Bullying among children and adolescents in the SAARC countries: A scoping review

Abstract

Background

Bullying is defined as repetitive and intentional aggression by an individual or group towards other individuals that happens in a power differential between the individuals being bullied and the bullies. There is increasing recognition of how bullying occurs among children and adolescents and its long-term effects. There is a dearth of research on bullying from the Lower- and Middle-Income Countries (LMIC). This scoping review focused on the research from the South Asian Association for Regional Cooperation (SAARC) nations that share a common history, similar demographics, and socio-cultural background.

Methodology

Various databases were searched using specific search terms and articles reviewed from the past 5 years.

Results

Of 194 articles identified, 53 met the criteria for inclusion in the review. There is a wide variation in the number of studies done across the SAARC nations. The prevalence of bullying victimization ranged from 4.1% to 95% and from 16 to 85% for perpetration. Only 3 interventions conducted in India and Pakistan showed some efficacy of play, the teaching of skills and multicomponent interventions to deal with bullying, each made culturally relevant.

Discussion

This review highlights the lacunae in the research conducted on bullying in the Indian sub-continent. It also highlights the need for contextually appropriate definitions, long term effects on the health and well-being of bullying, and socially appropriate interventions to address bullying.

Bullying; SAARC nations; Prevalence; Definitions; Intervention.

1. Introduction

The South Asian Association for Regional Cooperation (SAARC) was founded in 1985 by a group of countries (Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka) in the Indian subcontinent with a focus on addressing issues of agriculture, health, poverty and food security (Delinic, 2011). As of 2015, the population of the SAARC countries was approximately 1.75 billion individuals with a mean age of 27 years in an increasingly urbanized context (United Nations, n.d.). The countries have a long and closely-interwoven history, similar socio-cultural norms, structure and funding of the healthcare systems, kinds of illnesses and available resources in the management of the same, which was demonstrated during the recent COVID-19 pandemic (Sultana and Reza, 2020; The World Bank, n.d.). Given the similarities in the demographic distribution of youth in the SAARC countries and the mental health burden among the youth, creating contextualized interventions that address critical mental health related risk factors is essential. Due to these similarities, extending the interventions developed in one country may be possible for others in the region.

Bullying or peer victimization is a complex health and social issue that has been noted to have both immediate and long-term consequences on all the individuals involved (Rivara et al., 2016; Wolke & Lereya 2014). When viewed from an evolutionary perspective, this behaviour has been utilized to gain higher status and dominance in a group, access resources, survive, and reduce stress (Volk et al., 2012). The seminal work by Dan Olweus (1993) described bullying to have a few common characteristics: as being a condition of imbalance of power, presenting as a repetitive and intentional aggressive behaviour by one individual or group against another. This power differential may be actual or perceived by individuals involved. Across the sexes, ages, cultures and socioeconomic strata of society, the aggression associated with bullying is different from any other cause primarily because of the repetitive nature of this act (Smith et al., 2016; Tippett and Wolke, 2014).

Children can be involved in bullying in multiple roles, including victims, bullies, bystanders or a bully-victims (children who are victimized but bully other children). Victimization may occur by direct and indirect methods. Direct victimization includes physical and verbal aggression (hitting, stealing/snatching or name-calling), whereas indirect victimization includes acts such as social ostracism (exclusion from common activities) or rumour spreading. With the proliferation of the internet and other mobile devices, youth can now access their peers within their own homes. This is cyberbullying, which is defined as bullying that takes place over digital devices (like cell phones, computers and tablets) and occurs through SMS, text, and apps or online in social media, forums or gaming where people can view, participate in or share content (StopBullying.gov, 2020). Bullying typically occurs in settings where the individuals are unable to choose the groups they wish to be in, and the most common scenario is that of school. Acts of indirect victimization and cyberbullying do not restrict to the groups that children are in but may even be more widespread. Bullies exert their power with all children to establish their position in the social hierarchy. However, victims of bullying (repeated targets) are those who have an emotional response or do not have others to stand up for them. Evidence supports that conditions of higher density and more significant hierarchies in classrooms are associated with higher risks of the persistence of bullying (Wolke and Lereya, 2015).

This review aims to identify and compile the available scientific literature in the field of bullying in the SAARC countries to guide future work on developing appropriate interventions to address this problem.

2. Methodology

In order to replicate the best practices in scoping reviews, we utilized the methodological framework provided by Arksey and O'Malley (Arksey & O'Malley, 2005). Since this is a developing field in LMICs, the scoping review methodology was pursued to determine the current state and extent of research that has been done in the SAARC region. The following steps were followed:

2.1. Stage 1: development of research question

The primary research question was to evaluate the breadth of available literature on the topic of bullying in children and adolescents. The research question answered through this review was “In the SAARC region, what is the current state of knowledge about the definition, measurement, prevalence, risk and protective factors and the kinds of intervention for bullying among children and adolescents?”

2.2. Stage 2: Identifying relevant studies

After defining the research question, the following databases - PubMed, CINAHL, EMBASE, ERIC, PsycINFO, SCOPUS and Web of Science were searched using keywords/MeSH terms. The search terms included were—“bully∗” OR “bullying” AND (India OR Pakistan OR Bangladesh OR Sri Lanka OR Nepal OR Bhutan OR Maldives) searched in the text, title and abstract. Limits of Age (<18 years) and time (from June 2016–July 2021) were also applied.

2.3. Stage 3: study selection

The abstracts of the search results were then screened for details that would enable inclusion or exclusion from the review (see box 1). The references of the included articles and the rejected reviews were screened for any missed articles. These were included in the final analysis. Two independent researchers (AC and ER) conducted the initial screening of the titles and abstracts (N = 190). Any discrepancies were discussed with the lead author SP before they were finally included or excluded. Of the original search results, 33 articles did not address bullying directly, while 25 were not from the target countries. The articles' full texts were accessed and reviewed by SP before they were finally included (N = 53). Details of the search strategy are shown in figure 1.

Box 1. Inclusion and Exclusion criteria for studies included in the review.

Inclusion criteria

• •Age group of 6–18 years of age
• •Published in peer reviewed, indexed journals
• •Observational or interventional studies (e.g cross sectional, case-control and cohort studies, clinical trial, controlled clinical trial, RCT, pragmatic clinical trial, validation study)
• •Published within the last 5 years (i.e. 1st June 2016 to 1st June 2021)
• •Countries: Bangladesh, Bhutan, India, Maldives, Nepal, Pakistan and Sri Lanka

Exclusion criteria

• •Commentaries, editorials, reviews or meta-analysis on a related topic
• •Information from the grey literature
• •Articles not in English
• •Articles where the SAARC countries are not represented. If a multi country study is conducted, then if the data on the individual countries is not made available, then it will be excluded

Alt-text: Box 1

Figure 1.

Search strategy.

2.4. Stage 4: charting the data

Data from the studies that met the inclusion and exclusion criteria were collected and charted in an excel sheet. Details (box 2) were extracted from the full-text articles included in the study.

Box 2. Information charted from full text articles.

• •Authors
• •Year of publication
• •Journal
• •Country/countries where the study was conducted
• •Age group under study
• •Methodology of study
• •Location of data collection
• •Definition of bullying/Measurement tool used
• •Prevalence
• •Risk factors identified
• •Protective factors identified
• •Intervention
• •Degree of change shown
• •Key learnings from the study

Alt-text: Box 2

2.5. Stage 5: collating information, summarizing and reporting the results

The information obtained from the full texts of the studies was collated using an excel sheet. A narrative review of the studies was summarized under the themes of the distribution of research activities across the countries, definitions used in the research, epidemiology including associations of risk and protective factors, interventions done and their effects. Finally, a separate section on the studies that provided unique information about the bullying research in the target nations was included.

3. Results

The flowchart of the search strategy is shown in figure 1. Details of the 53 studies included in the review are shown in table 1. These included 49 articles from the original database search and 4 from the cross-references. There was a wide variation in the number of articles published in the past 5 years. These included 23 that had data from India, 12 from Pakistan, 8 each from Nepal and Bangladesh, 4 from Sri Lanka and 1 from Bhutan. There were no studies that were published from the Maldives. 3 studies had data from 2 of the SAARC countries. The information from the review of articles is summarized under the following headings.

Table 1.

Description of studies included in the review.

S.No	Authors	Country	N	Age group	Location of study	Type of study	Prevalence	Associations found
1.	Abid et al., 2017	Pakistan	200	8–11	School	Cross-sectional research	•NA	•Negative correlation between mindfulness and bullying behaviours •Single children were more likely to be bullied compared to children of other birth orders
2.	Asad et al., 2017	Pakistan	1752	12–14	School	RCT	•Victimization only Boys—17.8% •Victimization only Girls—28.5% •Any perpetration boys—75.5% •Any perpetration girls—50.6%	•Peer perpetration impacts girls’ risk of depression differently than boys, yielding females at greater risk to depression when reporting the same levels of perpetration as boys.
3.	Beattie et al., 2019	India	1191	13–14	Community	Cross-sectional research	•8% eve teasing •4.1% bullying or harassment in the past 12 months	•Recent eve teasing was significantly associated with not having hope for the future •Feeling down, depressed or hopeless was associated with recent eve teasing and a harassing/abusive school environment
4.	Bhatt et al., 2020	Nepal	602	13–17	Community—Child care homes	Cross sectional research	•16.6% ever bullied	•Clinically relevant depressive symptoms OR 1.97
5.	Chudal et al., 2021	India (Among other countries)	2016	13–15	School	Cross-sectional research	•Any Victimization 23.2% •Traditional 17.5% •Cyber only—1.3% •Combined—2.2%	•Boys being bullied OR 2.43 •Girls being bullied OR 1.14 •Both externalizing and internalizing symptoms were significantly higher when considering combined bullying as compared to either traditional or cyber bullying in both the sexes
6.	Dema et al., 2019	Bhutan	5809	13–17	School	Cross-sectional research	•25.6%	•Suicidal ideas aPR 1.3 •Suicidal attempts aPR 1.6
7.	Gunawardana et al., 2021	Sri Lanka	110	8–12	Clinic based	Cross-sectional research	•NA	•Psychiatric quality of life was significantly lower in obese/overweight children subject to bullying
8.	Haque et al., 2021	Bangladesh	1416	11–17	Community	Cross sectional research	•24.29% bullied by siblings	•Children who were bullied by their siblings had an increased risk of psychological abuse but, not neglect by adults
9.	Hettiarachchi et al., 2018	Sri Lanka	181	12–16	Children in detention	Cross-sectional research	•71.82% of all children had experienced bullying	•No significant association between self harm behaviour and bullying in the children in Juvenile Justice systems in Sri Lanka
10.	Irish and Murshid, 2020	Bangladesh	2883	11–17	School	Cross-sectional research	•24.43%	•Suicidal ideas—OR 2.23 •Suicide attempts—OR 2.94
11.	Karmaliani et al., 2020	Pakistan	1752	12–14	School	RCT	•NA	•Play based intervention to target depressive symptoms, and peer victimization and perpetration. •Boys peer victimization score—27.8% (Control arm) vs. -33.3% (intervention arm) •Girls Control -21.3% vs. 58.5% in intervention arm •The intervention was also noted to have a positive effect on the experience of physical punishment at school, home and gender attitudes
12.	Khan and Khan, 2020	Bangladesh	2883	11–17	School	Cross-sectional research	•NA	•aOR of anxiety 6.00
13.	Khan et al., 2020	Bangladesh	2989	11–17	School	Cross-sectional research	•NA	•aOR of Tobacco use 1.93 •aOR of other substance use—3.43
14.	Khan et al., 2020	Bangladesh	2989	11–17	School	Cross-sectional research	•NA	•25.3% age adjusted prevalence of suicidal behaviour •aRR of Suicidal behaviour 1.88
15.	Lee et al., 2019	Nepal Sri Lanka	914 1404	NA	School	Cross-sectional research	•Nepal 14% •Sri Lanka 23.5%	•Psychological distress OR in Nepal 3.34, in Sri Lanka 2.66
16.	Mallik and Radwan, 2020	Bangladesh	276	14–17	School	Cross-sectional research	•31.9% experienced cybervictimization	•Cybervictimization significantly associated with any psychiatric disorder (27.27%), MDD (9.09%) any emotional disorder (32.59%) and any behavioural disorder (12.5%)
17.	Mansoor and Shahzad, 2020	Pakistan	150	12–16	School	Cross-sectional research	•NA	•Urdu version of adolescent peer relation instrument was validated
18.	McFarlane et al., 2017	Pakistan	1752	12–14	School	RCT	•94% boys and 85% girls reported at least 1 act of victimization •85% boys and 66% girls reported at least 1 act of perpetration
19.	Menon and Hannah-Fisher, 2019	India	296	10–14	School	Cross-sectional research	•NA	•Boys scored higher on both victimization and aggression compared to girls. •Felt pressure, work sexism and entity view of gender differences were associated with higher victimization the effect that was more evident in girls.
20.	Mishra et al., 2018	Nepal	405	12–20	School	Cross-sectional research	•Bully—52.3% •Victim 58.0% •Bully-Victim 41.2%	•Boys had higher rates of bullying 62.25% and victimization 62.76% compared to girls •Bullies were more (55.8%) among the relatively advantaged Janajatis (Castes) and victims were more (64.86%) among the disadvantaged Janajatis
21.	Mukherjee et al., 2019	India	254	15–19	School	Cross-sectional research	•10.5% cyber bullying	•68.2% sought help from their friends
22.	Murshid, 2017	Pakistan Sri Lanka	4977 2500	12–16	School	Cross-sectional research	•NA	•APR of experiencing depressive symptoms in adolescents who were bullied was 1.68 and 1.52 in Pakistan and Sri Lanka respectively
23.	Murshid, 2018	Pakistan	4977	12–16	School	Cross-sectional research	•Bullying victimization 37.8%	•Students with good hygiene are less likely to be bullied as compared to students with poor hygiene.
24.	Nambiar et al., 2020	India	40	10–18	Clinic based	Cross-sectional research	•Peer victimization - 75%	•Peer victimization was significantly higher when the students were in a regular school compared to special schools
25.	Naveed et al., 2019	Pakistan	452	10–17	School	Cross-sectional research	•13.30% pure victims •16.80% pure bullies •15.50% bully victims	•20.6% bullied by girls •10.7% bullied by boys •21.5% bullied more than once a week •27.7% bullying in school •14% bullying out of school •Bully victimization strongest predictor of depression symptoms followed by perpetration and victimization
26.	Naveed et al., 2020	Pakistan	2315	10–17	School	Cross-sectional research	•26.6% bullied at school •17.9% bullied out of school •18.6% perpetration in school •16.3% perpetration out of school •31.7% bully victims	•Student bullying and victimization at school showed strong association to being bullied out of school
27.	Neupane et al., 2020	Nepal	6529	11–17	School	Cross-sectional research	•50.7% bullied •Males 55.7% •Females 46.2%	•Tobacco use aOR 2.05 •Physical fight aOR 2.04 •Attempted suicide aOR 2.08 •Anxiety aOR 2.04
28.	Nguyen et al., 2019	India (Among other countries)	967	15	Community	Cohort study	•79.1% at baseline	•Higher victimization associated with lower subjective wellbeing and more emotional difficulties •At follow up all associations had attenuated and were not significant
29.	Nguyen et al., 2020	India (Among other countries)	967	15	Community	Cohort study	•56.4% bullied •58.3% boys bullied •54.5% Girls bullied	•Taken something without permission, called you names and punched kicked and beaten were the most common forms of victimization for boys •Taken something without permission, tried to get you in trouble and refused to talk to you were the most common forms of victimization for girls.
30.	Nguyen et al., 2020	India (Among other countries)	967	15	Community	Cohort study	•NA	•4 groups were identified by Latent Class Analysis in the Indian subsample including Non-Victimized, Sometimes victimized—Direct and Indirect, highly victimized
31.	Pandey et al., 2020	Nepal	6529		School	Cross-sectional research	•NA	•Being bullied increase the odds of serious injuries (aOR 2.73)
32.	Pandey et al., 2021	Nepal	6529		School	Cross-sectional research	•50.86%	•Being bullied increased the odds of physical attack (OR 2.65), fights (OR 3.39) and sexual violence (OR 1.10)NS
33.	Patel et al., 2017	India	1106	12–15	School	Cross-sectional research	•49% any form of bullying •Bullies—29.9% •Victims—29.7%	•Males were more likely to be bullies and victims •Being overweight, having less than 7 friends and poor academic performance predicted victimization
34.	Prakash et al., 2017	India	2275	13–14	Community	RCT—baseline data	•8.1% harassment/bullying environment in school	•School dropout in children who endorsed harassment/bullying environment at school aOR 1.70 •School absenteeism in children who endorsed harassment/bullying environment in school aOR 3.39
35.	Pronk et al., 2017	India (and Netherlands)	480	12–14	School	Cross-sectional research	•NA	•Different roles in the bullying role behaviours (Bully, victim, follower, defender and outsider) have clear associations with the peer group status (preference and popularity). •Positive associations with popularity were found for the bully, follower and defender •Defenders and outsiders positively associated with preference across the 2 countries.
36.	Rahman et al., 2020	Bangladesh Nepal	2989 6529	13–17	School	Cross-sectional research	•24.5% in Bangladesh •50.9% in Nepal	•Students who experienced >10 days of bullying in the past 30days had aOR of engaging in physical violence of 16.16 (pooled data) •Suicidal ideas aOR 5.88 •Suicide attempt 6.50
37.	Rana et al., 2020	India	667	12–14	School	Cross-sectional research	•25.6% any bullying •Victimization 16% •Bullying 5.2% •Bully victimization 4.3%	•Girls had significantly higher verbal bullying compared to boys •Boys had significantly higher physical bullying compared to girls •Predictors of bullying behaviour included boys (bully OR 4.24), having emotional problems (bully-victim OR 4.36) peer relation problems (victim OR 2.77)
38.	Sethi et al., 2019	India	370	12–15	School	Cross-sectional research	•43% involved in bullying •19% victims •18% perpetrators •6% victim perpetrators	•Physical abuse 21% most common •Coercion 15% least common
39.	Shah et al., 2019	Pakistan	4102	11–16	School	Cross-sectional research	•Being bullied boys 44.5% •Being bullied girls 35.6% •Ever bullied—26.1% •14.8% frequently bullied	•Good hygiene (OR 0.62), physically active lifestyle (OR 0.55) and no tobacco use (OR 0.43)
40.	Shaikh et al., 2019	Pakistan	5177	11–16	School	Cross-sectional research	•15% reported being bullied	•Bully victimization increased the odds of physical fighting (OR 3.14)
41.	Shamsi et al., 2019	Pakistan	153	NA	School	Cross-sectional research	•NA	•53.6% of the teachers lacked adequate knowledge about bullying •66.2% did not identify suicidal thoughts as a result of bullying •Teachers with 1–5 years of experience, without formal training were better able to identify bullying
42.	Sharma et al., 2017	India	178	11–15	School	Cross-sectional research	•7.5% cyber bullies •17.2% victims of cyber bullying	•Victims and offenders of one kind of bullying are more likely to indulge in another form of bullying
43.	Sharma et al., 2020	India	174	11–13	School	Quasi experimental study	•NA	•Significant difference from baseline for physical aggression at 1 month post intervention •At 6 months the effect sizes: non physical aggression (-0.5), victimization (-0.4) and physical aggression (-0.9)
44.	Shinde et al., 2020	India	15232	13–15	School	RCT	•NA	•Significant reduction in frequency of bullying for intervention arm vs. teacher (aMD- 2.65) and vs. control (aMD -2.77)
45.	Shinde et al., 2018	India	10202	13–15	School	RCT	•NA	•Significantly lower mean bullying scores in intervention arm compared to control arm
46.	Singla et al., 2021	India	5539	13–15	School	RCT	•NA	•School climate accounted for 14.87% of the total direct effect on experience of bullying •Improved relationships at school accounted for 57.4% of the total mediating effects on experiences of bullying •Participation in school events accounted for 21.9% of the mediating effect on bullying
47.	Skrzypiec et al., 2018	India (Among other countries)	531	11–16	School	Cross-sectional research	•53.4% of students were victims of bullying (Intentional harm, repetition and power imbalance)	•90.5% of students in India reported being harmed from negative experiences with peers
48.	Thakkar et al., 2019	India	1238	11–16	School	Cohort study	•NA	•Psychopathic dimensions taken together serve as a predictor of bullying roles beyond sociodemographic •Gender (boys) predicted bullies and bully victims at different times •General caste and non Hindu children were more likely to be victims
49.	Thakkar et al., 2021	India	1238	11–16	School	Cohort study	•NA	•No concurrent association between self reported victimization and BMI •Concurrent association between BMI and peer reported victimization for boys and girls •Higher BMI prospectively predicted lesser victimization for boys
50.	Upadhaya et al., 2019	Nepal	35	13–18	School	Cross-sectional research	•NA	•Bullying is an experience of distress in the community •Adolescent bully others who are punished by the teachers or because of their social position •Bullying causes emotional harm more than physical harm
51.	Wang et al., 2020	Bangladesh (Among other countries)	2989	11–17	School	Cross-sectional research	•Bully victimization 1–2 days/week = 14.6% •Bully victimization ≥3 days/week = 9.8%	•Being bullied 1–2 days a week increased the odds of sleep loss over worry by moderate degree (aOR 2.21) and severe degree (aOR 2.81) •Being bullied more than 3 days increased risk of severe sleep loss over worry (aOR 6.0)
52.	Wright et al., 2017	India (Among other countries)	480	11–15	School	Cross-sectional research	•NA	•Indian boys used aggressor blame for public than for private cybervictimization •Indian boys used a normative attribution strategy for face to face rather than cyber victimization. •Indian girls used self blame as a strategy for private than for public victimization •Indian girls used aggressor blame and conflict as strategies for cyber rather than face to face victimization
53.	Wright et al., 2017	India (Among other countries)	480	11–15	School	Cross-sectional research	•NA	•Anger, sadness, embarrassment reported more for public than private face to face victimization

aMD = Adjusted mean difference, aOR = Adjusted Odds Ratio, aPR = adjusted prevalence Ratio, NA = Not available, OR = Odds Ratio, aRR = Adjusted Risk Ratio, RCT = Randomized Controlled Trial.

3.1. Types of studies

Most of the studies conducted were cross-sectional (40 of 53) and were conducted in schools (43 of 53). The other studies included 8 experimental designs (7 RCT and 1 quasi-experimental design) and 5 longitudinal cohort studies. 7 were conducted in the community, of which one was conducted in child care homes in Nepal (Bhatt et al., 2020), 2 in clinical settings and 1 in a juvenile detention facility (Hettiarachchi et al., 2018). The studies varied in size from 35 for a qualitative study (Upadhaya et al., 2019) and 40 for young people with special needs (Nambiar et al., 2020) to over 15,000 students (Shinde et al., 2020) and focused on children between the ages of 6–17 years of age. 13 of the studies were secondary analysis of the data from the Global School Based Health Survey (GSHS), 3 from the Young Lives study and 1 from the Eurasian Child Mental Health Study (ECAMH study).

3.2. Measurement of bullying

Different studies have measured bullying in different ways (see table 2). Some studies used single questions to evaluate various aspects of bullying behaviour. These ranged from questions directed towards bullying or questions taken from other scales or descriptive of specific varieties of bullying (e.g. eve-teasing– a colloquial term for publicly troubling or annoying women by use of offensive language or behaviour). 2 studies (Mukherjee et al., 2019; Upadhaya et al., 2019) created a questionnaire (or an In-depth interview guide) measuring specific aspects of bullying or distress inducing scenarios in adolescents. The other method of quantifying bullying behaviour was by using definitions. The definitions varied across studies, with each definition focused on a specific aspect of bullying behaviour. Typically, a definition was followed up by a question enquiring into the frequency of occurrence of the bullying behaviour. The studies described cut-offs in the frequency of occurrence of the incidents or would label it present or absent.

Table 2.

Bullying Definition used in different studies.

		Studies included	Validity and reliability described
Definitions used	“A classmate says something really nasty and humiliating to you at school in front of everyone.” For cyber victimization at school was replaced by online. For private scenario in front of everyone was replaced by but nobody is around to see/hear it.	Wright et al., 2017; Wright et al., 2017	NA
	“A student is getting bullied, if another student or a group of students repeatedly treats him/her negatively or in an insulting manner. It is difficult for the bullied student to defend himself/herself. Bullying can be intermittent or continuous. Bullying can be verbal (e.g. calling names, threatening), physical (e.g. hitting, pushing) or psychological (e.g. spreading rumors, avoiding, excluding). Continuous nasty or insulting teasing is also bullying”. Cyberbullying as “Repeated mocking on the Internet, bullying via emails or text messages or spreading insulting material about another person on the Internet.”	ECAMHS study—Chudal et al., 2021	NA
	“Bullying occurs when a student or group of students say or do bad and unpleasant things to another student. It is also bullying when a student is teased a lot in an unpleasant way or when a student is left out of things on purpose. It is not bullying when two students of about the same strength or power argue or fight or when teasing is done in a friendly and fun way.”	Global School Based Health Survey—Dema et al., 2019; Irish and Murshid, 2020; Khan et al., 2020; Khan and Khan, 2020; Murshid, 2017, 2018; Neupane et al., 2020; Pandey et al., 2020, 2021; Rahman et al., 2020; Shah et al., 2019; Shaikh et al., 2019; Wang et al., 2020	NA
Instruments used:	Adolescent Peer Relation Instrument	Mansoor and Shahzad, 2020	Validation study,
	Bangla translation of the cyber victim part of the cyber victim and bullying scale	Mallik and Radwan, 2020	No details given
	Bullying behaviour scale	Abid et al., 2017	No details given
	Bullying victimization questionnaire	SEHER intervention trial—Shinde et al., 2018, 2020; Singla et al., 2021	Coefficient alpha mentioned, translated and contextualized
	Hindi version of the Korean-Peer Nomination Inventory (K–PNI)	Sethi et al., 2019	No details given
	Illinois Bully scale	Sharma et al., 2020; Thakkar et al., 2019, 2021	Coefficient alpha mentioned for English and Hindi versions
	Illinois Bully scale and Cyber harassment student survey	Sharma et al., 2017	No details given
	International Society for the Prevention of Child Abuse and Neglect Child Abuse Screening Tool for Children	Haque et al., 2021	Not details given
	Interview	Naveed et al., 2019; Naveed et al., 2020	11 items questions, Coefficient alpha mentioned
	Multidimensional Peer Victimization Scale	Nambiar et al., 2020	Face validity conducted, previously used in India
	Olweus Bully/Victim Questionnaire	Mishra et al., 2018; Rana et al., 2020s	No details given
	Participant Roles Scales	Pronk et al., 2017	CFA done
	Peer Interaction in Primary Schools	Patel et al., 2017	No details given
	Peer perpetrating and victimization scale	Right to play Intervention study - Asad et al., 2017; Karmaliani et al., 2020; McFarlane et al., 2017	Coefficient alpha mentioned for both the scales
	Peer report of bullies and victims	Thakkar et al., 2019, 2021	NA
	Social and Health assessment Peer Victimization scale	Young Lives Study - Nguyen et al., 2019; Nguyen et al., 2020; Nguyen et al., 2020	Validated, and reliability established
	Student Aggression and Victimization Questionnaire	Skrzypiec et al., 2018	CFA and coefficient alpha mentioned
Questionnaire	Predesigned questionnaire measuring use of Social networking sites, online risk behaviours, details of being cyberbullied and attitude towards cyberbullying	Mukherjee et al., 2019	No details given
	Interview guide created to evaluate areas of psychological distress in adolescents	Upadhaya et al., 2019	NA
Single questions	Single question—“How many days were your bullied in the past 30 days?”	Lee et al., 2019	NA
	Single question—bullying by other students (Clubbed with 4 other questions related to harassment at school)	Prakash et al., 2017	NA
	Single question—from the Aggression/Victimization scale—“the number of times you have been a victim or perpetrator of aggressive behaviours in at school in the past 7 days”	Menon and Hannah-Fisher, 2019	NA
	Single question for eve teasing in three scenarios (at school, on the way to school and in the village)—Sometimes when girl's bodies mature, they begin to attract unwanted attention from boys and men. In the past 12 months have you been sexually harassed or teased?	Beattie et al., 2019	NA
No information mentioned	No information mentioned	Bhatt et al., 2020; Gunawardana et al., 2021; Hettiarachchi et al., 2018; Mansoor and Shahzad, 2020	NA

CFA = Confirmatory Factor Analysis, NA = Not applicable.

A number of scales have been utilized to measure peer victimization in the studies. It is notable that those used as a part of multisite studies or RCTs were more likely to have a description of the validation. Of the 17 scales used, 7 did not have any details regarding the validation of the translated version of the scales. A set of studies (Thakkar et al., 2019, 2021) used a list method of peers nominating bullies and victims and the method of victimization that was witnessed. A similar approach was utilized by (Sethi et al., 2019) using the Korean Peer Nomination Inventory in Hindi. All other scales were self-report or interviewer-administered.

No information on the assessment of bullying or victimization was provided for 4 of the studies included in the review (Bhatt et al., 2020; Gunawardana et al., 2021; Hettiarachchi et al., 2018; Mansoor and Shahzad, 2020).

3.3. Prevalence data

Data on the prevalence of peer victimization was reported in 31 of the 53 articles that were included in the review. The prevalence of youth having experienced bullying ranged from 4.1% in the past year (Beattie et al., 2019) to 94% of boys and 85% of girls ever being victimized (McFarlane et al., 2017). In community-based studies, the prevalence was varied, with 8% of the girls endorsing a harassing/bullying environment in school, 8% endorsing eve-teasing, 16.6% of Child care home residents endorsing having been bullied, 24.3% reporting being bullied by siblings, to 79.1% in a cohort study (Beattie et al., 2019; Bhatt et al., 2020; Haque et al., 2021; Nguyen et al., 2019; Nguyen et al., 2020a).

The prevalence of bullying in Bhutan in the GSHS was 25.6%, while in Bangladesh, the population weighted percentage was 24.43% (Dema et al., 2019; Irish and Murshid, 2020; Murshid, 2017; Shaikh et al., 2019). The combined weighted prevalence of bullying experience in Sri Lanka, Pakistan and Myanmar was 37.5% in the GSHS. In the study of the GSHS sample from Pakistan, 15% of the total sample had experienced bullying (Shaikh et al., 2019). In the results of the GSHS survey from Nepal (Neupane et al., 2020; Pandey et al., 2021), 50.86% of the respondents endorsed having been bullied in the past month.

The prevalence of bullying perpetration ranged from about 16% to 85% of boys and 66% of girls being involved in bullying perpetration ever. The prevalence of cyberbullying was reported only in Bangladesh and India in a total of 4 studies. In Bangladesh (Mallik and Radwan, 2020), cybervictimization was experienced by 31.9% of students from a school in Dhaka. In India, as a part of the cross-national ECAMH study (Chudal et al., 2021), cybervictimization only was reported by about 1.3% of the students, while 10.5% of students from a school in West Bengal had reported being victimized online (Mukherjee et al., 2019). Another study from India (Sharma et al., 2017) found that 7.5% of the students engaged in cyberbullying while 17.2% had been victims.

3.4. Associations of bullying

Across studies, boys scored higher on both victimization and aggression than girls, which was evident both in cross-sectional studies and longitudinal studies. (Menon and Hannah-Fisher, 2019; Mishra et al., 2018; Rana et al., 2020; Thakkar et al., 2019). Notably, while boys endorsed significantly higher rates of victimization and physical and non-physical aggression, girls were noted to have significantly higher verbal bullying compared to boys (Rana et al., 2020; Sharma et al., 2020). In Nepal, it was observed that bullies were more (55.8%) among the children who came from a higher social standing/caste (Janajatis), while the victims were more (64.86%) among the disadvantaged castes (Janajatis). This was similar to a study done in India which found that Hindu boys were more likely to be bullies and general caste and non-Hindu children more likely to be victims (Mishra et al., 2018; Thakkar et al., 2019). In a related study among school students in India (Sethi et al., 2019), it was noted that students belonging to the lowest SES had the least odds of being a bully. Bullying behaviours were generally lower in public schools than in private schools and regular schools compared to special schools for children with mild Intellectual disabilities (Nambiar et al., 2020).

Significant associations have been found in relation to bullying behaviours (for victims, bullies and bully victims) and various mental health outcomes, including depressive symptoms (OR = 1.97) (Bhatt et al., 2020), Anxiety symptoms (aOR = 2.40–6.00) (Khan and Khan, 2020; Neupane et al., 2020), psychological distress (OR = 2.66–3.44) (Lee et al., 2019), suicidal ideas and attempts (OR = 2.23–5.88 & OR = 2.94–6.60) (Irish and Murshid, 2020; Rahman et al., 2020), and tobacco use (aOR 1-93-2.05) (Khan et al., 2020; Neupane et al., 2020). Both externalizing and internalizing symptoms were significantly higher when considering combined bullying as compared to either traditional or cyberbullying in both sexes (Chudal et al., 2021). Cybervictimization among school students (Mallik and Radwan, 2020) was noted to be significantly associated with any psychiatric disorder (27.27%), major depressive disorder (9.09%), any emotional disorder (32.59%) and any behavioural disorder (12.5%).

Peer victimization was also associated with physical health outcomes, including risk of serious injuries (aOR = 2.73) (Pandey et al., 2020), being in physical fights (OR = 2.04–3.39) (Neupane et al., 2020; Pandey et al., 2020; Shaikh et al., 2019) and sleep loss (being bullied more than 3 days increased severe sleep loss over worry aOR = 6.0) (Wang et al., 2020). Bullying was also associated with a higher risk of school absenteeism and school dropouts (Prakash et al., 2017).

Among the various contributors and associated factors with bullying, felt pressure, work sexism, and entity view of gender differences were associated with higher victimization. This was more evident in girls. Similarly, felt atypicality was associated with higher victimization in younger children with high levels of felt pressure and work sexism (Menon and Hannah-Fisher, 2019). In a unique study between the Netherlands and India (Pronk et al., 2017), the authors found that different roles in the bullying role behaviours (Bully, victim, follower, defender and outsider) have clear associations with the peer group status (preference and popularity). Positive associations with popularity were found for the bully, follower and defender, while defenders and outsiders positively associated with preference across the 2 countries. In Pakistan (Abid et al., 2017), a negative correlation between mindfulness and bullying behaviour was documented. It was also noted that good hygiene (Murshid, 2018), not using tobacco, a physically active lifestyle, and having more than 7 friends were identified as factors protective against being bullied (Shah et al., 2019).

In a clinical sample of children in an obesity/overweight clinic in Sri Lanka (Gunawardana et al., 2021), psychosocial quality of life was found to be significantly lower in children subject to bullying. In a longitudinal study (Thakkar et al., 2021), it was noted that higher BMI was associated with an increased risk of being an overt bully for boys. There was no significant association observed between self-harm behaviour and bullying in the children in the Juvenile Justice System in Sri Lanka (Hettiarachchi et al., 2018).

3.5. Effect of interventions on bullying

A cluster randomized controlled trial (Shinde et al., 2018) conducted in Bihar, India, delivered a multicomponent (whole school, group and targeted) intervention—SEHER that aimed to improve the school climate and thus have an impact on health outcomes of the youth. This study used a counsellor (SEHER Mitra—SM) and a teacher as SEHER Mitra (TSM) to deliver interventions. It demonstrated that SM was significantly superior to both TSM and the control group in reducing bullying, violence victimization, or perpetration.

In India, the Setu (Bridge) intervention (Sharma et al., 2020) was developed in conjunction with educators and public health professionals. Culturally sensitive methods including storytelling, group discussions, mindfulness and various activities were utilized to deliver four 2-hour sessions to students in schools that addressed teaching mindfulness, problem-solving, empathy and effective communication. There was a significant difference from the baseline for physical aggression at one month post-intervention. Non-physical aggression, victimization and physical aggression showed significant reductions within 6 months of the intervention. The effect sizes were medium for non-physical aggression (−0.5) and victimization (−0.4) and large for physical aggression (−0.9). The intervention was also found to be more effective for the older children compared to the younger ones.

The intervention “Right to Play” in Pakistan (Karmaliani et al., 2020) used a play-based intervention for 11–12 years to target depressive symptoms. The intervention included manualized play-based learning activities delivered by adult coaches followed by a three-step discussion (Reflect-Connect-Apply) that encouraged reflection on the activity and had the children apply their learnings. The study showed significant reductions in the mean scores in the secondary outcome as measured by the Peer perpetrating and victimization scale post-intervention, with an estimated mean difference in the scores ranging from -0.79 to -1.98 among boys and girls. The children receiving the intervention also showed less negative affect on the experience of physical punishment at school and home and more positive gender attitudes.

3.6. Other information related to bullying

A unique study from Pakistan (Shamsi et al., 2019) assessed teachers’ knowledge about bullying and its effects using the Peer Relations Assessment Questionnaire - Revised for Teachers. It reported that even though 81% of the teachers could define bullying correctly, less than a third acknowledged mimicking, teasing, blackmailing or spreading rumours as an act of bullying. Similarly, headaches, depression or low self-esteem were identified as physical or emotional consequences of the same. Only one study from Pakistan (Mansoor and Shahzad, 2020) established the psychometric properties of the Urdu version of the Adolescent Peer relation instrument.

In the longitudinal follow up of the Young Lives study (Nguyen et al., 2019), it was noted that while at age 15, victimization was associated with more significant emotional difficulties, on follow up, the association was attenuated and non-significant. The study concluded that outcomes of peer victimization are not unique, and multiple potential pathways can be areas of intervention. A latent class analysis of the data from the Young Lives study (Nguyen et al., 2020) identified that males were significantly more likely to be victimized than females and urban students were less likely to be victimized compared to their rural counterparts.

Cyberbullying is a new and developing area of indirect victimization with longstanding effects, and in this review, 5 of the 53 studies had some information about the same. Face to face victimization was noted to provide more negative affect as compared to cyber/private victimization, and the response was more of aggressor blame in 2 of the studies (Wright et al., 2017; Wright et al., 2017b).

4. Discussion

The prevalence of bullying has varied across different studies. A meta-analysis done by Modecki et al., (2014) across 80 international studies summarized the mean prevalence for traditional bullying involvement at 35% and cyberbullying at 15%. In our review, the prevalence of bullying victimization has wide variation from as low as 4% to as high as 94%. This might be due to the difference in the way bullying was defined. While the CDC (Gladden et al., 2014) has standardized the definitions of the different types of bullying, including physical, verbal, relational and property damage, it may not be possible to eliminate the societal and cultural influences on the understanding of bullying (Campbell et al., 2018). These could vary from the language used to describe specific forms of bullying to cultural phenomena such as individualism vs collectivism or power distance or masculinity or femininity. This may have an additional impact on how the questions related to bullying are framed in the countries involved in this review.

Similar to studies done in high income countries (Wolke and Lereya, 2015), significant associations between mental health problems and bullying were seen among bullies, victims and bully victims. This is corroborated in the studies included in this review for depression (Asad et al., 2017; Murshid, 2017; Naveed et al., 2019), anxiety (Khan and Khan, 2020; Neupane et al., 2020), physical violence and injuries (Pandey et al., 2021; Shaikh et al., 2019) and substance use (Khan et al., 2020; Neupane et al., 2020). Suicidal thoughts and behaviours were analysed in the meta-analysis performed by van Geel et al. (2014) among bullied youth (OR = 2.23 and 2.55, respectively). In comparison, in a study conducted in Bhutan (Dema et al., 2019), suicidal behaviour was lower (aPR of 1.3 and 1.6, respectively), while the one conducted in Bangladesh (Irish and Murshid, 2020) had nearly comparable odds (2.23 and 2.94 respectively). This could indicate a possible effect of the psychosocial stressors and strengths present in the youth in different countries.

One of the key findings of a meta-analysis by Tippett & Wolke in 2014 showed the odds of victims and bully victims being higher in the lower socioeconomic status, and this was reflected in a few studies done in Nepal and India, where disadvantaged classes were more likely to be bullied (Mishra et al., 2018; Sethi et al., 2019). Unlike the meta-analysis, the studies identified that the youth from relatively advantaged backgrounds were more likely to be bullies. This was also seen in a study from schools in India (Rana et al., 2020) which found that the youth in government schools were significantly less likely to report either victimization or bullying behaviours than adolescents from private schools.

A 2017 systematic review of antibullying interventions found only 18 studies that met the quality threshold, and all were conducted in high income countries (Silva et al., 2017). Without measuring the quality of the studies, our review identified three unique interventions addressing bullying through the development of either play skills (McFarlane et al., 2017), the teaching of specific skills (Sharma et al., 2020) or the use of unique providers (Shinde et al., 2018) that were useful in the cultural contexts of the Southern Asia Region. This review also highlights the need for a multistakeholder, multilevel intervention. At the children's level, gender and social equality issues need to be addressed, while skills such as empathy and mindfulness through interactive interventions using young facilitators need to be developed. At the level of teachers and the school, sensitization about the kinds of bullying, the effects on physical health and academic well-being and identification of risk factors (such as being overweight among boys at a young age) is essential. They also need to be active partners in promoting a healthy school climate that facilitates positive interactions among youth while reducing the risk and effects of victimization. At the policy level, cyberbullying needs to be taken seriously as it impacts children and adults. Locally relevant forms of victimization such as eve teasing and socially relevant risk factors such as casteism/religious discrimination would require policy level interventions.

The strength of this review lies in capturing important information about bullying in children and adolescents that is associated with significant physical and mental health consequences. This review focuses on a group of LMICs with a shared history and similar socio-cultural background, which lends itself to a translation of locally relevant information to implementation, such as eve-teasing, caste, and BMI's role in perceived/victimization and the emerging trends in cyberbullying. One of the unique aspects of this review is addressing the issue of defining and measuring bullying in the research done in SAARC countries. It highlights the issue of transferring a concept en masse without appropriate contextualization. The review also captures the little work done with other stakeholders, including teachers and school systems, to create a safe environment for the children.

As with any scoping review, one of the primary limitations has been that the quality of the individual studies was not taken into consideration. This review also restricted itself to the past 5 years to elicit the latest information on the topic of bullying. The scoping review did not consider the research from grey literature, and non-peer reviewed research publications. This may be one of the reasons why no information on studies related to policy or legal effects of antibullying measures was available. Furthermore, it is essential to recognize that the data collected in most of the studies related to the epidemiology and risk factors associated with bullying were parts of a multicounty comprehensive data set from which the associations were made within the past 5 years. Bullying was not the primary focus of the data collection. This may limit the generalizability of the conclusions drawn from the studies.

5. Future directions

The evidence gathered from the studies related to bullying shows a significant impact on the health and well-being of young people, even in the SAARC countries. This review lays the groundwork for the following future work:

• •Creation of contextual and socially appropriate definitions of bullying that can be standardized for the SAARC countries. This may apply to other countries in the South-East Asia region.
• •Focus on bullying in the community and its longitudinal effects.
• •A focus on cyberbullying, especially after the transposition of academics to an online medium due to the COVID-19 pandemic
• •Creating cross-country partnerships in developing contextually relevant interventions co-designed with the youth that can be implemented in the community and academic settings.
• •The perception and subjective experience of bullying, either in person or online, in the context of multiple other adverse childhood experiences experienced by the youth in the SAARC nations need to be studied.
• •The policy decisions to make the academic environment bullying-free are essential to enhance youth's learning and emotional outcomes.

Declarations

Author contribution statement

All authors listed have significantly contributed to the development and the writing of this article.

Funding statement

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Data availability statement

Data will be made available on request.

Declaration of interests statement

The authors declare no conflict of interest.

Additional information

No additional information is available for this paper.