020). Smart- Bor, W., Dean, A.J., Najman, J., & Hayatbakhsh, R. (2014). Are
phones, social media use and youth mental health. CMAJ, child and adolescent mental health problems increasing in
192, E136–E141. the 21st century? A systematic review. Australian & New
Abraham, Z.K., & Sher, L. (2019). Adolescent suicide as a Zealand Journal of Psychiatry, 48, 606–616.
global public health issue. International Journal of Adoles- Bozzola, E., Spina, G., Agostiniani, R., Barni, S., Russo, R.,
cent Medicine and Health, 31. Available from: https://www. Scarpato, E., . . . & Staiano, A. (2022). The use of social Media
degruyter.com/document/doi/10.1515/ijamh-2017-0036/ in Children and Adolescents: Scoping review on the potential
html [last accessed 13 July 2023]. risks. International Journal of Environmental Research and
Abramovich, A., & Cleverley, K. (2018). A call to action: The Public Health, 19, 9960.
urgent need for trans inclusive measures in mental health Bradford, S.A., Cbe, H., Frcp, D., & Frs. (1965). The
research. Canadian Journal of Psychiatry. Revue Canadi- environment and disease: Association or causation? Pro-
enne de Psychiatrie, 63, 532. ceedings of the Royal Society of Medicine, 58, 295.
Allen, K.A., Ryan, T., Gray, D.L., McInerney, D.M., & Waters, L. Brener, N.D., Bohm, M.K., Jones, C.M., Puvanesarajah, S.,
(2020). Social media use and social connectedness in Robin, L., Suarez, N., . . . & Moyse, D. (2022). Use of tobacco
adolescents: The positives and the potential pitfalls. The products, alcohol, and other substances among high school
Educational and Developmental Psychologist, 31, 18–31. students during the COVID-19 pandemic – Adolescent
Anderson, C.A., Shibuya, A., Ihori, N., Swing, E.L., Bushman, behaviors and experiences survey, United States, January–
B.J., Sakamoto, A., . . . & Saleem, M. (2010). Violent video June 2021. MMWR Supplements, 71, 8–15.
game effects on aggression, empathy, and prosocial behavior Britannica. (2009). Great Recession|Causes, effects, statistics,
in eastern and Western countries: A meta-analytic review. & facts. Available from: https://www.britannica.com/topic/
Psychological Bulletin, 136, 151–173. great-recession [last accessed 26 January 2023]
Anderson, M., & Jiang, J. (2018). Teens, social media & Bronfenbrenner, U. (1994). Ecological models of human
technology 2018. Pew Research Center. Available from: development. International Encyclopedia of Education, 3,
https://www.pewresearch.org/internet/2018/05/31/ 37–43.
teens-social-media-technology-2018/#vast-majority-of- Bubonya, M., Cobb-Clark, D.A., Christensen, D., Johnson,
teens-have-access-to-a-home-computer-or-smartphone S.E., & Zubrick, S.R. (2019). The great recession and
[last accessed 26 January 2023]. children’s mental health in Australia. International Journal
Andreassen, C.S., Pallesen, S., & Griffiths, M.D. (2017). The of Environmental Research and Public Health, 16, 537.
relationship between addictive use of social media, narcis- Burnell, K., & Odgers, C.L. (2023). Trajectories of perceived
sism, and self-esteem: Findings from a large national survey. technological impairment and psychological distress in
Addictive Behaviors, 64, 287–293. adolescents. Journal of Youth and Adolescence, 52, 258.
Andreassen, C.S., TorbjØrn, T., Brunborg, G.S., & Pallesen, S. Campbell, O.L.K., Bann, D., & Patalay, P. (2021). The gender
(2012). Development of a Facebook Addiction Scale. Psycho- gap in adolescent mental health: A cross-national investi-
logical Reports, 110, 501–517. gation of 566,829 adolescents across 73 countries. SSM –
Andrews, J.L., Birrell, L., Chapman, C., Teesson, M., Newton, Population Health, 13, 100742.
N., Allsop, S., . . . & Slade, T. (2022). Evaluating the Campisi, S.C., Carducci, B., Akseer, N., Zasowski, C., Szat-
effectiveness of a universal eHealth school-based prevention mari, P., & Bhutta, Z.A. (2020). Suicidal behaviours among
programme for depression and anxiety, and the moderating adolescents from 90 countries: A pooled analysis of the
role of friendship network characteristics. Psychological global school-based student health survey. BMC Public
Medicine, 1–10. Available from: https://www.cambridge. Health, 20, 1–11.
org/core/journals/psychological-medicine/article/abs/ Case, A., & Deaton, A. (2015). Rising morbidity and mortality
evaluating-the-effectiveness-of-a-universal-ehealth- in midlife among white non-Hispanic Americans in the 21st
schoolbased-prevention-programme-for-depression-and- century. Proceedings of the National Academy of Sciences of
anxiety-and-the-moderating-role-of-friendship-network- the United States of America, 112, 15078–15083.
characteristics/745721E4F8E63892095BB0551F267F7F Castellanos-Ryan, N., Parent, S., Vitaro, F., Tremblay, R.E., &
[last accessed 13 July 2023]. S
eguin, J.R. (2013). Pubertal development, personality, and
Angold, A., Costello, E.J., & Worthman, C.M. (1998). Puberty substance use: A 10-year longitudinal study from childhood
and depression: The roles of age, pubertal status and to adolescence. Journal of Abnormal Psychology, 122, 782–
pubertal timing. Psychological Medicine, 28, 51–61. 796.
Appel, M., Marker, C., & Gnambs, T. (2020). Are social media CDC. (2023a). Youth risk behavior survey data summary &
ruining our lives? A review of meta-analytic evidence. Review trends report. Centers for Disease Control and Prevention.
of General Psychology, 24, 60–74. Available from: https://www.cdc.gov/healthyyouth/data/
Asher, M., Asnaani, A., & Aderka, I.M. (2017). Gender yrbs/pdf/YRBS_Data-Summary-Trends_Report2023_508.
differences in social anxiety disorder: A review. Clinical pdf [last accessed 23 February 2023].
Psychology Review, 56, 1–12. CDC. (2023b). CDC report shows concerning increases in
Barth, J., Bermetz, L., Heim, E., Trelle, S., & Tonia, T. (2013). sadness and exposure to violence among teen girls and
The current prevalence of child sexual abuse worldwide: A LGBQ+ youth. Centers for Disease Control and Prevention.
systematic review and meta-analysis. International Journal Available from: https://www.cdc.gov/nchhstp/newsroom/
of Public Health, 58, 469–483. fact-sheets/healthy-youth/sadness-and-violence-among-
Baxter, A.J., Scott, K.M., Ferrari, A.J., Norman, R.E., Vos, T., & teen-girls-and-LGBQ-youth-factsheet.html [last accessed 21
Whiteford, H.A. (2014). Challenging the myth of an “epi- February 2023].
demic” of common mental disorders: Trends in the global Chan, K.L. (2011). Gender differences in self-reports of
prevalence of anxiety and depression between 1990 and intimate partner violence: A review. Aggression and Violent
2010. Depression and Anxiety, 31, 506–516. Behavior, 16, 167–175.

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 19

Chapman, C., Slade, T., Swift, W., Keyes, K., Tonks, Z., & problem of intracategory variability. Psychological Bulletin,
Teesson, M. (2017). Evidence for sex convergence in 132, 477–495.
prevalence of cannabis use: A systematic review and meta- Dougherty, C. (2005). Why are the returns to schooling higher
regression. Journal of Studies on Alcohol and Drugs, 78, for women than for men? Journal of Human Resources, XL,
344–352. 969–988.
Charlson, F., van Ommeren, M., Flaxman, A., Cornett, J., Earnshaw, V.A., Bogart, L.M., Poteat, V.P., Reisner, S.L., &
Whiteford, H., & Saxena, S. (2019). New WHO prevalence Schuster, M.A. (2016). Bullying among lesbian, gay, bisex-
estimates of mental disorders in conflict settings: A system- ual, and transgender youth. Pediatric Clinics of North
atic review and meta-analysis. Lancet, 394, 240–248. America, 63, 999–1010.
Chavira, D.A., Ponting, C., & Ramos, G. (2022). The impact of Eckert-Lind, C., Busch, A.S., Petersen, J.H., Biro, F.M., Butler,
COVID-19 on child and adolescent mental health and G., Br€auner, E., & Juul, A. (2020). Worldwide secular trends
treatment considerations. Behaviour Research and Therapy, in age at pubertal onset assessed by breast development
157, 104169. among girls: A systematic review and meta-analysis. JAMA
Cheng, C., Lau, Y.C., Chan, L., & Luk, J.W. (2021). Prevalence Pediatrics, 174, e195881.
of social media addiction across 32 nations: Meta-analysis Engeln, R., Loach, R., Imundo, M.N., & Zola, A. (2020).
with subgroup analysis of classification schemes and Compared to Facebook, Instagram use causes more appear-
cultural values. Addictive Behaviors, 117, 106845. ance comparison and lower body satisfaction in college
Children, Youth, Families and Socioeconomic Status. (2010). women. Body Image, 34, 38–45.
American Psychological Association. Available from: Fardouly, J., & Vartanian, L.R. (2016). Social media and body
https://www.apa.org/pi/ses/resources/publications/child image concerns: Current research and future directions.
ren-families [last accessed 16 February 2023]. Current Opinion in Psychology, 9, 1–5.
Clark, O., Lee, M.M., Jingree, M.L., O’Dwyer, E., Yue, Y., Ferguson, C.J. (2007). The good, the bad and the ugly: A meta-
Marrero, A., . . . & Mattei, J. (2021). Weight stigma and social analytic review of positive and negative effects of violent
media: Evidence and public health solutions. Frontiers in video games. Psychiatric Quarterly, 78, 309–316.
Nutrition, 8, 913. Ferguson, C.J. (2015). Do angry birds make for angry
Connell, R. (1987). Gender and power: Society, the person, and children? A meta-analysis of video game influences on
sexual politics (334 pp.). Palo Alto, CA: Stanford University Children’s and adolescents’ aggression, mental health,
Press. Available from: http://www.sup.org/books/title/? prosocial behavior, and academic performance. Perspectives
id=2532 [last accessed 16 February 2023]. on Psychological Science, 10, 646–666.
Cosma, A., Bjereld, Y., Elgar, F.J., Richardson, C., Bilz, L., Ferguson, C.J., Kaye, L.K., Branley-Bell, D., Markey, P., Ivory,
Craig, W., . . . & Walsh, S.D. (2022). Gender differences in J.D., Klisanin, D., . . . & Wilson, J. (2021). Like this meta-
bullying reflect societal gender inequality: A multilevel study analysis: Screen media and mental health. Professional
with adolescents in 46 countries. Journal of Adolescent Psychology: Research and Practice, 53, 205–214.
Health, 71, 601–608. Finkelhor, D. (2020). Trends in adverse childhood experiences
Cosma, A., Stevens, G., Martin, G., Duinhof, E.L., Walsh, S.D., (ACEs) in the United States. Child Abuse & Neglect, 108,
Garcia-Moya, I., . . . & de Looze, M. (2020). Cross-national 104641.
time trends in adolescent mental well-being from 2002 to Finkelhor, D., Saito, K., & Jones, L. (2018). Updated trends in
2018 and the explanatory role of schoolwork pressure. The child maltreatment, 2016. Crimes Against Children Research
Journal of Adolescent Health, 66(6 Suppl), S50. Center. Available from: https://calio.dspacedirect.org/
€
Cosma, N.C., Usekes, B., Otto, L.R., Gerike, S., Heuser, I., handle/11212/3770 [last accessed 26 January 2023].
Regen, F., & Hellmann-Regen, J. (2021). M1/M2 polariza- Finkelhor, D., Turner, H., Ormrod, R., & Hamby, S.L. (2010).
tion in major depressive disorder: Disentangling state from Trends in childhood violence and abuse exposure: Evidence
trait effects in an individualized cell-culture-based from 2 National Surveys. Archives of Pediatrics & Adolescent
approach. Brain, Behavior, and Immunity, 94, 185–195. Medicine, 164, 238–242.
Cotti, C., & Simon, D. (2018). The impact of stock market Foulkes, L., & Andrews, J.L. (2023). Are mental health
fluctuations on the mental and physical well-being of awareness efforts contributing to the rise in reported mental
children. Economic Inquiry, 56, 1007–1027. health problems? A call to test the prevalence inflation
Cunningham, E. (2018). Great Recession, great recovery? hypothesis. New Ideas in Psychology, 69, 101010.
Trends from the current population survey. Monthly Labor Gallup. (2022). Global emotions report. Available from: https://
Review. Bureau of Labor Statistics. www.gallup.com/analytics/349280/gallup-global-emotions
Damodar, S., Gurusamy, V., Parrill, A., Lokemoen, C., Bishev, -report.aspx [last accessed 16 February 2023].
D., Takhi, M., . . . & Branch, R. (2021). 1.1 #Trending: Social Gassman-Pines, A., Gibson-Davis, C.M., & Ananat, E.O.
Media’s influence on adolescent anxiety and depression. (2015). How economic downturns affect children’s develop-
Journal of the American Academy of Child & Adolescent ment: An interdisciplinary perspective on pathways of
Psychiatry, 60, S139. influence. Child Development Perspectives, 9, 233.
Davison, K.K., Schmalz, D.L., & Downs, D.S. (2010). Hop, skip Gassman-Pines, A., Oltmans Ananat, E., & Gibson-Davis,
. . . No! Explaining adolescent Girls’ disinclination for phys- C.M. (2014). Effects of statewide job losses on adolescent
ical activity. Annals of Behavioral Medicine, 39, 290–302. suicide-related behaviors. American Journal of Public Health,
Dickson, K., Richardson, M., Kwan, I., Macdowall, W., 104, 1964–1970.
Burchett, H., Stansfield, C., . . . & Thomas, J. (2019). Ge, X., Conger, R.D., & Elder, G.H. (1996). Coming of age too
Screen-based activities and children and young people’s early: Pubertal influences on girls’ vulnerability to psycho-
mental health and psychosocial wellbeing: A systematic map logical distress. Child Development, 67, 3386–3400.
of reviews. Department of Health Reviews Facility To Ge, X., & Natsuaki, M.N. (2009). In search of explanations for
support national policy development and implementation. early pubertal timing effects on developmental psychopa-
Available from: http://eppi.ioe.ac.uk/ [last accessed 16 thology. Current Directions in Psychological Science, 18, 327–
February 2023]. 331.
DiPrete, T.A., & Buchmann, C. (2006). Gender-specific trends Ghandour, R.M., Sherman, L.J., Vladutiu, C.J., Ali, M.M.,
in the value of education and the emerging gender gap in Lynch, S.E., Bitsko, R.H., & Blumberg, S.J. (2019). Preva-
college completion. Demography, 43, 1–24. lence and treatment of depression, anxiety, and conduct
Dohrenwend, B.P. (2006). Inventorying stressful life events as problems in US children. Journal of Pediatrics, 206, 256–
risk factors for psychopathology: Toward resolution of the 267.e3.

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
20 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

Gimbrone, C., Bates, L.M., Prins, S.J., & Keyes, K.M. (2022). review of the evidence. Children and Youth Services Review,
The politics of depression: Diverging trends in internalizing 121, 105772.
symptoms among US adolescents by political beliefs. SSM. Hirschtritt, M.E., Howard, C.A., & Simon, G.E. (2023).
Mental Health, 2, 100043. Fulfilling the goals of 988 through crisis stabilization care.
Global Gender Gap Report 2020. (2019). World economic Washington, DC: Psychiatric Services. Available from:
forum. Available from: https://www.weforum.org/reports/ https://pubmed-ncbi-nlm-nih-gov.ezproxy.cul.columbia.
gender-gap-2020-report-100-years-pay-equality// [last edu/36718601/ [last accessed 19 February 2023].
accessed 16 February 2023]. H€ogberg, B. (2021). Educational stressors and secular trends
Godwin, J.W. (2020). The fast track intervention’s impact on in school stress and mental health problems in adolescents.
behaviors of despair in adolescence and young adulthood. Social Science & Medicine, 270, 113616.
Proceedings of the National Academy of Sciences of the Holland, G., & Tiggemann, M. (2016). A systematic review of
United States of America, 117, 31748–31753. the impact of the use of social networking sites on body
Golberstein, E., Gonzales, G., & Meara, E. (2019). How do image and disordered eating outcomes. Body Image, 17,
economic downturns affect the mental health of children? 100–110.
Evidence from the National Health Interview Survey. Health Holmes, T.H., & Rahe, R.H. (1967). The social readjustment
Economics, 28, 955. rating scale. Journal of Psychosomatic Research, 11, 213–
Goodwin, R.D., Weinberger, A.H., Kim, J.H., Wu, M., & Galea, 218.
S. (2020). Trends in anxiety among adults in the United Hope, D., & Martelli, A. (2019). The transition to the knowledge
States, 2008–2018: Rapid increases among young adults. economy, labor market institutions, and income inequality
Journal of Psychiatric Research, 130, 441–446. in advanced democracies. World Politics, 71, 236–288.
Graber, J.A. (2013). Pubertal timing and the development of Hossain, M.M., Nesa, F., Das, J., Aggad, R., Tasnim, S.,
psychopathology in adolescence and beyond. Hormones and Bairwa, M., . . . & Ramirez, G. (2022). Global burden of
Behavior, 64, 262–269. mental health problems among children and adolescents
Greenland, S., & Robins, J.M. (1986). Identifiability, exchange- during COVID-19 pandemic: An umbrella review. Psychiatry
ability, and epidemiological confounding. International Jour- Research, 317, 114814. Available from: https://www.ncbi.
nal of Epidemiology, 15, 413–419. nlm.nih.gov/pmc/articles/PMC9420079/pdf/main.pdf
Haidt, J., & Allen, N. (2020). Scrutinizing the effects of digital [last accessed 13 July 2023].
technology on mental health. Nature, 578, 226–227. Irwin, V., Wang, K., Cui, J., & Thompson, A. (2021). A
Hancock, J., Liu, S.X., Luo, M., & Mieczkowski, H. (2022). publication of the National Center for Education Statistics at
Psychological well-being and social media use: A meta- IES. Available from: https://bjs.ojp.gov [last accessed 26
analysis of associations between social media use and January 2023].
depression, anxiety, loneliness, Eudaimonic, hedonic and Ivie, E.J., Pettitt, A., Moses, L.J., & Allen, N.B. (2020). A meta-
social well-being. SSRN Electronic Journal. Available from: analysis of the association between adolescent social media
https://papers.ssrn.com/abstract=4053961 [last accessed use and depressive symptoms. Journal of Affective Disor-
16 February 2023]. ders, 275, 165–174.
Harrison, L., Carducci, B., Klein, J.D., & Bhutta, Z.A. (2022). Jackman, K., Kreuze, E.J., Caceres, B.A., & Schnall, R. (2020).
Indirect effects of COVID-19 on child and adolescent mental Bullying and peer victimization of minority youth: Intersec-
health: An overview of systematic reviews. BMJ Global tions of sexual identity and race/ethnicity. The Journal of
Health, 7, e010713. School Health, 90, 368–377.
Hartman-Munick, S.M., Lin, J.A., Milliren, C.E., Braverman, Jalal, H., Buchanich, J.M., Roberts, M.S., Balmert, L.C.,
P.K., Brigham, K.S., Fisher, M.M., . . . & Richmond, T.K. Zhang, K., & Burke, D.S. (2018). Changing dynamics of
(2022). Association of the COVID-19 pandemic with adoles- the drug overdose epidemic in the United States from 1979
cent and young adult eating disorder care volume. JAMA through 2016. Science, 361, eaau1184.
Pediatrics, 176, 1225–1232. Jorns-Presentati, A., Napp, A.K., Dessauvagie, A.S., Stein,
Hase, C.N., Goldberg, S.B., Smith, D., Stuck, A., & Campain, D.J., Jonker, D., Breet, E., . . . & Groen, G. (2021). The
J. (2015). Impacts of traditional bullying and cyberbullying prevalence of mental health problems in sub-Saharan
on the mental health of middle school and high school adolescents: A systematic review. PLoS One, 16, e0251689.
students. Psychology in the Schools, 52, 607–617. Kalb, L.G., Stapp, E.K., Ballard, E.D., Holingue, C., Keefer, A.,
Hawes, T., Zimmer-Gembeck, M.J., & Campbell, S.M. (2020). & Riley, A. (2019). Trends in psychiatric emergency depart-
Unique associations of social media use and online appear- ment visits among youth and young adults in the US.
ance preoccupation with depression, anxiety, and appear- Pediatrics, 143, e20182192.
ance rejection sensitivity. Body Image, 33, 66–76. Keles, B., McCrae, N., & Grealish, A. (2019). A systematic
Hern
an, M.A., & Robins, J.M. (2010). Causal inference. Boca review: The influence of social media on depression, anxiety
Raton, FL: CRC. and psychological distress in adolescents. International
Hernan, M.A., & Robins, J.M. (2010). Causal inference: What Journal of Adolescence and Youth, 25, 79–93.
if. Boca Raton, FL: CRC Press. Available from: https://www. Kelly, Y., Zilanawala, A., Booker, C., & Sacker, A. (2018). Social
taylorfrancis.com/books/mono/10.1201/9781315374 media use and adolescent mental health: Findings from the
932/causal-inference-miguel-hernan-james-robins [last UK millennium cohort study. E Clinical Medicine, 6, 59–68.
accessed 16 February 2023] Kennedy, R.S. (2019). Bullying trends in the United States: A
Herrman, H., Patel, V., Kieling, C., Berk, M., Buchweitz, C., meta-regression. Trauma, Violence, and Abuse, 22, 914–
Cuijpers, P., . . . & Wolpert, M. (2022). Time for united action 927.
on depression: A lancet–world psychiatric association com- Kessler, R.C., Birnbaum, H.G., Shahly, V., Bromet, E., Hwang,
mission. The Lancet, 399, 957–1022. I., McLaughlin, K.A., . . . & Stein, D.J. (2010). Age differences
Hickman, C., Marks, E., Pihkala, P., Clayton, S., Lewan- in the prevalence and comorbidity of DSM-IV major depres-
dowski, R.E., Mayall, E.E., . . . & van Susteren, L. (2021). sive episodes: Results from the WHO World Mental Health
Climate anxiety in children and young people and their Survey Initiative. Depression and Anxiety, 27, 351.
beliefs about government responses to climate change: A Kessler, R.C., & McLeod, J.D. (1984). Sex differences in
global survey. The Lancet Planetary Health, 5, e863–e873. vulnerability to undesirable life events. American Sociologi-
Hiilamo, A., Hiilamo, H., Ristikari, T., & Virtanen, P. (2021). cal Review, 49, 620.
Impact of the great recession on mental health, substance Kim, K.M., Kim, D., & Chung, U.S. (2020). Investigation of the
use and violence in families with children: A systematic trend in adolescent mental health and its related social

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 21

factors: A multi-year cross-sectional study for 13 years. McCrory, A., Best, P., & Maddock, A. (2020). The relationship
International Journal of Environmental Research and Public between highly visual social media and young people’s
Health, 17, 1–14. mental health: A scoping review. Children and Youth
Kim, Y., & Hagquist, C. (2018). Trends in adolescent mental Services Review, 115, 105053.
health during economic upturns and downturns: A multi- McGorry, P.D., Mei, C., Chanen, A., Hodges, C., Alvarez-
level analysis of Swedish data 1988–2008. Journal of Jimenez, M., & Killackey, E. (2022). Designing and scaling
Epidemiology and Community Health, 72, 101–108. up integrated youth mental health care. journal World
Klein, D.F. (1980). Anxiety reconceptualized. Comprehensive Psychiatry, 21(1), 61–76. https://doi.org/10.1002/wps.
Psychiatry, 21, 411–427. 20938
Koch, J.M., McLachlan, C., Victor, C.J., Westcott, J., & Yager, Meier, A., & Reinecke, L. (2020). Computer-mediated commu-
C. (2019). The cost of being transgender: Where socio- nication, social media, and mental health: A conceptual and
economic status, global health care systems, and gender empirical meta-review. Communication Research, 48(1),
identity intersect. Psychology & Sexuality, 11, 103–119. 1182–1209.
Kreski, N.T., Chen, Q., Olfson, M., Cerd
a, M., Hasin, D., Memon, A.M., Sharma, S.G., Mohite, S.S., & Jain, S. (2018).
Martins, S.S., & Keyes, K.M. (2022). Explaining US adoles- The role of online social networking on deliberate self-harm
cent depressive symptom trends through declines in reli- and suicidality in adolescents: A systematized review of
gious beliefs and service attendance. Journal of Religion and literature. Indian Journal of Psychiatry, 60, 384–392.
Health, 61, 300–326. Miech, R.A., Johnston, L.D., Patrick, M.E., O’malley, P.M.,
Kreski, N.T., Chen, Q., Olfson, M., Cerd
a, M., Martins, S.S., Bachman, J.G., & Schulenberg, J.E. (2019). Monitoring the
Mauro, P.M., . . . & Keyes, K.M. (2022). Experiences of online future national survey results on drug use, 1975–2018:
bullying and offline violence-related behaviors among a Overview, key findings on adolescent drug use. Sponsored
nationally representative sample of US adolescents, 2011 by The National Institute on Drug Abuse at The National
to 2019. The Journal of School Health, 92, 376–386. Institutes of Health. Available from: http://
Krieger, N. (2001). Theories for social epidemiology in the 21st monitoringthefuture.org/results/publications/ [last
century: An ecosocial perspective. International Journal of accessed 26 January 2023].
Epidemiology, 30, 668–677. Miranda-Mendizabal, A., Castellv
ı, P., Par
es-Badell, O., Alayo,
Lawlor, D.A., Tilling, K., & Smith, G.D. (2016). Triangulation in I., Almenara, J., Alonso, I., . . . & Alonso, J. (2019). Gender
aetiological epidemiology. International Journal of Epidemi- differences in suicidal behavior in adolescents and young
ology, 45, 1866–1886. adults: Systematic review and meta-analysis of longitudinal
Lebrun-Harris, L.A., Ghandour, R.M., Kogan, M.D., & Warren, studies. International Journal of Public Health, 64, 265–283.
M.D. (2022). Five-year trends in US children’s health and Miron, O., Yu, K.H., Wilf-Miron, R., & Kohane, I.S. (2019).
well-being, 2016–2020. JAMA Pediatrics, 176, e220056. Suicide rates among adolescents and young adults in the
Levey, D.F., Stein, M.B., Wendt, F.R., Pathak, G.A., Zhou, H., United States, 2000-2017. JAMA, 321, 2362.
Aslan, M., . . . & Gelernter, J. (2021). Bi-ancestral depression Moroney, E., Lee, S.S., Ebbert, A.M., & Luthar, S.S. (2022).
GWAS in the Million Veteran Program and meta-analysis in Digital media use frequency, online behaviors, and risk for
>1.2 million individuals highlight new therapeutic direc- negative mental health in adolescents in high-achieving
tions. Nature Neuroscience, 24, 954–963. schools, 28, 237–254.
Lin, J.A., Hartman-Munick, S.M., Kells, M.R., Milliren, C.E., Morris, J.N. (1975). Uses of epidemiology (3rd edn). Edinburgh:
Slater, W.A., Woods, E.R., . . . & Richmond, T.K. (2021). The Churchill Livingstone.
impact of the COVID-19 pandemic on the number of Moylan, A., Appelbaum, N., Clarke, J., Feather, C., Flahault
adolescents/young adults seeking eating disorder-related Tairraz, A., Maconochie, I., & Darzi, A. (2019). Suicidal
care. Journal of Adolescent Health, 69, 660–663. attempts and ideation among children and adolescents in
Luthar, S.S., Kumar, N.L., & Zillmer, N. (2020). High-achieving US Emergency Departments, 2007–2015. JAMA Pediatrics,
schools connote risks for adolescents: Problems documen- 173, 598–600.
ted, processes implicated, and directions for interventions. Munaf o, M.R., Tilling, K., Taylor, A.E., Evans, D.M., & Smith,
The American Psychologist, 75, 983–995. G.D. (2018). Collider scope: When selection bias can
Lynch, J., & Smith, G.D. (2005). A life course approach to substantially influence observed associations. International
chronic disease epidemiology. Annual Review of Public Journal of Epidemiology, 47, 226.
Health, 26, 1–35. Murray, S.B., Nagata, J.M., Griffiths, S., Calzo, J.P., Brown,
March-Llanes, J., Marqu
es-Feixa, L., Mezquita, L., Fa~nan
as, T.A., Mitchison, D., . . . & Mond, J.M. (2017). The enigma of
L., & Moya-Higueras, J. (2017). Stressful life events during male eating disorders: A critical review and synthesis.
adolescence and risk for externalizing and internalizing Clinical Psychology Review, 57, 1–11.
psychopathology: A meta-analysis. European Child and Nesi, J., & Prinstein, M.J. (2015). Using social media for social
Adolescent Psychiatry, 26, 1409–1422. comparison and feedback-seeking: Gender and popularity
Marciano, L., Driver, C.C., Schulz, P.J., & Camerini, A.L. moderate associations with depressive symptoms. Journal of
(2022). Dynamics of adolescents’ smartphone use and well- Abnormal Child Psychology, 43, 1427–1438.
being are positive but ephemeral. Scientific Reports, 12, 1– Neumann, A., Nolte, I.M., Pappa, I., Ahluwalia, T.S., Petters-
15. son, E., Rodriguez, A., . . . & Tiemeier, H. (2022). A genome-
Mathur, M.B., & VanderWeele, T.J. (2019). Finding common wide association study of total child psychiatric problems
ground in meta-analysis “wars” on violent video games. scores. PLoS One, 17, e0273116.
Perspectives on Psychological Science, 14, 705–708. Nobari, T.Z., Whaley, S.E., Crespi, C.M., Prelip, M.L., & Wang,
McCann, E., Donohue, G., & Brown, M. (2021). Experiences M.C. (2018). Widening socio-economic disparities in early
and perceptions of trans and gender non-binary people childhood obesity in Los Angeles County after the Great
regarding their psychosocial support needs: A systematic Recession. Public Health Nutrition, 21, 1–10.
review of the qualitative research evidence. International Nolen-Hoeksema, S., Larson, J., & Grayson, C. (1999).
Journal of Environmental Research and Public Health, 18, Explaining the gender difference in depressive symptoms.
3403. Journal of Personality and Social Psychology, 77, 1061–
McCrae, N., Gettings, S., & Purssell, E. (2017). Social media 1072.
and depressive symptoms in childhood and adolescence: A Nottelmann, E.D., Inoff-Germain, G., Susman, E.J., & Chrou-
systematic review. Adolescent Research Review, 2, 315–330. sos, G.P. (1990). Hormones and behavior at puberty. In

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
22 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

Adolescence and puberty. The Kinsey Institute series (Vol. 3, Prescott, A.T., Sargent, J.D., & Hull, J.G. (2018). Metaanalysis
pp. 88–123). New York: Oxford University Press. of the relationship between violent video game play and
NPR. (2022). The teen mental health crisis. Available from: physical aggression over time. Proceedings of the National
https://www.npr.org/2022/12/19/1144124269/the-teen- Academy of Sciences of the United States of America, 115,
mental-health-crisis [last accessed 19 February 2023]. 9882–9888.
Odgers, C.L. (2015). Income inequality and the developing Przybylski, A.K., & Weinstein, N. (2017). A large-scale test of
child: Is it all relative? The American Psychologist, 70, 722. the goldilocks hypothesis: Quantifying the relations between
Odgers, C.L., & Jensen, M.R. (2020a). Adolescent development digital-screen use and the mental well-being of adolescents.
and growing divides in the digital age. Dialogues in Clinical Psychological Science, 28, 204–215.
Neuroscience, 22, 143. Rabkin, J.G., & Struening, E.L. (1976). Live events, stress, and
Odgers, C.L., & Jensen, M.R. (2020b). Annual Research illness. Science, 194, 1013–1020.
Review: Adolescent mental health in the digital age: Facts, Racine, N., McArthur, B.A., Cooke, J.E., Eirich, R., Zhu, J., &
fears, and future directions. Journal of Child Psychology and Madigan, S. (2021). Global prevalence of depressive and
Psychiatry, 61, 336–348. anxiety symptoms in children and adolescents during
Ogburn, E.L., & VanderWeele, T.J. (2012). On the nondiffer- COVID-19: A meta-analysis. JAMA Pediatrics, 175, 1142–
ential misclassification of a binary confounder. Epidemiol- 1150.
ogy, 23, 433. Radtke, T., Apel, T., Schenkel, K., Keller, J., & von Lindern, E.
Ohlsson, C., Bygdell, M., Celind, J., Sond
en, A., Tidblad, A., (2022). Digital detox: An effective solution in the smartphone
S€ avendahl, L., & Kindblom, J.M. (2019). Secular trends in era? A systematic literature review. Mobile Media and
pubertal growth acceleration in Swedish boys born from Communication, 10, 190–215.
1947 to 1996. JAMA Pediatrics, 173, 860–865. Rathmann, K., Pf€ ortner, T.K., Hurrelmann, K., Osorio, A.M.,
Ong, K.K., Ahmed, M.L., & Dunger, D.B. (2006). Lessons from Bosakova, L., Elgar, F.J., & Richter, M. (2016). The great
large population studies on timing and tempo of puberty recession, youth unemployment and inequalities in psycho-
(secular trends and relation to body size): The European logical health complaints in adolescents: A multilevel study
trend. Molecular and Cellular Endocrinology, 254–255, 8–12. in 31 countries. International Journal of Public Health, 61,
Orben, A. (2020). Teenagers, screens and social media: A 809–819.
narrative review of reviews and key studies. Social Psychi- Reisner, S.L., Poteat, T., Keatley, J.A., Cabral, M., Mothopeng,
atry and Psychiatric Epidemiology, 55, 407–414. T., Dunham, E., . . . & Baral, S.D. (2016). Global health
Orben, A., & Przybylski, A.K. (2019). The association between burden and needs of transgender populations: A review. The
adolescent well-being and digital technology use. Nature Lancet, 388, 412–436.
Human Behaviour, 3, 173–182. Reiss, F. (2013). Socioeconomic inequalities and mental health
Ortiz, P., & Khin Khin, E. (2018). Traditional and new media’s problems in children and adolescents: A systematic review.
influence on suicidal behavior and contagion. Behavioral Social Science & Medicine, 90, 24–31.
Sciences & the Law, 36, 245–256. Risman, B.J., & Davis, G. (2013). From sex roles to gender
Panova, T., & Carbonell, X. (2018). Is smartphone addiction structure. Current Sociology, 61, 733–755.
really an addiction? Journal of Behavioral Addictions, 7, 252. Robinson, A., Bonnette, A., Howard, K., Ceballos, N., Dailey,
Parent, A.S., Teilmann, G., Juul, A., Skakkebaek, N.E., S., Lu, Y., & Grimes, T. (2019). Social comparisons, social
Toppari, J., & Bourguignon, J.P. (2003). The timing of media addiction, and social interaction: An examination of
normal puberty and the age limits of sexual precocity: specific social media behaviors related to major depressive
Variations around the world, secular trends, and changes disorder in a millennial population. Journal of Applied
after migration. Endocrine Reviews, 24, 668–693. Biobehavioral Research, 24, e12158.
Parodi, K.B., Holt, M.K., Green, J.G., Porche, M.V., Koenig, B., Roh, B.R., Jung, E.H., & Hong, H.J. (2018). A comparative
& Xuan, Z. (2022). Time trends and disparities in anxiety study of suicide rates among 10–19-year-olds in 29 OECD
among adolescents, 2012–2018. Social Psychiatry and countries. Psychiatry Investigation, 15, 376.
Psychiatric Epidemiology, 57, 127. Rose, G. (2001). Sick individuals and sick populations.
Passari, E. (2020). The Great Recession and the rise of International Journal of Epidemiology, 30, 427–432.
populism. Intereconomics, 55, 17–21. Ross, C.E., & Mirowsky, J. (2006). Sex differences in the effect
Patterson, M.W., Mann, F.D., Grotzinger, A.D., Tackett, J.L., of education on depression: Resource multiplication or
Tucker-Drob, E.M., & Harden, K.P. (2018). Genetic and resource substitution? Social Science & Medicine, 63,
environmental influences on internalizing psychopathology 1400–1413.
across age and pubertal development. Developmental Psy- Rothman, K.J. (1976). Causes. American Journal of Epidemi-
chology, 54, 1928. ology, 104, 587–592.
Pickett, W., Molcho, M., Elgar, F.J., Brooks, F., De Looze, M., Rubinow, D.R., & Schmidt, P.J. (1996). Androgens, brain, and
Rathmann, K., . . . & Currie, C. (2013). Trends and socioeco- behavior. The American Journal of Psychiatry, 153, 974–
nomic correlates of adolescent physical fighting in 30 984.
countries. Pediatrics, 131, e18–e26. Ruby, D. (2023). Social media users – How many people use
Pine, R., Fleming, T., McCallum, S., & Sutcliffe, K. (2020). The social media in 2023. Available from: https://www.
effects of casual videogames on anxiety, depression, stress, demandsage.com/social-media-users/ [last accessed 26
and low mood: A systematic review. Games for Health January 2023].
Journal, 9, 255–264. Ruscio, A.M., Hallion, L.S., Lim, C.C.W., Aguilar-Gaxiola, S.,
Pirkis, J., Gunnell, D., Shin, S., del Pozo-Banos, M., Arya, V., Al-Hamzawi, A., Alonso, J., . . . & Scott, K.M. (2017). Cross-
Aguilar, P.A., . . . & Spittal, M.J. (2022). Suicide numbers sectional comparison of the epidemiology of DSM-5 gener-
during the first 9–15 months of the COVID-19 pandemic alized anxiety disorder across the globe. JAMA Psychiatry,
compared with pre-existing trends: An interrupted time 74, 465–475.
series analysis in 33 countries. eClinicalMedicine, 51, Rutter, M. (1981). Stress, coping and development: Some
101573. issues and some questions. Journal of Child Psychology and
Platt, J.M., Bates, L., Jager, J., McLaughlin, K.A., & Keyes, Psychiatry, 22, 323–356.
K.M. (2021). Is the US gender gap in depression changing Rutter, M., & Sroufe, L.A. (2000). Developmental psychopa-
over time? A meta-regression. American Journal of Epidemi- thology: Concepts and challenges. Development and Psycho-
ology, 190, 1190. pathology, 12, 265–269.

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
doi:10.1111/jcpp.13864 Sex differences in internalizing conditions in adolescents 23

Saiphoo, A.N., & Vahedi, Z. (2019). A meta-analytic review of Sullivan, P.F., Neale, M.C., & Kendler, K.S. (2000).
the relationship between social media use and body image Genetic epidemiology of major depression: Review and
disturbance. Computers in Human Behavior, 101, 259–275. meta-analysis. American Journal of Psychiatry, 157, 1552–
Salk, R.H., Hyde, J.S., & Abramson, L.Y. (2017). Gender 1562.
differences in depression in representative national samples: Sun, Y., & Zhang, Y. (2021). A review of theories and models
Meta-analyses of diagnoses and symptoms. Psychological applied in studies of social media addiction and implications
Bulletin, 143, 783. for future research. Addictive Behaviors, 114, 106699.
Sanyaolu, A., Okorie, C., Qi, X., Locke, J., & Rehman, S. Susman, E.J., Nottelmann, E.D., Inoff-Germain, G.E., Dorn,
(2019). Childhood and adolescent obesity in the United L.D., Cutler, G.B., Loriaux, D.L., & Chrousos, G.P. (1985).
States: A public health concern. Global Pediatric Health, 6, The relation of relative hormonal levels and physical
1–11. development and social-emotional behavior in young ado-
Sedgwick, R., Epstein, S., Dutta, R., & Ougrin, D. (2019). lescents. Journal of Youth and Adolescence, 14, 245–264.
Social media, internet use and suicide attempts in adoles- Sweeting, H., West, P., Young, R., & Der, G. (2010). Can we
cents. Current Opinion in Psychiatry, 32, 534–541. explain increases in young people’s psychological distress
Seedat, S., Scott, K.M., Angermeyer, M.C., Berglund, P., over time? Social Science & Medicine, 71, 1819.
Bromet, E.J., Brugha, T.S., . . . & Kessler, R.C. (2009). Sweeting, H., Young, R., & West, P. (2009). GHQ increases
Cross-National Associations between gender and mental among Scottish 15 year olds 1987–2006. Social Psychiatry
disorders in the World Health Organization World Mental and Psychiatric Epidemiology, 44, 579.
Health Surveys. Archives of General Psychiatry, 66, 785– Tang, S., Werner-Seidler, A., Torok, M., Mackinnon, A.J., &
795. Christensen, H. (2021). The relationship between screen
Sevar, K., Vythilingum, B., & Castle, D. (2015). Anxiety and time and mental health in young people: A systematic review
related disorders in women. In D. J. Stein . . . B. Vythilingum of longitudinal studies. Clinical Psychology Review, 86,
(Eds.), Anxiety disorders and gender (pp. 137–154). Cham, 102021.
Switzerland: Springer International Publishing/Springer Tolin, D.F., & Foa, E.B. (2006). Sex differences in trauma and
Nature. https://doi.org/10.1007/978-3-319-13060-6_7 posttraumatic stress disorder: A quantitative review of
Shorey, S., Ng, E.D., & Wong, C.H.J. (2022). Global prevalence 25 years of research. Psychological Bulletin, 132, 959–992.
of depression and elevated depressive symptoms among Twenge, J.M. (2020). Increases in depression, self-harm, and
adolescents: A systematic review and meta-analysis. British suicide among U.S. adolescents after 2012 and links to
Journal of Clinical Psychology, 61, 287–305. technology use: Possible mechanisms. Psychiatric Research
Sil

en, Y., & Keski-Rahkonen, A. (2022). Worldwide prevalence and Clinical Practice, 2, 19–25.
of DSM-5 eating disorders among young people. Current Twenge, J.M., Martin, G.N., & Spitzberg, B.H. (2018). Trends in
Opinion in Psychiatry, 35, 362–371. U.S. Adolescents’ Media use, 1976–2016: The rise of digital
Slade, T., Chapman, C., Swift, W., Keyes, K., Tonks, Z., & media, the decline of TV, and the (near) demise of print.
Teesson, M. (2016). Birth cohort trends in the global Psychology of Popular Media Culture, 8, 329–345.
epidemiology of alcohol use and alcohol-related harms in Twenge, J.M., & Park, H. (2019). The decline in adult activities
men and women: Systematic review and metaregression. among U.S. adolescents, 1976–2016. Child Development,
BMJ Open, 6, e011827. 90, 638–654.
Smith, P.K., L
opez-Castro, L., Robinson, S., & G€ orzig, A. Ullsperger, J.M., & Nikolas, M.A. (2017). A meta-analytic
(2019). Consistency of gender differences in bullying in review of the association between pubertal timing and
cross-cultural surveys. Aggression and Violent Behavior, 45, psychopathology in adolescence: Are there sex differences
33–40. in risk? Psychological Bulletin, 143, 903–938.
Sohn, S., Rees, P., Wildridge, B., Kalk, N.J., & Carter, B. UNHCR. (2021). Global trends: Forced displacement in 2020.
(2019). Prevalence of problematic smartphone usage and Geneva: United Nations High Commissioner for Refugees.
associated mental health outcomes amongst children and Available from: https://www.unhcr.org/statistics/
young people: A systematic review, meta-analysis and unhcrstats/60b638e37/global-trends-forced-displacement-
GRADE of the evidence. BMC Psychiatry, 19, 1–10. 2020.html [last accessed 25 May 2023].
Sørensen, K., Aksglaede, L., Petersen, J.H., & Juul, A. (2010). Valkenburg, P.M., Meier, A., & Beyens, I. (2022). Social media
Recent changes in pubertal timing in healthy Danish boys: use and its impact on adolescent mental health: An umbrella
Associations with body mass index. The Journal of Clinical review of the evidence. Current Opinion in Psychology, 44,
Endocrinology & Metabolism, 95, 263–270. 58–68.
Sørensen, K., Mouritsen, A., Aksglaede, L., Hagen, C.P., Vandenbosch, L., Fardouly, J., & Tiggemann, M. (2022). Social
Mogensen, S.S., & Juul, A. (2012). Recent secular trends media and body image: Recent trends and future directions.
in pubertal timing: Implications for evaluation and diagnosis Current Opinion in Psychology, 45, 101289.
of precocious puberty. Hormone Research in Pædiatrics, 77, VanderWeele, T.J. (2009). Concerning the consistency
137–145. assumption in causal inference. Epidemiology, 20, 880–883.
Starting Over. (2021). Michael Ohlrogge tracks post- VanderWeele, T.J., & Richardson, T.S. (2012). General theory
foreclosure outcomes during the Great Recession. NYU for interactions in sufficient cause models with dichotomous
School of Law. Available from: https://www.law.nyu.edu/ exposures. Annals of Statistics, 40, 2128.
news/ideas/michael-ohlrogge-great-recession-foreclosures Verduyn, P., Ybarra, O., R
esibois, M., Jonides, J., & Kross, E.
[last accessed 26 January 2023]. (2017). Do social network sites enhance or undermine
Steers, M.L.N. (2016). ‘It’s complicated’: Facebook’s relation- subjective well-being? A critical review. Social Issues and
ship with the need to belong and depression. Current Policy Review, 11, 274–302.
Opinion in Psychology, 9, 22–26. Vicary, J.R., Klingaman, L.R., & Harkness, W.L. (1995). Risk
Stice, E., Presnell, K., & Bearman, S.K. (2001). Relation of early factors associated with date rape and sexual assault of
menarche to depression, eating disorders, substance abuse, adolescent girls. Journal of Adolescence, 18, 289–306.
and comorbid psychopathology among adolescent girls. Vidal, C., Lhaksampa, T., Miller, L., & Platt, R. (2020). Social
Developmental Psychology, 37, 608–619. media use and depression in adolescents: A scoping review.
Sullivan, P.F., Daly, M.J., Ripke, S., Lewis, C.M., Lin, D.Y., International Review of Psychiatry, 32, 235–253.
Wray, N.R., . . . & Sullivan, P.F. (2013). A mega-analysis of Vogels, E.A., Gelles-Watnick, R., & Massarat, N. (2022). Teens,
genome-wide association studies for major depressive dis- social media and technology 2022. Pew Research Center.
order. Molecular Psychiatry, 18, 497–511. Available from: https://www.pewresearch.org/internet/

Ó 2023 Association for Child and Adolescent Mental Health.

 14697610, 0, Downloaded from https://acamh.onlinelibrary.wiley.com/doi/10.1111/jcpp.13864 by University Of Iowa, Wiley Online Library on [07/01/2024]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License
24 Katherine M. Keyes and Jonathan M. Platt J Child Psychol Psychiatr 2023; 0(0): 1–24

2022/08/10/teens-social-media-and-technology-2022/ guided by ideation-to-action theory. Clinical Psychology

[last accessed 26 January 2023]. Review, 63, 93–105.
Vuorre, M., Orben, A., & Przybylski, A.K. (2021). There is No Wray, N.R., Ripke, S., Mattheisen, M., Trzaskowski, M., Byrne,
evidence that associations between Adolescents’ digital E.M., Abdellaoui, A., . . . & Major Depressive Disorder
technology engagement and mental health problems have Working Group of the Psychiatric Genomics Consortium.
increased. Clinical Psychological Science, 9, 823–835. (2018). Genome-wide association analyses identify 44 risk
Ward, J.B., Gartner, D.R., Keyes, K.M., Fliss, M.D., Mcclure, variants and refine the genetic architecture of major
E.S., & Robinson, W.R. (2019). How do we assess a racial depression. Nature Genetics, 50, 668–681.
disparity in health? Distribution, interaction, and interpre- Wray-Lake, L., Schulenberg, J., Keyes, K.M., & Shubert, J.
tation in epidemiological studies HHS public access. Annals (2017). The developmental course of community service
of Epidemiology, 29, 1–7. across the transition to adulthood in a national U.S. sample.
Werner-Seidler, A., Spanos, S., Calear, A.L., Perry, Y., Torok, Developmental Psychology, 53, 2397–2408.
M., O’Dea, B., . . . & Newby, J.M. (2021). School-based Xiao, Y., Brown, T.T., Snowden, L.R., Julian Chow, C.-C., &
depression and anxiety prevention programs: An updated John, M.J. (2023). COVID-19 policies, pandemic disrup-
systematic review and meta-analysis. Clinical Psychology tions, and changes in child mental health and sleep in the
Review, 89, 102079. United States. JAMA Network Open, 6, e232716.
WHO. (2021a). Depression. Geneva, Switzerland: World Health Young, E., & Korszun, A. (2009). Sex, trauma, stress hormones
Organization. Available from: https://www.who.int/news- and depression. Molecular Psychiatry, 15, 23–28.
room/fact-sheets/detail/depression [last accessed 16 Feb- Zabelski, S., Kaniuka, A.R., Robertson, A., & Cramer, R.J.
ruary 2023]. (2022). Crisis lines: Current status and recommendations
WHO. (2021b). Suicide. Geneva, Switzerland: World Health for research and policy. Psychiatric Services, 74, 505–512.
Organization. Available from https://www.who.int/news- Zhao, L., Han, G., Zhao, Y., Jin, Y., Ge, T., Yang, W., . . . & Li, B.
room/fact-sheets/detail/suicide [last accessed 21 February (2020). Gender differences in depression: Evidence from
2023]. genetics. Frontiers in Genetics, 11, 562316.
Wills, T.A., & Filer, M. (1996). Stress—Coping model of Zlan, E., & Debeuf, K. (2020). 2008–2009: The years that
adolescent substance use. In Advances in clinical child almost broke the euro. Available from: https://euobserver.
psychology (pp. 91–132). Boston, MA: Springer. Available com/20th-anniversary/150087 [last accessed 26 January
from: https://link.springer.com/chapter/10.1007/978-1- 2023].
4613-0323-7_3 [last accessed 26 January 2023].
Wolford-Clevenger, C., Frantell, K., Smith, P.N., Flores, L.Y., & Accepted for publication: 7 June 2023
Stuart, G.L. (2018). Correlates of suicide ideation and
behaviors among transgender people: A systematic review

Ó 2023 Association for Child and Adolescent Mental Health.