Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Global variation in the prevalence of suicidal ideation, anxiety and their
correlates among adolescents: A population based study of 82 countries
Tuhin Biswas
a,b,
*, James G. Scott
c,d,e,f
, Kerim Munir
g
, Andre M.N. Renzaho
h
, Lal B. Rawal
h,i
,
Janeen Baxter
a,b
, Abdullah A. Mamun
a,b
a
Institute for Social Science Research, The University of Queensland, Brisbane, Australia
b
ARC Centre of Excellence for Children and Families over the Life Course, The University of Queensland, Brisbane, Australia
c
QIMR Berghofer Medical Research Institute, Herston, Qld, 4006, Australia
d
Centre for Clinical Research, Faculty of Medicine, The University of Queensland, Herston, QLD, Australia
e
Metro North Mental Health, Royal Brisbane and Women’s Hospital, Herston, Brisbane, QLD, Australia
f
Queensland Centre for Mental Health Research, Wacol, Brisbane, QLD, Australia
g
Developmental Medicine Center, Boston Children’s Hospital, Harvard Medical School, Boston, MA, United States
h
School of Social Sciences and Psychology, Western Sydney University, Penrith 2751 NSW, Australia
i
School of Health, Medical and Applied Sciences, Central Queensland University, Australia
ARTICLE INFO
Article History:
Received 8 April 2020
Revised 12 May 2020
Accepted 12 May 2020
Available online xxx
ABSTRACT
Background: Suicidal ideation and anxiety are common among adolescents although their prevalence has
predominantly been studied in high income countries. This study estimated the population prevalence of sui-
cidal ideation and anxiety and their correlates with peer support, parent-adolescent relationship, peer vic-
timization, conflict, isolation and loneliness across a range of low-income, lower-middle-income, upper-
middle-income countries and high-income countries (LMICHICs).
Methods: Data were drawn from the Global School-based Student Health Survey (GSHS) of adolescents aged
1217 years between 2003 and 2015 in 82 LM-HICs from the six World Health Organization (WHO) regions.
For those countries with repeated time point data in this study, we used data from the most recent survey.
We estimated weighted prevalence of suicidal ideation and anxiety by country, region and at a global level
with the following questions:-“Did you ever seriously consider attempting suicide during the past 12 months?”
and “During the past 12 months, how often have you been so worried about something that you could not sleep at
night?”. We used multiple binary logistic regression to estimate the adjusted association between adolescent
age, sex, socioeconomic status, peer support, parent-adolescent relationship, peer victimization, conflict, iso-
lation and loneliness with suicidal ideation and anxiety.
Findings: The sample comprised of 275,057 adolescents aged 1217 years (mean age was 14.6 (SD 1.18) years
of whom 51.8% were females). The overall 12 months pooled prevalence of suicidal ideation and anxiety
were 14.0% (95% CI 10.017.0%) and 9.0% (7.012.0%) respectively. The highest pooled prevalence of suicidal
ideation was observed in the Africa Region (21.0%; 20.021.0%) and the lowest was in the Asia region (8.0%,
8.09.0%). For anxiety, the highest pooled prevalence was observed in Eastern Mediterranean Region (17.0%,
16.017.0%) the lowest was in the European Region (4.0%, 4.05.0%). Being female, older age, having a lower
socioeconomic status and having no close friends were associated with a greater risk of suicidal ideation and
anxiety. A higher levels of parental control was positively associated with a greater likelihood of experiencing
suicidal ideation (OR: 1.65, 1.451.87) and anxiety (1.53, 1.301.80). Parental understanding and monitoring
were negatively associated with mental health problems. Similarly, the odds of experiencing suicidal ideation
and anxiety were higher among adolescents who had been experiencing peer conflict (1.36, 1.241.50; 1.54,
1.401.70), peer victimization (1.26, 1.151.38; 1.13, 1.021.26), peer isolation (1.69, 1.531.86; 1.76,
1.611.92) and reported loneliness (2.56, 2.332.82; 5.63, 5.216.08).
Interpretations: Suicidal ideation and anxiety are prevalent among adolescents although there is significant
global variation. Parental and peer supports are protective factors against suicidal ideation and anxiety. Peer
based interventions to enhance social connectedness and parent skills training to improve parent-child
* Corresponding author: Tuhin Biswas, Institute for Social Science Research, The Uni-
versity of Queensland, 80 Meiers Road, Indooroopilly, Queensland 4068, Australia,
Telephone: +61 470216336.
E-mail address: t.biswas@uqconnect.edu.au (T. Biswas).
https://doi.org/10.1016/j.eclinm.2020.100395
2589-5370/© 2020 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license. (http://creativecommons.org/licenses/by-nc-nd/4.0/)
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
EClinicalMedicine 000 (2020) 100395
Contents lists available at ScienceDirect
EClinicalMedicine
journal homepage: https://www.journals.elsevier.com/eclinicalmedicine
relationships may reduce suicidal ideation and anxiety. Research to inform the factors that influence country
and regional level differences in adolescent mental health problems may inform preventative strategies.
Funding: None
© 2020 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license.
(http://creativecommons.org/licenses/by-nc-nd/4.0/)
1. Introduction
Adolescents represent almost a quarter of the world’spopulation[1]
and are projected to proportionally increase in coming decades [2].Ado-
lescence is a pivotal developmental stage where improvements in health
and education can establish an improved quality of lifetime health trajec-
tories [3,4]. Nearly half of the global burden of disease has its origins in
adolescence [5] and mental disorders account for 16% of the global bur-
den of disease and injury in people aged 1019 years [6,7]. In terms of
disability-adjusted life years (DALYs) mental and substance use disorders
ranked 6th with 55.5 million DALYs and rise to 5th when mortality
burden of suicide is attributed [6]. Data informing country and regional
variation in prevalence estimates of mental health problems in adoles-
cents are needed to identify potential modifiable risk factors and to
inform differences in service needs globally.
Positive relationships with family and friends are consistently
linked with improved mental health and wellbeing during adoles-
cence [810]. For example, a study in Vietnam reported that greater
parental understanding and monitoring was significantly associated
with reduced risk of mental health problems. By contrast, greater
parental control was significantly associated with greater likelihood
of having suicidal ideation [11]. Similarly, positive peer support is
important to the mental wellbeing of adolescents. Supportive peer
relations following aversive peer experiences reduces the risk of
internalizing and externalizing symptoms [12].
Globally 1020% of adolescents have mental health problems and
adolescents from low-resource settings are particularly vulnerable to
mental ill-health [13]. A systematic review conducted in 27 countries
including every world region reported that the worldwide prevalence of
mental disorders was 13.4% (95% CIs 11.315.9) with the prevalence of
any anxiety disorder reported as 6.5% (4.79.1) [14]. A recent study
reported that more than three-quarters of suicides globally occur in low-
income and middle-income countries (LMICs) [15].However,inLMICs
mental disorders still remain under-reported because of social stigma,
religious or cultural taboos and inadequate reporting systems [16].Some
country-specific data are available, for example a study from Vietnam
showed that more than 30% of adolescents self-reported life time experi-
ences of low mood, and the prevalence rates of suicidal behaviours were
12.2% [17]. Another study in Pakistan reported that prevalence of anxiety
and suicidal ideation were 8.4% and 7.3%, respectively. To date, there are
a paucity of studies examining the prevalence of suicidal ideation and
anxiety among adolescents in low- and middle-income country and
comparing to high income country (LMICHIC) settings. Uddin et al.
(2019) used the Global School-based student Health Survey (GSHS) data
to examine suicidal ideation among adolescents in 59 LMICs without
comparison with HICs. This study did not examine the correlates of sui-
cidal ideation [15]. In the current study we aimed to estimate the preva-
lence of suicidal ideation and anxiety and examine their correlates
among adolescents aged 12 to 17 years across 82 countries with varying
income levels from different world regions.
2. Methods
2.1. Data sources
Data were from the Global School-based Student Health Survey
(GSHS), which commenced in 2003. The GSHS was jointly developed
by the WHO and the United States centre for Disease Control and Pre-
vention (CDC) in collaboration with The United Nations International
Children's Fund (UNICEF), The United Nations Educational, Scientific
and Cultural organisation (UNESCO), and The Joint United Nations
Programme on HIV and AIDS (UNAIDS). GSHS is administered to ado-
lescents aged 1217 years to capture information on a wide range of
health indicators using validated items from ten core modules includ-
ing: nutrition, physical activity, hygiene, mental health, alcohol use,
tobacco use, drug use, sexual behaviors, violence/injury, and protec-
tive factors [18]. In collecting this information, GSHS employed a
two-stage cluster sampling technique. In the first stage, the schools
were selected randomly. Classes that provided a representative sam-
ple of the general population aged 1217 years were selected within
Research in context
Evidence before this study
We systematically searched PubMed, EMBASE, PsycNIFO with a
combination of MeSH heading terms and keywords. The key
words used in the search (“suicide”OR “self-harm”OR “anxi-
ety”) and (“adolescents”OR “child*”OR “teenager”OR “youth”)
and (“developing country”OR “low socioeconomic status”OR
“low income country”OR “middle income country”OR “low-
and middle-income country”OR “high income country”OR
“low and middle income to high income countries”OR
“LMICHICs”OR “LMICs”). The literature search was conducted
up to August 27, 2019. We identified only one multicounty
study which used data from the Global School-based Students
Health Survey (GSHS) from 59 countries. We did not find any
comparative study of the prevalence of suicidal ideation, anxi-
ety among adolescents across low and middle to high income
countries (LMICHICs). In addition, no prior study has evalu-
ated the relationship between suicidal ideation and anxiety
with peer support, parent-adolescent relationship, peer victim-
ization, conflict, isolation and loneliness.
Added value of this study
This is the first study to comprehensively estimate the popula-
tion prevalence of suicidal ideation, anxiety and its relationship
with peer and parental support among adolescents across
LMIC
HICs. We used data from the GSHS of adolescents, ages
1217 years, in 82 LMICHICs in the six World Bank regions
to show the geographic variation in prevalence of suicidal idea-
tion, anxiety victimization in 82 LMICHICs. We demonstrated
that parental involvement and peer support were strongly
associated with reduced levels of suicidal ideation and anxiety.
Implications of all the available evidence
This study shows a large proportion of adolescents in all coun-
tries irrespective of income status experience suicidal thinking
and anxiety although there is a high variation between coun-
tries. In every country, those adolescents with lower levels of
peer and parental support and higher levels of parental control
were more likely to report experiencing suicidal ideation and
anxiety. Adolescent suicidal ideation and anxiety prevention
strategies should address family and peer relationships which
are socioculturally specific and sensitive.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
2T. Biswas et al. / EClinicalMedicine 00 (2020) 100395
the schools at the second stage of sampling [19]. The study design
and selection procedure of the participants were similar across the
GSHS countries. For this study we included data from 82 LMICHICs
from inception year 2003 to 2015. Only 10 countries had two-time
point’s data and that information presented in supplementary table-
2. For those countries with repeated time point data in this study, we
used data from the most recent survey.
2.1.1. Ethics statement
In each of the participating countries, the GSHS received ethics
approval from the Ministry of Education or a relevant institutional
ethics review committee, or both. Only those adolescents and their
parents or guardians who provided written or verbal consent partici-
pated. As the current study used retrospective publicly available data,
we did not require ethics approval.
2.2. Suicidal ideation and anxiety among adolescents
Suicidal ideation was examined with the question “Did you ever
seriously consider attempting suicide during the past 12 months?”with
a binary response of “yes”or “no.”
Anxiety was assessed with the question “During the past 12
months, how often have you been so worried about something that you
could not sleep at night?”With response ‘never’,‘rarely’,‘sometimes’,
‘most of the time’and ‘always’.
Anxiety responses were recoded and classified as ‘yes’,with
endorsement of ‘most of the time/always’,and ‘no’,which included
‘never’,‘rarely’,and ‘sometimes’.
2.3. Peer support
Peer support was assessed using a proxy variable based on the
question “During the past 30 days, how often were most of the stu-
dents in your school kind and helpful?”to which students could
respond ‘never’,‘rarely’,‘sometimes’,‘most of the time’or ‘always’.
Responses were recoded as never/rarely, sometimes, most of the
times and always.
2.4. Parentadolescent relationship
Parental understanding, parental monitoring and parental control are
components of the relationship between the adolescent and their
parents as perceived by the adolescent [20]. These were assessed with
three questions: i) parental understanding of student’sproblems(“Dur-
ing the past 30 days, how often did your parents or guardians understand
your problems and worries?”), ii) parental monitoring of student’s activi-
ties during their free time (“During the past 30 days, how often did your
parents or guardians really know what you were doing with your free
time?”). iii) Parental control was examined with the question “How often
did your parents or guardians go through your things without your approval
during the past 30 days?”Possible response options to each of these ques-
tions were ‘never’,‘rarely’,‘sometimes’,‘most of the time’and ‘always’.
These variables were recoded and classified as never/rarely, sometimes,
mostofthetimeandalways.
2.5. Socio-demographic factors
The sex and age of the participants were included in the survey.
Participants were categorized into three age groups: 1213 years,
1415 years and 1617 years. Socioeconomic status (SES) was mea-
sured by the variable, “During the past 30 days, how often did you go
hungry because there was not enough food in your home?”
Responses of “never to rarely”were recoded as ‘average’, and “some-
times to always”as ‘below average’SES [21]. We included country
policy or plan for mental health as reported by the World Health
Organization.
2.6. Peer victimization, conflict and loneliness
Conflict was assessed with two questions “During the past 12 months,
how many times you were physically attacked”and “During the past 12
months, how many times were you in a physical fight?”.Studentresponses
for conflict were recoded as ‘yes’(reported being attacked or fighting one
or more times) or ‘no’. Peer victimization was assessed with the question
“How many days where you bullied during the past 30 days?”and the
response was recoded as “yes”foranswerofoneormoredaysor“no”.
Loneliness was examined using the question “How often have you
felt lonely during the past 12 months?”with responses ranging from
“never”to “always.”The responses were dichotomized to “lonely”
where the student endorsed “most of the time”or “always”or “no”
meaning “never/rarely or sometimes”.
2.7. Statistical analysis
The data were weighted by primary sampling unit (PSU). The
weighting is based on the PSU which is derived from the probability
of a school being selected, a classroom being selected, school and stu-
dent level non-response and gender. Therefore, the samples were
nationally representative in respect to the study population. This
included using strata and primary sampling units at the country level.
Weighted mean estimates of prevalence (with corresponding 95%
CIs) were calculated by country and sex. We used random-effects
meta-analysis to generate regional and overall pooled estimates of
suicide and anxiety data, using the DerSimonian and Laird inverse-
variance method. Forest plots show the prevalence of each indicator
by country and its corresponding weight, and the pooled prevalence
by region and its associated 95% confidence intervals (CI). Heteroge-
neity was examined using the I
2
-statistic and a high level of inconsis-
tency (I
2
>50%) warranted the use of a random effect.
Bivariate analysis was performed to calculate the prevalence of
suicidal ideation and anxiety over background characteristics at the
global and regional level. We conducted binary logistic regression
analysis to examine the associations between suicidal ideation and
anxiety with peer support, parent-adolescent relationship, peer vic-
timization, conflict, isolation and loneliness. We conducted multiple
regression models which included all significant variables binary
logistic regression and survey year to explore independent factors
associated with suicidal ideation and anxiety. Variations due to clus-
tering were controlled in all analyses.
3. Results
The sample comprised of 275,057 adolescents aged 1217 years
(mean age was 14.6 (SD 1.18) years of whom 51.8% were females.
Response rates ranged from 60% in Chile to 99.8% in Jordan (supple-
mentary Table 1). Of the 82 participating countries by World Bank
classification, 25.0% of the data came from low-income countries,
31.5% from lower-middle-income countries, and 20.1% from upper-
middle-income countries and 23.4% from high-income countries.
The overall pooled prevalence of suicidal ideation and anxiety
were 14.0% (95% CI: 10.017.0%) and 9.0% (7.012.0%) respectively.
There was significant regional variation in the prevalence of mental
health problems. The highest pooled prevalence of suicidal ideation
was observed in the Africa Region (21.0%; 20.021.0%) and the low-
est was in the Asia region with 8.0% (8.09.0%; Fig. 1). The highest
pooled prevalence of anxiety was observed in Eastern Mediterranean
Region (17.0%, 16.017.0%) whereas the lowest prevalence of anxiety
was in the European Region (4.0%, 4.05.0%).
Figure 1.1. and 1.2.
Figure 2.1. and 2.2.
According to the country income classification, pooled prevalence
of suicidal ideation among the adolescents was highest in upper-mid-
dle-income countries (17.0%; 17.018.0%) and lowest in lower-
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
T. Biswas et al. / EClinicalMedicine 00 (2020) 100395 3
income group of LMICs (11.0%; 10.012.0; Fig. 1). On the other hand,
pooled prevalence of anxiety (10.0%, 10.011.0%) was highest in
lower-income economies and lowest in Lower-middle-income econ-
omies countries (9.0%, 9.010.0%). Similarly, there was wide varia-
tion in the mental health problems at the country level. The country-
specific prevalence of suicidal ideation and anxiety are presented in
Fig. 2. Prevalence of suicidal ideation ranged from 1.0% in Myanmar
to 40.0% in Samoa. Prevalence of anxiety ranged from 2.0% in Myan-
mar to 84.0% in Djibouti (Fig. 2). A large variation in prevalence
among both male and female adolescents was observed (Fig. 2). The
prevalence of suicidal ideation (male: 10.13% vs female: 13.08%) and
anxiety (male: 7.35% vs female: 9.21%) was higher among the female
than male adolescents. In almost all countries suicidal ideation and
anxiety were more prevalent in adolescent females compared to
males (Fig. 2). Countries had two time points data and that informa-
tion presented in supplementary Table 2.
Table 1 shows the prevalence of suicidal ideation and anxiety by
age group, socioeconomic status, peer support, parent-adolescent
relationship, peer victimization, conflict, isolation and loneliness.
Mental health problems were more prevalent in those of older age,
those from a lower SES, and those without peer support. Almost
without exception, suicidal ideation and anxiety in adolescents was
negatively associated with higher levels of peer support and parental
support (table 1). At the same time, suicidal ideation and anxiety
were positively associated with greater parental control, peer victim-
ization, conflict, isolation and loneliness.
Fig. 1.1. Pooled prevalence of suicidal ideation and anxiety, by WHO region, among adolescents aged 1217 years.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
4T. Biswas et al. / EClinicalMedicine 00 (2020) 100395
Table 2 shows the multivariate associations (adjusted model) of sui-
cidal ideation and anxiety for the overall sample and supplementary
Table 34 by WHO region. Overall, mental health problems were associ-
ated with being female, older age, and having a lower socioeconomic sta-
tus and an absence of close friends. For example, suicidal ideation was
higher among females (OR; 1.68; 95% CI: 1.551.81) older age (1.55;
1.411.71), and having a lower socioeconomic status (1.10; 1.011.19)
and those reporting no close friends (2.14; 1.872.46) compared to their
counterparts. Those adolescents who perceived they experienced paren-
tal understanding and experienced more parental monitoring had signif-
icantly lower rates of suicidal ideation (0.69; 0.610.79) and (0.68;
0.620.76) respectively. Conversely, suicidal ideation and anxiety were
more likely to be reported by those adolescents who perceived higher
levels of parental control (1.65; 1.451.87; and 1.53; 1.301.80
respectively). Similarly, peer victimization, peer conflict, peer isolation
and loneliness were all strongly correlated with suicidal ideation and
anxiety. For example, anxiety was higher among adolescents who had
experienced peer conflict (1.54; 1.401.70), peer victimization (1.13;
1.021.26), peer isolation (1.76; 1.611.92) or reported loneliness (5.63;
5.216.08) compared to those who did not endorse these items.
4. Discussion
The present study based on the GSHS data provides the most com-
prehensive summary to date of the prevalence of suicidal ideation
and anxiety among adolescents across 82 HIC-LMICs from six WHO
regions. This is the first global study of its kind to examine the rela-
tionship between suicidal ideation and anxiety with peer and
Fig.. 1.2. Pooled prevalence of suicidal ideation and anxiety, by the World Bank income groups, among adolescents aged 1217 years.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
T. Biswas et al. / EClinicalMedicine 00 (2020) 100395 5
parental relationships in a wide range of countries. There are four
major findings. First, a high prevalence of suicidal ideation and
anxiety was observed in most of the 82 countries, irrespective of
income classification. Second, there was a wide variation between
countries in the prevalence of suicidal ideation and anxiety. Third,
there were strong and consistent associations between suicidal
ideation and anxiety in adolescents and the nature of the rela-
tionships they had with their parents and peers across all world
regions. Finally, negative experiences with peers such as peer vic-
timization, conflict, isolation and loneliness were strongly corre-
latedwithsuicidalideationandanxiety.
A previous study of 59 LMICs using GSHS data reported that the
prevalence of suicidal ideation was 16.9% [15]. Studies in Nepal and
China [22]. reported that the prevalence of suicidal ideation was
13.6% in general, and 17.4% among adolescents [23]. We found a
higher prevalence of suicidal ideation and anxiety in females com-
pared to male adolescents which aligns with a study of loneliness
among adolescents in the US, Czech Republic and Russia [24]. The
higher prevalence of suicidal ideation among female adolescents is
consistent with other studies of internalizing disorders for example,
higher rates of any mood disorder among females (18.3%) than
males (10.5%) [25]. Suicidal ideation is only loosely connected with
Fig.. 2.1. Prevalence of suicidal ideation among adolescents aged 1217 years for 82 LM-HICs, 20032015.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
6T. Biswas et al. / EClinicalMedicine 00 (2020) 100395
completed suicide which is higher among males. It is likely that sui-
cidal ideation is a proxy for psychological distress. Its higher preva-
lence in female adolescents, especially in LMICs may be explained
by a range of biopsychosocial factors which would include gender
inequity, a tendency for females to internalize their distress more
than males, greater exposure of females to forms of maltreatment,
particularly childhood sexual abuse in addition to biological factors
such as estrogen [26].
Adolescents who reported they had higher levels of peer support,
parental understanding and monitoring had significantly lower rates
of mental health problems. Conversely suicidal ideation and anxiety
were approximately double in adolescents who perceived high
levels of parental control. Other studies have also reported child
anxiety to be significantly associated with observed parental con-
trol [2730].Alignedtoourfindings, other studies have also
highlighting the relationship between peer support, paternal rela-
tionship and depressive symptoms in adolescents [31,32]. At the
same time both cross-sectional and longitudinally studies
reported, higher levels of peer support also predict reduced onset
of depressive symptoms over time [32,33].
Resources for mental health services in LMICs are scarce. The
World Health Assembly adopted the Comprehensive Mental Health
Fig.. 2.2. Prevalence of anxiety among adolescents aged 1217 years for 82 LM-HICs, 20032015.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
T. Biswas et al. / EClinicalMedicine 00 (2020) 100395 7
Action Plan 20132020 [33]. However, of the 82 countries in this
study, 36 countries have no specific mental health policy (Supple-
mentary figure-1). The strong association between parental and
peer relationships on adolescent anxiety and suicidal ideation should
inform national policies to improve population mental health. Cultur-
ally appropriate interventions that modify the parent- adolescent
relationship and promote the adolescent’s individuation-separation
whilst maintaining parental monitoring and understanding may also
promote mental wellbeing in adolescents. Similarly, establishing
school-based programmes or community activities that increase peer
connectedness may also help reduce distress, anxiety, and alleviate
progression to suicidal ideation.
The present study needs to be viewed in the context of a number
of important limitations. First, there is a risk of selection bias because
school attendance is low in some counties and only children that
attend school participated. Non-attendance at school is differentially
likely to be more so among students in LMICs than in HICs. Further,
some students were absent from school on the day of data collection.
Second, suicidal ideation and anxiety were each measured in the
GSHS using a single self-report item. While self-report is an accepted
method of measuring suicidal ideation and anxiety in adolescents,
there is a limitation of possible shared method variance. Further,
there would be cultural and legal factors that might influence self-
report, particularly of suicidal ideation. For example, in some coun-
tries, suicide is illegal (eg India, Singapore and Kenya) and discussion
of suicidal ideation is discouraged. Another limitation is the measure
of socioeconomic status, peer support, parent-adolescent relation-
ship, peer victimization, conflict, isolation which were based on a
proxy measures generally derived from a single item. Some regions
(European and African) were not well represented in the study. The
study design was cross-sectional, therefore the temporality between
the variables in unclear and causality must not be inferred.
On the other hand, the study has a number of strengths that help
us to uniquely estimate the global prevalence of suicidal ideation and
anxiety amongst adolescents in various settings. First, the GSHS
methodology uses a standardized questionnaire. Collection of the
data was standardized and always occurred during a regular class
period. The questionnaire did not allow skip patterns in questions
enabling consistency and uniformity of comparison across participant
sites. Another strength is the use of survey data with large random
Table. 1
Prevalence of suicidal ideation and anxiety by age group, socioeconomic sta-
tus, parent-adolescent bonding, peers support, peers victimization, peers con-
flict, peers isolation.
Variables Suicidal ideation (%, 95% CI) Anxiety (%, 95% CI)
Sex
Boys 10.1 (9.810.5) 7.4 (7.17.7)
Girls 13.1 (12.713.5) 9.2 (8.99.5)
Age
1213 years 9.3 (8.99.7) 6.8 (6.47.1)
1415 years 11.6 (11.212) 8.3 (88.6)
1617 years 14.3 (13.615) 10.6 (10.111.1)
Socioeconomic status
Average 10.9 (10.611.2) 7.2 (6.87.2)
Below average 13.1 (12.713.6) 10.1 (10.411.3)
Close friends
No 24.3 (22.925.8) 12.7 (11.713.7)
Yes 10.7 (10.511) 7.9 (7.78.1)
Peers supportive
Never/rarely 15.1 (14.515.7) 10.5 (10.111.3)
Sometimes 10.1 (9.610.5) 6.8 (6.57.2)
Most of the times 9.5 (8.910.1) 7.8 (7.38.3)
Always 9.1 (8.59.6) 7.8 (7.38.3)
Parental understanding
Never/rarely 15.5 (15.116.3) 10.6 (10.211.0)
Sometimes 10.1 (9.610.6) 6.4 (6.36.7)
Most of the times 8.4 (7.99.1) 7.3 (6.87.8)
Always 8.5 (8.18.9) 7.3 (6.87.7)
Parental Monitoring
Never/rarely 15.4 (14.916) 10.1 (9.610.4)
Sometimes 10.4 (9.911) 7.2 (6.87.6)
Most of the times 9.1 (8.69.7) 7.8 (7.38.4)
Always 8.7 (8.39.1) 7.2 (6.87.6)
Parental control
Never/rarely 9.8 (9.510.2) 7.1 (6.87.4)
Sometimes 10.8 (10.111.6) 7.3 (6.77.9)
Most of the times 15.8 (14.317.4) 13.2 (11.914.6)
Always 14.3 (13.115.8) 11.8 (10.713.1)
Peer conflict
No 9.3 (9.09.6) 6.4 (6.26.7)
Yes 15.8 (15.216.3) 12.5 (12.113.3)
Peer victimization
No 9.2 (8.89.5) 6.6 (6.36.8)
Yes 13.4 (12.913.9) 10.5 (1010.9)
Peer isolation
No 8.1 (7.88.4) 5.4 (5.25.6)
Yes 17.3 (16.717.9) 14 (13.514.5)
Loneliness
No 10.0 (9.49.9) 6.0 (5.35.7)
Yes 27.0 (26.128.3) 30.0 (29.431.6)
Table. 2
Multiple logistic regression analysis (adjusted ORs) to identify factors associated
with suicidal ideation and anxiety among the young adolescent.
Suicidal ideation Anxiety
Variables OR (95% CI) p- value OR (95% CI) p- value
Sex
Boys Ref Ref
Girls 1.68 (1.551.81) <0.001 1.49 (1.361.65) <0.001
Age
1213 years Ref Ref
1415 years 1.22 (1.111.34) <0.001 1.19 (1.081.31) <0.001
1617 years 1.55 (1.411.71) <0.001 1.68 (1.491.9) <0.001
Socioeconomic status
Average Ref Ref
Below average 1.10 (1.011.19) 0.032 1.30 (1.191.42) <0.001
Close friends
No 2.14 (1.872.46) <0.001 1.06 (0.881.28) 0.551
Yes Ref Ref
Peers supportive
Never/rarely Ref Ref
Sometimes 0.76 (0.680.85) <0.001 0.83 (0.720.96) 0.011
Most of the times 0.77 (0.680.87) <0.001 0.96 (0.841.11) 0.582
Always 0.78 (0.70.87) <0.001 1.03 (0.911.17) 0.641
Parental understanding
Never/rarely Ref Ref
Sometimes 0.77 (0.680.87) <0.001 0.73 (0.650.83) <0.001
Most of the times 0.61 (0.550.69) <0.001 0.82 (0.70.95) 0.010
Always 0.69 (0.610.79) <0.001 0.93 (0.811.06) 0.273
Parental monitoring
Never/rarely Ref Ref
Sometimes 0.77 (0.690.86) <0.001 0.91 (0.81.04) 0.158
Most of the times 0.67 (0.60.74) <0.001 0.94 (0.811.1) 0.440
Always 0.68 (0.620.76) <0.001 0.92 (0.81.06) 0.256
Parental control
Never/rarely Ref Ref
Sometimes 1.27 (1.131.42) <0.001 1.09 (0.941.26) 0.241
Most of the times 1.60 (1.351.89) <0.001 1.69 (1.441.98) <0.001
Always 1.65 (1.451.87) <0.001 1.53 (1.301.80) <0.001
Peer conflict
Yes 1.36 (1.241.5) <0.001 1.54 (1.401.70) <0.001
No Ref Ref
Peer victimization
Yes 1.26 (1.151.38) <0.001 1.13 (1.021.26) 0.019
No Ref Ref
Peer isolation
Yes 1.69 (1.531.86) <0.001 1.76 (1.611.92) 0.000
No Ref Ref
Loneliness
Yes 2.56 (2.332.82) <0.001 5.63 (5.216.08) <0.001
No Ref Ref
*Adjusted by survey year, gender, age, socioeconomic status, peer support, par-
ent-adolescent bonding, injuries, violence and loneliness.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
8T. Biswas et al. / EClinicalMedicine 00 (2020) 100395
sample sizes taken from a wide variety of international geographical
and cultural settings. Finally, the analyses were inclusive of data from
82 countries.
Mental health problems such as suicidal ideation and anxiety in
adolescents are a major public health concern globally [34]. Although
there is enormous global variation, there is consistency across cul-
tures and across country development status (LMIC vs HIC) that men-
tal health in adolescents is strongly influenced by parental and peer
relationships. Similarly, there is a consistently higher prevalence of
mental ill-health in female adolescents compared with males. These
findings are important to inform national and global policies and
related actions to address adolescent suicidal ideation and anxiety.
Adolescent suicidal ideation and anxiety prevention strategies should
include female-specific initiatives, family and peer relationships
which are sociocultural specific and sensitive. Given the substantial
variation among countries and regions, more work is needed to
understand the sociocultural context of the antecedents of adoles-
cents’mental health related behaviours in low-income and middle-
income countries.
Contributors
All authors critically reviewed earlier versions of the draft and
approved the final manuscript. TB and AAM conceived the paper. TB,
JGS and AAM developed the analysis plan. TB did the analysis and
wrote the initial draft. AMNR, KM, LBR and JB contributed to the write
up and editing.
Declaration of interests
All other authors declare no competing interests.
Acknowledgments
We thank the US Centers for Disease Control and WHO for making
Global School-based Student Health Survey (GSHS) data publicly
available for analysis. We thank Md. Mehedi Hasan, for helping us
data management. This research was partially supported by the Aus-
tralian Research Council Centre of Excellence for Children and Fami-
lies over the Life Course (project number CE140100027). James Scott
is supported by a National Health and Medical Research Council Prac-
titioner Fellowship Grant (APP1105807).
Supplementary materials
Supplementary material associated with this article can be found
in the online version at doi:10.1016/j.eclinm.2020.100395.
References
[1] Magadi MA. multilevel determinants of teenage childbearing in sub-saharan
Africa in the context of Hiv/aids. Health & place 2017;46:37–48.
[2] Weiss HA, Ferrand RAJTL. Improving adolescent health: an evidence-based call to
action 2019;393(10176): 1073-5.
[3] Steinberg L, Morris AS. Adolescent development. Annual review of psychology
2001;52(1).
[4] Patton GC, Olsson CA, Skirbekk V, et al. Adolescence and the next generation
2018;554(7693): 458.
[5] The LJL Health an afterthought in a single-issue election 2017;389(10080):1670.
[6] Erskine H, Moffitt TE, Copeland W, et al. A heavy burden on young minds: the
global burden of mental and substance use disorders in children and youth. Psy-
chological medicine 2015;45(7):1551–63.
[7] https://www.who.int/news-room/fact-sheets/detail/adolescent-mental-health
[8] Moore GF, Cox R, Evans RE, et al. School, peer and family relationships and adoles-
cent substance use, subjective wellbeing and mental health symptoms in wales: a
cross sectional study 2018;11(6): 1951-65.
[9] Tough H, Siegrist J, Fekete CJBph. Social relationships, mental health and wellbe-
ing in physical disability: a systematic review2017;17(1): 414.
[10] Bowes L, Maughan B, Caspi A, Moffitt TE. Arseneault LJJoCP, Psychiatry Families
promote emotional and behavioural resilience to bullying: evidence of an envi-
ronmental effect2010;51(7): 809-17.
[11] Nguyen HTL, Nakamura K, Seino K, Al-Sobaihi SJBp. Impact of parentadolescent
bonding on school bullying and mental health in Vietnamese cultural setting: evi-
dence from the global school-based health survey2019;7(1): 16.
[12] Sweeting H, Young R, West P, Der GJBjoep. Peer victimization and depression in
earlymid adolescence: A longitudinal study2006;76(3): 577-94.
[13] Kieling C, Baker-Henningham H, Belfer M, et al Child and adolescent mental
health worldwide: evidence for action2011;378(9801): 1515-25.
[14] Polanczyk GV, Salum GA, Sugaya LS, Caye A, Rohde LA. Annual Research Review:
A metaanalysis of the worldwide prevalence of mental disorders in children
and adolescents. Journal of Child Psychology and Psychiatry 2015;56(3):345–65.
[15] Uddin R, Burton NW, Maple M, Khan SR, Khan AJTLC. Health A Suicidal ideation,
suicide planning, and suicide attempts among adolescents in 59 low-income and
middle-income countries: a population-based study 2019;3(4): 223-33.
[16] Jordans M, Rathod S, Fekadu A, et al. Suicidal ideation and behaviour among com-
munity and health care seeking populations in five low-and middle-income coun-
tries: a cross-sectional study 2018;27(4): 393-402.
[17] Le MTH, Nguyen HT, Tran TD, Fisher JRJJoah. Experience of low mood and suicidal
behaviors among adolescents in Vietnam: findings from two national population-
based surveys 2012;51(4): 339-48.
[18] WHO Global School-based Student Health Survey (GSHS); http://www.who.int/
ncds/surveillance/gshs/en/ (accessed July 25).
[19] https://www.cdc.gov/gshs/background/pdf/gshs-data-users-guide.pdf.
[20] Biswas T, Scott JG, Munir K, et al. Global variation in the prevalence of bullying
victimisation amongst adolescents: Role of peer and parental supports. 2020:
100276.
[21] Rao S, Shah N, Jawed N, Inam S, Shafique KJBph Nutritional and lifestyle risk
behaviors and their association with mental health and violence among Pakistani
adolescents: results from the National Survey of 4583 individuals2015;15(1):
431.
[22] Cheng Y, Tao M, Riley L, et al Protective factors relating to decreased risks of ado-
lescent suicidal behaviour2009;35(3): 313-22.
[23] Pandey AR, Bista B, Dhungana RR, Aryal KK, Chalise B, Dhimal MJPo Factors associ-
ated with suicidal ideation and suicidal attempts among adolescent students in
Nepal: Findings from Global School-based Students Health Survey2019;14(4):
e0210383
[24] Stickley A, Koyanagi A, Koposov R, et al. Loneliness and its association with psy-
chological and somatic health problems among Czech. Russian and US adoles-
cents 2016;16(1):128.
[25] Merikangas KR, Nakamura EF, Kessler RC. Epidemiology of mental disorders in
children and adolescents 2009;11(1):7.
[26] Altemus M, Sarvaiya N, Epperson CNJFin. Sex differences in anxiety and depres-
sion clinical perspectives. 2014;35(3): 320-30.
[27] Van Der Bruggen CO, Stams GJJ, B€
ogels SMJJoCP, Psychiatry Research Review: The
relation between child and parent anxiety and parental control: a metaanalytic
review2008;49(12): 1257-69.
[28] De Rosnay M, Cooper PJ, Tsigaras N, Murray LJBr, therapy Transmission of social
anxiety from mother to infant: An experimental study using a social referencing
paradigm2006;44(8): 1165-75.
[29] Gerull FC, Rapee RMJBr, therapy. Mother knows best: effects of maternal model-
ling on the acquisition of fear and avoidance behaviour in toddlers2002;40(3):
279-87.
[30] Waite P, Whittington L, Creswell CJPR. Parent-child interactions and adolescent
anxiety: A systematic review2014;1(1): 51-76.
[31] MacPhee AR, Andrews JJJA. Risk factors for depression in early adolescence
2006;41(163): 435.
[32] Chester C, Jones DJ, Zalot A, Sterrett EJJoCC, Psychology A The psychosocial adjust-
ment of African American youth from single mother homes: The relative contri-
bution of parents and peers 2007;36(3): 356-66.
[33] Allen JP, Insabella G, Porter MR, et al A social-interactional model of the develop-
ment of depressive symptoms in adolescence. 2006;74(1): 55.
[34] Saxena S, Funk M, Chisholm DJTLWorld health assembly adopts comprehensive
mental health action plan 20132020 2013;381(9882): 1970-1.
ARTICLE IN PRESS
JID: ECLINM [m5G;June 16, 2020;10:06]
Please cite this article as: T. Biswas et al., Global variation in the prevalence of suicidal ideation, anxiety and their correlates among
adolescents: A population based study of 82 countries, EClinicalMedicine (2020), https://doi.org/10.1016/j.eclinm.2020.100395
T. Biswas et al. / EClinicalMedicine 00 (2020) 100395 9