Youth self-harm in low- and middle-income countries: Systematic review of the risk and protective factors

Abstract

Background:

Self-harm (defined here as an act of intentionally causing harm to own self, irrespective of the type, motive or suicidal intent) is one of the strongest antecedents of suicide in youth. While there have been a number of studies of youth self-harm in low- and middle-income countries (LMICs), there is currently no systematic review of studies of prevalence rates and risk and protective factors.

Aim:

To systematically review the evidence relating to the prevalence rates and forms of self-harm in youth in LMICs and its relationship to family economic status, family functioning, relationship with the peer group, social relationships and academic performance.

Methods:

Electronic searches of three databases, MEDLINE, PsycINFO and Scopus, were performed. In total, 27 school-, community- and hospital-based studies evaluating self-harm in LMICs with youth focus (with participants between 12 and 25 years) were included.

Results:

The self-harm was divided into suicidal and non-suicidal depending on the nature of self-harm. The 12-month prevalence rates of non-suicidal self-harm varied from 15.5% to 31.3%, and the range of suicidal behaviour rates was from 3.2% to 4.7%. Banging and hitting were the most common in the community-based studies, followed by cutting, scratching and wound picking. Self-poisoning with pesticides was most commonly reported in hospital-based studies. Risk factors for self-harm were family conflict, peer groups with members indulging in self-harm, truancy and school absenteeism. Protective factors were having an understanding family, having friends and higher school competence. Risk factors for suicidal thoughts and attempts were lack of close friends and history of suicide by a friend.

Conclusion:

The 12-month prevalence rates of youth self-harm in LMICs are comparable to high-income countries (HICs). The profile of risk and protective factors suggests that family-based interventions could be useful in treatment and prevention. Future studies should aim for greater consistency in assessment methods and the constructs under evaluation.

Keywords

Self-harm youth low- and middle-income countries risk factors protective factors

Background

Suicide in youth is a major public health concern in low- and middle-income countries (LMICs; McKinnon, Gariepy, Sentenac, & Elgar, 2016; Sidhartha & Jena, 2006). It is the second biggest cause of death in those aged 15–29 years in India (Patel et al., 2012), the world’s largest democracy and the home to its largest adolescent population. Self-harm is defined as an act of intentionally causing harm to own self, irrespective of the type, motive or suicidal intent (Hawton et al., 2003; National Collaborating Centre for Mental Health, 2012). Self-harm predicts an elevated risk of future suicide by 30–100 times as compared to the general population in the following year (Cooper et al., 2005; Hawton et al., 2015; Whitlock et al., 2013). The studies in adults suggest that the risk persists beyond the first year with 1 in 15 dying by suicide after 9 or more years following the self-harm attempt (Owens, Horrocks, & House, 2002). The risk increases with the number of self-harm episodes and is more prominent in those aged 10–24 years as compared to older age groups (Zahl & Hawton, 2004).

Various terms are used to describe self-harm such as self-harm behaviour (Washburn et al.), self-injurious behaviour, deliberate self-harm (DSH), self-mutilatory behaviour (SMB) and suicidal behaviour. Sometimes suicide intent is used to make a dichotomous separation between suicidal and non-suicidal self-harm. Non-suicidal self-injury (NSSI) is one such term defined as a deliberate direct destruction or alteration of body tissue without a conscious suicidal intent (Whitlock et al., 2013). The substantial overlap of suicidal and non-suicidal self-injurious behaviours, with the dimensional quality of suicidal intent makes this classification arbitrary (Kapur et al., 2013). Thus, many standard guidelines focus on self-harm irrespective of the intent (Hawton et al., 2003; Hawton, Saunders, & O’Connor, 2012; National Collaborating Centre for Mental Health, 2012; Swahn et al., 2012).

Our understanding of self-harm in youth and its relationship with suicide is based on mostly the research conducted in high-income countries (HICs) (Muehlenkamp, Claes, Havertape, & Plener, 2012). About 10% of adolescents in the community harm themselves at some point (Hawton, Rodham, Evans, & Weatherall, 2002; Moran et al., 2012). The risk factors for self-harm and suicide in youth include socio-economic disadvantage and psychiatric illnesses such as depression, anxiety disorders and substance use problems (Hawton et al., 2012; Skegg, 2005). Other important individual- and family-related risk factors include adverse childhood experiences, interpersonal difficulties, parental separation, divorce or death, parental mental disorder and family history of suicidal behaviour (Hawton et al., 2012). Additionally, characteristics related to self-harm that are associated with suicide include male gender, self-harm attempt requiring a hospital presentation and onset of self-harm early in life (Hawton et al., 2012).

A longitudinal study exploring the relationship between NSSI and suicidal behaviour (suicidal-plan and -attempt) in young adults (baseline average age of 20.3 years) found having parents as confidants was protective against transition from NSSI to suicidal behaviour (Whitlock et al., 2013). A good social support, confiding relationship with the family members, religious affiliations, cultural norms and optimistic outlook are some of the other protective factors against self-harm.

The paucity of evidence in this area in LMICs impedes our understanding of origin, purpose and forms of self-harm in youth. Additionally, the societal and cultural differences between LMICs and HICs could have a significant impact on the overall presentation and course of self-harm in youth (Amitai & Apter, 2012; Muehlenkamp et al., 2012). This is important as the role of overarching culture and specific transcultural factors such as the media environment and family structure in self-harm is becoming increasingly clear. There is growing evidence that some cultural factors offer protection from self-harm (Skegg, 2005). For example, the suicide rates are lower in some Catholic and Islam countries where it is strongly disapproved (Skegg, 2005). Furthermore, identifying and addressing the context-specific risk and protective factors are likely to improve the effectiveness of interventions that aim to reduce self-harm.

With this background, we systematically reviewed the current available evidence on self-harm in youth in LMICs with the aim of answering the following research questions:

What is the prevalence of self-harm in youth in LMICs and is it different from HICs?

What are the most common forms of self-harm in youth in LMICs, and in particular, what proportion is suicidal self-harm?

What is the relationship between family economic status, family functioning, peer group, social relationships, academic performance and self-harm in youth in LMICs?

This article is focused on self-harm in 12–25 years old. We chose this age range for a number of reasons, including the lengthening of the period of adolescence with the earlier onset of puberty and the delay in the social transitions related to adulthood such as marriage (Patton et al., 2016). The age range of the current review is broad enough to capture the developmental trajectory of this developmental phase and includes mid-adolescence, a period of higher prevalence of self-harm (Patton et al., 2007). This age range also ensures developmental consistency, as young adults are relatively close in age to older adolescents (Hawton et al., 2012). In most LMICs, young people of this age are still financially dependent on their family members (Lai, 2011). We have included the research articles published in Turkish language to make the review more comprehensive. To our knowledge, this is the first review of its kind.

Method

A systematic search of bibliographic electronic databases, such as MEDLINE, PsycINFO and Scopus, was performed in August 2015. The Covidence software that is used as a primary screening, and data extraction tool for Cochrane authors was used to assist in the production of the systematic review. The following terms formed the basis of the search strategy: ‘self-harm’ OR ‘deliberate self-harm’ OR ‘DSH’ OR ‘self-injurious behavior’ OR self-injury OR ‘self-inflicted wounds’ OR ‘suicide attempt’ OR ‘suicidal behaviour’ AND ‘low middle income country/ies’ OR ‘an individual country from LMIC list (WHO Classification, 135 countries)’. Further papers were identified through hand searching the references of all subsequently included studies and relevant reviews (Figure 1).

Figure 1.

PRISMA study selection flow chart.

Two authors, including the first author, independently screened the titles of all retrieved articles and then conferred to resolve any differences of opinion. Thereafter, the same authors assessed the relevance and appropriateness of the selected articles by screening the abstracts. The full text versions of potentially relevant articles after screening were assessed for inclusion in the review. The first author and a co-author classified all studies independently, and disagreements were resolved through discussion in person. The nature of the self-harm (suicidal vs non-suicidal) was decided by the description provided in the study and discussed by the two authors to reach a joint consensus. For example, the construct of DSH has been used to describe both suicidal and non-suicidal self-harm in different studies (Sidhartha & Jena, 2006; Wan, Hu, Hao, Sun, & Tao, 2011).

Inclusion criteria

Cross-sectional, case–control and longitudinal studies were included if they were conducted in a school, community or a hospital setting targeting self-harm or suicidal behaviour, in youth in LMICs. All the included studies had a youth person focus. There were no participants over 25 years of age in any of the studies included.

Exclusion criteria

The following studies were excluded:

Systematic reviews, correspondence, personal viewpoints, brief reports and other non-data articles;

Studies in specific populations in which self-harm is a comorbid condition and not the primary concern or the presenting problem, for example, self-harm in psychosis;

Studies in which the sample mainly consisted of the older adult population with a small proportion of adolescents.

The review conforms to the recommended guidelines by PRISMA 2009 checklist. The quality assessment was performed using the following scales: Newcastle–Ottawa Scale adapted version for cross-sectional studies, Newcastle–Ottawa Scale for case–control studies and quality assessment tool for systematic reviews of observational studies (QATSO) scale for longitudinal studies (Modesti et al., 2016; Wells et al., 2000; Wong, Cheung, & Hart, 2008).

Results

In total, 27 studies were included in the review. Of these, 16 were conducted in the school or community setting and 11 were hospital-based studies. In total, 14 studies were from Turkey and 4 each from China and India. There was one study from Albania, Lithuania, Pakistan, Seychelles and Zambia that met the inclusion criteria (Table 1).

Table 1.

Type of studies and countries.

Country	English language, school and community based	English language, hospital based	Non-English language, school and community based	Non-English language, hospital based
Albania		1
China	4
India	3	1
Lithuania	1
Pakistan		1
Seychelles	1
Turkey	5		1	8
Zambia	1

Prevalence of self-harm

Results showed considerable variation in 12-month prevalence rates of self-harm rates in youth in LMICs. Comparing the prevalence rates between studies is complicated by large variations in methodology including differences in the constructs under evaluation and in the ages of the included populations (Appendices 1 –4). Eight studies assessed self-harm prevalence rates over the previous 12 months (Table 2). One study evaluated both non-suicidal and suicidal self-harm separately and tried to explore the relationship between them (Tang et al., 2011). All the others focused on either suicidal or non-suicidal self-harm (Evren, Evren, Bozkurt, & Can, 2014; Kharsati & Bhola, 2015; Muula, Kazembe, Rudatsikira, & Siziya, 2008; Sidhartha & Jena, 2006; Somer et al., 2015; Wan et al., 2011; Wilson, Dunlavy, Viswanathan, & Bovet, 2012).

Table 2.

12-month prevalence (school and community based).

Author (year); country	Age (years), M (SD)	Sample size	NSSI/DSH/SHB	12-month prevalence (%)
Studies evaluating non-suicidal self-harm
Wan, Hu, Hao, Sun, and Tao (2011); China	16.1 (2.82)	17,622	DSH	17.00
Evren, Evren, Bozkurt, and Can (2014); Turkey (Istanbul)	15.58 (2.85)	4,957	SHB	14.38
Kharsati and Bhola (2015); India	20.25 (1.7)	470	NSSI	31.20
Somer et al. (2015); Turkey (Izmir and outlying counties)	16.8 (1.26)	1,656	NSSI	31.30
Studies evaluating suicidal and non-suicidal self-harm
Tang et al. (2011); China	15.6 (2.8)	2,013	NSSI, SI, SA	NSSI: 15.50, SI: 8.8, SA: 3.5
Studies evaluating suicidal self-harm
Sidhartha and Jena (2006); India	14.73 (1.44)	1,205	DSH	SB: 6.10, SA: 3.5
Muula, Kazembe, Rudatsikira, and Siziya (2008); Zambia	13–15 (not specified)	2,257	SI	SI: 31.3
Wilson, Dunlavy, Viswanathan, and Bovet (2012); Seychelles	14 (1.5)	1,427	SE	SE: 16, SI + SP: 6.2

NSSI: non-suicidal self-injury; SI: suicidal ideation; SA: suicidal attempt; DSH: deliberate self-harm; SE: suicidal expression; SP: suicidal plan; SHB: self-harm behaviour; SD: standard deviation.

The rates of non-suicidal self-harm varied from 15.5% to 31.3% (Table 2). The variation is partly been explained by the method of assessment. For example, the rates were higher in the cross-sectional surveys with standardized scales as compared to the interview-based studies or those using branched assessment method (evaluation of self-harm using a single question followed by exploratory questions to ascertain the nature of self-harm in case of an affirmative response; Evren et al., 2014; Kharsati & Bhola, 2015; Somer et al., 2015; Wan et al., 2011).

There was a very little variation in the rates of suicide attempt in the studies. However, the rates of suicidal thoughts in adolescents in Zambia were much higher as compared to China and India (Muula et al., 2008; Sidhartha & Jena, 2006; Tang et al., 2011; Table 2).

Form of self-harm

In total, six school- and community-based (Aktepe, Çalişkan, & Sönmez, 2014; Kharsati & Bhola, 2015; Somer et al., 2015; Tang et al., 2011; Wan et al., 2011; Zoroglu et al., 2003) and nine hospital/clinic-based studies reported on the forms of self-harm (Akdemir, Zeki, Ünal, Kara, & Çetin, 2013; Akin & Berkem, 2012; Çelik et al., 2011; Fidan, Ceyhun, & Kirpinar, 2009; Fidan & Kirpinar, 2009; Gökçen & Köylü, 2011; Grover, Sarkar, Chakrabarti, Malhotra, & Avasthi, 2015; Kola, Kola, Petrela, & Zaimi, 2013; Soylu, Taneli, & Taneli, 2013; Syed & Khan, 2008). Banging and hitting were the most common form of self-harm in the community-based studies, followed by cutting, scratching and wound picking (Aktepe et al., 2014; Kharsati & Bhola, 2015; Somer et al., 2015; Tang et al., 2011; Wan et al., 2011; Zoroglu et al., 2003). Burning and use of chemicals were the least common forms of self-harm (Wan et al., 2011; Zoroglu et al., 2003). In the hospital-based studies, self-poisoning was the commonest method (Akin & Berkem, 2012; Çelik et al., 2011; Fidan et al., 2009; Fidan & Kirpinar, 2009; Gökçen & Köylü, 2011; Grover et al., 2015; Kola et al., 2013; Soylu et al., 2013; Syed & Khan, 2008). Pesticides were a common choice of poisoning (Grover et al., 2015; Kola et al., 2013; Syed & Khan, 2008; Tables 3 and 4). The details of the measurement tools used in the various studies are shown in the appendices.

Table 3.

Form of self-harm (school/community-based studies).

Author (year); country	Sample size; mean age (SD) (years)	Construct	Form
Wan, Hu, Hao, Sun, and Tao (2011); China	17,622; 16.1 (2.82)	DSH	Self-hitting (8.2%), pinching (6.7%), pulling hair (6.2%), commonest f/b banging head (4.1%), biting (2.9%), cutting (2.3%), scratching (1.9%) and firing/burning (0.4%)
Tang et al. (2011); China	2,013; 15.6 (2.8)	NSSI	Self-hitting, hair pulling, wound picking and biting self (10.5% – considered as minor self-harm)Cutting/carving, burning, self-tattooing and scraping (5% – moderate/severe)
Kharsati and Bhola (2015); India	470; 20.25 (1.7)	NSSI	Self-hitting – 15.2%, cutting/carving skin – 13.2%, biting self – 10.4% and wound picking – 10.3%
Somer et al. (2015); Turkey (Izmir and outlying counties)	1,656; 16.8 (1.26)	NSSI	Banging/hitting–50.8%, cutting–44.1% and wound picking–46.9%.
Aktepe, Çalişkan, and Sönmez (2014); Turkey	1,358; 15.5 (1.1)	NSSI	Hitting – 27.9%, cutting – 20.9%, preventing wound healing – 14%, biting – 13.4% and drug overdose – 13.18%
Zoroglu et al. (2003); Turkey	862; 15.9 (1.8)	SMB	Banging/hitting – 11.3%, cutting – 8.4%, pulling hair – 4.3% (F > M), cigarette burns – 2.1% and other types – 2.3%

DSH: deliberate self-harm behaviour; NSSI: non-suicidal self-injury; SHB: self-harm behaviour; SMB: self-mutilatory behaviour; F: female; M: male; SD: standard deviation.

Table 4.

Form of self-harm (clinic/hospital-based studies).

Author (year); country	Sample size; mean age (SD) (years)	Construct	Form
Kola, Kola, Petrela, and Zaimi (2013); Albania	582; 10–19 (not specified)	SA	Intoxication (97.6%) – mainly phystoxin (germ killer, in 35% of intoxication cases); other ways such as hanging or vein cutting almost inconsiderable (2.4%)
Grover, Sarkar, Chakrabarti, Malhotra, and Avasthi (2015); India	101; 17(1.6)	ISH	Ingestion of insecticides (65%); corrosives (10.9%); and violent methods such as stabbing, hanging, jumping from height and self-immolation used by 11 (9 males)
Syed and Khan (2008); Pakistan	69; 16 (2.2)	DSH	Benzodiazepine overdose (30.4%) f/b insecticide ingestion
Akdemir, Zeki, Ünal, Kara, and Çetin (2013); Turkey	48; 14.7 (1.2)	NSSI	Self-cutting, scraping, piercing skin, head/hand hitting, nail biting and hair plucking
Soylu, Taneli, and Taneli (2013); Turkey	63; 16.01 (1.31)	SI, SA	Medications, toxic drug ingestion, cutting (14.8%), jumping in front of vehicles and jump from height (7.4%)
Akin and Berkem (2012); Turkey	36; 14.74 (1.74)	SA	Overdose drug intake and fall from height
Çelik et al. (2011); Turkey	64; 15.5 (1.7)	SA	Self-poisoning (79.7%), jumping from height (9.4), cutting (6.3%) and other methods (4.7%)
Gökçen and Köylü (2011); Turkey	121; 16.08 (not specified)	SA	All cases-self-poisoning
Fidan, Ceyhun, and Kirpinar (2009), Fidan and Kirpinar (2009); Turkey	30; 17.21 (3.14)	SA	Drug/poison ingestion (90%)

SA: suicidal attempt, ISH: intentional self-harm, SIB: self-injurious behaviour, NSSI: non-suicidal self-injury, DSH: deliberate self-harm, SI: suicidal ideation; SD: standard deviation.

Risk and protective factors

We classified the risk and protective factors into three domains, that is, family-related factors, peer-related factors, and school-related factors (Table 5). We have focused on the school- and community-based studies exploring the risk factors for self-harm for a period of up to 12 months preceding a self-harm episode for the conclusions to be generalizable and the association to be most relevant.

Table 5.

Risk and protective factors for self-harm (school/community-based studies).

Domain	Risk factors	Protective factors
Family	SB Not living with parents Unsatisfying/strained/conflictual relationships Parents divorced and remarried to new partner Physical punishment style of parenting Physical abuse by parents Poor maternal general healthSHB Education of father – senior middle school Perceived family economic status poor	SB Higher family functioning Positive youth development Understanding parentsSHB Positive youth development
Peer	SB Suicide by friend Lack of close friendsSHB Self-injury by best friend risk factor self-injury 6 months later Self-harming youth join self-harming peer groups	SHB Three or more friends – repeat incident Six or more friends – single incident
School	SB Not attending regular school or collegeSHB Below average school performance Higher truancy School absenteeism	SHB Higher academic/school competence

SB: suicidal behaviour; SHB: self-harm behaviour.

Family-related factors

A couple of studies found higher family functioning to be protective against self-harm (Law & Shek, 2013; Shek & Yu, 2012). Additionally, two other studies established unsatisfying/strained/conflictual familial relationships, divorced parents with new partners and negative opinion of family members as risk factors for self-harm and suicidal behaviour (Pillai, Andrews, & Patel, 2009; Shek & Yu, 2012). An understanding set of parents was protective against suicidal behaviour (Wilson et al., 2012). Parental neglect, physical abuse by parents and authoritarian discipline style were the other risk factors identified for suicidal behaviour (Pillai et al., 2009; Sidhartha & Jena, 2006). Perceived and real socio-economic status of the family as poor was significantly related to self-harm (Shek & Yu, 2012; Wan et al., 2011). There was some evidence for the association of poor maternal general health with suicidal behaviour and the education level of fathers having been restricted to senior middle school with self-harm in youth (Liu, Sun, & Yang, 2008; Wan et al., 2011).

Peer-related factors

Lack of close friends (Wilson et al., 2012) and history of suicide by a friend (Sidhartha & Jena, 2006) were identified as risk factors for suicidal thoughts and attempts. In one study, having three or more friends was protective against repeated self-harm and having six or more friends was protective against a single self-harm episode (Wan et al., 2011). In another study, self-injury by the best friend was predictive of self-injury in youth 6 months later. Additionally, youth indulging in self-harm were more likely to join peer groups with self-injuring members at 6-month follow-up (You, Lin, Fu, & Leung, 2013). Also, one study found that the negative opinion of the family members of the peer group was significantly associated with self-harm in youth (Evren et al., 2014).

School-related factors

Below average school performance, truancy, school absenteeism and dropping out of school or college were significantly associated with self-harm (Evren et al., 2014; Pillai et al., 2009). Higher academic achievement and superior school competence were associated with lower incidence of self-harm (Shek & Yu, 2012). Furthermore, there were a few studies reporting a correlation between substance use and self-harm behaviours (Evren et al., 2014; Toprak, Cetin, Guven, Can, & Demircan, 2011; Table 5).

Important findings that need replication

In a large community-based study in a rural India, loss of traditional lifestyles indexed by behaviours such as premarital sex and independent decision-making was associated with increased suicidal behaviour in young women (Pillai et al., 2009).

Discussion

This review, the first of its kind to review the evidence for the prevalence rates of self-harm in youth in LMICs, as well as the associated risk and protective factors, identified 27 studies from eight countries. The prevalence rates of non-suicidal self-harm in LMICs varied from 15.5% to 31.3%. This range is comparable to 12-month prevalence rates of NSSI (7.3%–37.2%) in HICs (Moran et al., 2012; Muehlenkamp et al., 2012). Furthermore, the 12-month prevalence rates of suicide attempts of 3.2%–4.7% are similar to HICs (Cash & Bridge, 2009).

Hitting is the commonest method of non-suicidal self-harm followed by cutting, wound picking and biting (Table 5). These findings are similar to the available evidence from HICs of NSSI generally involving low lethality methods (e.g. cutting, burning and biting) as opposed to suicidal behaviour that tends to involve high lethality methods (e.g. overdose, wrist cutting and hanging; Hamza, Stewart, & Willoughby, 2012).

Self-poisoning is the commonest method of suicide attempt in youth in LMICs according to the hospital-based studies (Akin & Berkem, 2012; Çelik et al., 2011; Fidan & Kirpinar, 2009; Gökçen & Köylü, 2011; Kola et al., 2013; Soylu et al., 2013). In a large hospital-based study conducted in Sri Lanka, the authors found that most of the suicide attempts were using pesticides and were precipitated by triggers in the social environment (Konradsen, Van Der Hoek, & Peiris, 2006). There was no demarcation between suicidal and non-suicidal self-harm in a couple of studies that used constructs such as intentional self-harm and DSH to evaluate self-harm (Grover et al., 2015; Syed & Khan, 2008). Poisoning remained the commonest method in these studies. This could mean that the demarcation between suicidal and non-suicidal self-harm may be comparatively blurred in LMICs.

In many studies, the pesticides stored in the household were used for poisoning (Grover et al., 2015; Kola et al., 2013; Syed & Khan, 2008). An additional method of self-poisoning was the use of pharmaceutical drugs (Kola et al., 2013; Syed & Khan, 2008), many of which are classified as restricted in HICs requiring a doctor’s prescription. These medicines are freely available without a doctor’s prescription in LMICs. These factors become important while planning any suicide/self-harm prevention programmes by restriction of means. For example, a central pesticide storage facility was proven to be an effective measure in reducing completed and attempted suicide in Southern India (Vijayakumar et al., 2013).

There are limited studies that have explored the risk and protective factors for self-harm in youth in LMICs making the comparisons with the evidence from HICs harder. However, this review provides us with a few helpful insights into the risk factors associated with self-harm in youth in this context and what possibly could be protective towards it. The most robust finding of the current review is the importance of family-related factors such as unsatisfying/strained/conflictual familial relationships, divorced parents with new partners and childhood abuse and maltreatment as risk factors for self-harm and suicidal behaviour in youth in LMICs (Table 5). A couple of studies found an association between low paternal education and self-harm (Oksuz & Malhan, 2005; Wan et al., 2011). These associations are similar to those found in studies conducted in HICs (Beautrais, 2000; Skegg, 2005). An adequate parenting is essential for the development of social skills required for effective interpersonal functioning during adolescence, and a lack of it could result in increasing the risk of self-harm (Beautrais, 2000). The protective function of higher family functioning and understanding parents with regard to self-harm in LMICs could be attributable to dense family networks and greater reliance of youth on their family in this context as compared to many western countries (Aggarwal & Berk, 2015a). Thus, interventions with an incorporated family component could work better for self-harm in youth in LMICs. There is some evidence for the effectiveness of interpersonal psychotherapy in self-harm in young people (Washburn et al., 2012). Further research will be needed in LMICs to see how effective the existing evidence-based approaches are for self-harm and whether a family-related component could be incorporated to make them better suited to the local conditions.

Formation of peer groups and establishing friendships is an important psychosocial maturational task of adolescence. It becomes all the more important in LMICs due to the dense and proximal social networks that most of them share (Aggarwal & Berk, 2015b). Furthermore, the social networks act as reliable source of multiple types of support in LMICs (Perkins, Subramanian, & Christakis, 2015). Not participating in these social connections, being excluded from desirable groups or negative feedback from peers could result in negative impacts on the adolescent identity formation (Harter, Stocker, & Robinson, 1996). The self-harm could occur in the friendship groups by selection effects (where adolescents befriend peers already indulging in the behaviour) or socialization effects (where they start doing it to win the approval of peers and for social conformity; Prinstein et al., 2010).

Some indirect association has been found between school-related factors such as below average school performance and negative attitude towards school and school work with suicidal behaviour in studies in HICs (Evans, Hawton, & Rodham, 2004). The higher school functioning was found to be protective against self-harm in a study with poor academic performance, truancy and school absenteeism identified as risk factors for self-harm in youth in many studies (Evren et al., 2014; Shek & Yu, 2012). This could be due to the higher value placed at academic achievement in LMICs and the stress that ensues due to poor academic performance (Aggarwal & Berk, 2015b).

Conclusion

The available evidence suggests that the 12-month prevalence rates of self-harm in LMICs in youth are comparable to those in HICs. The risk and protective factors for self-harm in youth in LMICs identified in the review require further investigation due to the limited evidence and the variability in the assessment methods, as well as the constructs studied. The distinction between suicidal and non-suicidal self-harm is blurred, and these constructs are very often used interchangeably. The association between family functioning and self-harm suggests that family-based interventions could be useful for treating and preventing self-harm in young in LMICs. More research in this area could give us clarity about the mechanisms leading to and maintaining self-harm in youth, the relationship between self-harm and suicide in adolescents and young adults in LMICs and will help in devising effective intervention programmes in this setting.

Footnotes

Appendix

Appendix 4.

Hospital/outpatient clinic-based studies on self-harm (non-English language).

Turkey (Soylu, Taneli, & Taneli, 2013)	University hospital–based cross-sectional study (June 2008–March 2009)	Social, emotional and cognitive factors leading to SI and SA in patients with MDD	63 cases between ages 12 and 18 years Mean age: 16.01 years (SD: 1.31 years)	Children’s Depression Inventory (CDI) State-Trait Anxiety Inventory (STAI-I, II) Beck Hopelessness Scale Coppersmith Self-Esteem Inventory Multi Dimensional Perceived Social Support Scale Strengths and Difficulties Questionnaire (SDQ)	Prevalence: not assessed Form of self-harm: medications, toxic drug ingestion, cutting (14.8%), jumping in front of vehicles and high-jump (7.4%) Risk factors for SI: high depression and anxiety levels, hopelessness and low self-esteem. Risk factors for SA: separated family background, lower family support and conduct difficulties	5/10, as per Newcastle–Ottawa Scale adapted for cross-sectional studies
Turkey (Elif and Berkem, 2013)	University hospital–based case–control study (June 2009–May 2010)	Adverse Life events in adolescent SA	36 cases between ages 12 and18 years Mean age: 14.74 years (SD: 1.74 years) 34 controls between ages 12 and 18 years Mean age: 14.42 years (SD: 1.43 years)	Adverse Life Events List	Prevalence: not assessed Form of self-harm: not described Risk factors: adverse life events, poor school performance, family history of mental/emotional problems, negative financial state of family, separation from significant other and school failure	4/10, as per Newcastle–Ottawa Scale adapted for cross-sectional studies
Turkey (Akdemir, Zeki, Ünal, Kara, & Çetin, 2013)	University hospital–based cross-sectional case–control study	NSSI and its a/w psychiatric symptoms, identity formation, self-esteem	48 cases between 12 and 17 years Mean age: 14.7 years (SD: 1.2 years) 49 controls between 12 and 17 years Mean age: 14.7 years (SD: 1.5 years)	Brief Symptom Inventory (BSI) Sense of Identity Assessment Form (KDE) Rosenberg Self-Esteem Inventory (RSES)	Prevalence: not assessed Forms of self-harm: self-cutting, scraping, piercing skin, head/hand hitting, nail biting and hair plucking Risk factors: F > M, low self-esteem, associated psychiatry behaviours such as MDD and conduct disorder	6/10, as per Newcastle–Ottawa Scale adapted for cross-sectional studies
Turkey (Akin & Berkem, 2012)	University hospital–based retrospective case–control study from 2009 to 2010	SA and its a/w anger and impulsivity	36 cases of SA between ages 12 and 18 years Mean age: 14.74 years (SD: 1.74 years) 34 healthy subjects as controls Mean age: 14.42 years (SD: 1.43 years)	Kiddie Schedule for Affective Disorders and Schizophrenia Present and Lifetime Version Spielberg State-Trait Anger Expression Inventory (SSTAEI) Barratt’s Impulsivity Scale (BIS) Suicide Intent Scale (SIS) Wechsler Intelligence Scale for Children–Revised (WISC-R)	Prevalence: not assessed Form of self-harm: overdose drug intake and fall from height Risk factors: F > M, depression, ADHD, conduct disorder, anger a/w SA and impulsivity a/w SA	4/10, NOS-A
Turkey (Çelik et al., 2011)	University hospital–based cross-sectional assessment	SA and its r/w self-image, domestic violence and socio-demographic variables	64 adolescents 12–18 years Mean age: 15.5 years (SD: 1.7 years), M: 25%, F: 75%	Offer Self-Image Questionnaire (OSIQ) Child Depression Inventory (CDI) State-Trait Anxiety Inventory For Children (STAIC) Martial Adjustment Test (MAT) Diagnosis and severity evaluated according to DSM-IV criteria and Clinical Global Impression Scale	Prevalence: first SA (70.3%), second SA (18.8%) and third SA (10.9%) Form of self-harm: self-poisoning (79.7%), jumping from height (9.4), cutting (6.3%) and other methods (4.7%) Risk factors: mood disorders, substance use, conflict with family a/w first SA, subsequent SA a/w problems in romantic relationships and grief and loss	4/10, NOS-A
Turkey (Gökçen & Köylü, 2011)	University hospital–based study from 2009 to 2010	SA	121 adolescents 12–18 years, mean age: 16.08 years, M: 17.4%, F: 82.6%	Hospital administered ‘Suicide Attempt Record Form’ and evaluation by the hospital ‘Psychosocial Support Unit’	Prevalence: first SA (81.8%) and second SA (18.2%) Form of self-harm: all cases self-poisoning Risk factors: school problems (38.8%), problems with other gender (21.5%), familial problems (14%) and other (25.7%) Preponderance of female gender in sample, 23.1% treated for psychiatric disorders in previous 6 months
Turkey (Fidan, Ceyhun, & Kirpinar, 2009)	University hospital–based case–control study	SA and family functionalitySA and coping strategies	30 cases of SA between ages 11 and 22 yearsMean age: 17.21 years (SD: 3.14 years)	Socio-demographic FormFamily Assessment ScaleHamilton Depression Inventory (HAM-D)Coping Strategies Scale (COPE)	Prevalence: not assessedForm of self-harm: drug/poison ingestion (90%)Risk factors: F > M, SA a/w family conflicts/discord, tuition failure, problems a/w emotional relationships, SA a/w stressful life events, lack of coping mechanisms such as positive reinterpretation, seeking social support and mental disengagement	4/10, NOS-A

DSH: deliberate self-harm; SHB: self-harm behaviour; SIB: self-injurious behaviour; NSSI: non-suicidal self-injury; NFSB: non-fatal suicidal behaviour; SB: suicidal behaviour; SA: suicide attempt; SI: suicidal ideations; a/w: associated with; r/t: related to; r/w: relation with; NOS-A: Newcastle–Ottawa Scale adapted; F: female; M: male.

Conflict of interest

The author(s) declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: M.B. is supported by a NHMRC Senior Principal Research Fellowship 1059660. G.P. is supported by a NHMRC Senior Principal Research Fellowship 1117873. N.R. is supported by a NHMRC Career Development Fellowship 1083394.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Aggarwal

Berk

(2015a). Evolution of adolescent mental health in a rapidly changing socioeconomic environment: A review of mental health studies in adolescents in India over last 10 years. Asian Journal of Psychiatry, 13, 3–12.

Aggarwal

Berk

(2015b). Nonsuicidal self-injury in Indian adolescents: Nonexistent or unacknowledged? International Journal of Social Psychiatry, 61, 516–517.

Akdemir

Zeki

Ünal

D. Y.

Kara

Çetin

F. Ç

. (2013). Identity status and self-esteem in adolescents with non-suicidal self-injurious behavior. Anatolian Journal of Psychiatry / Anadolu Psikiyatri Dergisi, 14, 69–76.

Akin

Berkem

(2012). Anger and impulsivity in adolescents attempting suicide. Marmara Medical Journal, 25, 148–152.

Aktepe

Çalişkan

A. M.

Sönmez

(2014). Self-injurious behaviour in high school students in city center of Isparta and factors psychiatrically related with this behavior. Anatolian Journal of Psychiatry/Anadolu Psikiyatri Dergisi, 15, 257–264.

Amitai

Apter

(2012). Social aspects of suicidal behavior and prevention in early life: A review. International Journal of Environmental Research and Public Health, 9, 985–994.

Beautrais

A. L.

(2000). Risk factors for suicide and attempted suicide among young people. Australian and New Zealand Journal of Psychiatry, 34, 420–436.

Cash

S. J.

Bridge

J. A.

(2009). Epidemiology of youth suicide and suicidal behavior. Current Opinion in Pediatrics, 21, 613–619.

Çelik

Yildirim

Metin

Ö.

Tahiroğlu

Toros

Avci

Karayazi

. (2011). Psychiatric disorders, self and family functions in adolescents with suicide attempts. Anatolian Journal of Psychiatry / Anadolu Psikiyatri Dergisi, 12, 280–286.

10.

Cooper

Kapur

Webb

Lawlor

Guthrie

Mackway-Jones

Appleby

(2005). Suicide after deliberate self-harm: A 4-year cohort study. American Journal of Psychiatry, 162, 297–303.

11.

Elif

Berkem

(2012). Evaluation of psychiatric diagnosis, demographic and clinical features in adolescents with suicide attempts. Fırat Medical Journal, 17(4), 228–232.

12.

Evans

Hawton

Rodham

(2004). Factors associated with suicidal phenomena in adolescents: A systematic review of population-based studies. Clinical Psychology Review, 24, 957–979.

13.

Evren

Bozkurt

Can

(2014). Non-suicidal self-harm behavior within the previous year among 10th-grade adolescents in Istanbul and related variables. Nordic Journal of Psychiatry, 68, 481–487.

14.

Fidan

Ceyhun

H. E.

Kirpinar

(2009). Family functionality and suicide attempts in children and youths. Klinik Psikofarmakoloji Bulteni, 19, S220–S222.

15.

Fidan

Kirpinar

(2009). Coping strategies and suicide attempts in youths. Klinik Psikofarmakoloji Bulteni, 19(Suppl. 1), S214–S216.

16.

Gökçen

Köylü

(2011). Evaluation of cases under 18 years admitted for attempted suicide to the emergency service and transferred to the psychosocial support unit. Journal of Academic Emergency Medicine, 10, 18–21.

17.

Grover

Sarkar

Chakrabarti

Malhotra

Avasthi

(2015). Intentional self-harm in children and adolescents: A study from psychiatry consultation liaison services of a tertiary care hospital. Indian Journal of Psychological Medicine, 37, 12–16.

18.

Hamza

C. A.

Stewart

S. L.

Willoughby

(2012). Examining the link between nonsuicidal self-injury and suicidal behavior: A review of the literature and an integrated model. Clinical Psychology Review, 32, 482–495.

19.

Harter

Stocker

Robinson

N. S.

(1996). The perceived directionality of the link between approval and self-worth: The liabilities of a looking glass self-orientation among young adolescents. Journal of Research on Adolescence, 6, 285–308.

20.

Hawton

Bergen

Cooper

Turnbull

Waters

Ness

Kapur

(2015). Suicide following self-harm: Findings from the Multicentre Study of self-harm in England, 2000–2012. Journal of Affective Disorders, 175, 147–151.

21.

Hawton

Hall

Simkin

Bale

Bond

Codd

Stewart

(2003). Deliberate self-harm in adolescents: A study of characteristics and trends in Oxford, 1990–2000. Journal of Child Psychology and Psychiatry, 44, 1191–1198.

22.

Hawton

Rodham

Evans

Weatherall

(2002). Deliberate self harm in adolescents: Self report survey in schools in England. British Medical Journal, 325, 1207–1211.

23.

Hawton

Saunders

K. E.

O’Connor

R. C.

(2012). Self-harm and suicide in adolescents. The Lancet, 379, 2373–2382.

24.

Kapur

Gunnell

Hawton

Nadeem

Khalil

Longson

… Cooper

(2013). Messages from Manchester: Pilot randomised controlled trial following self-harm. British Journal of Psychiatry, 203, 73–74.

25.

Kharsati

Bhola

(2015). Patterns of non-suicidal self-injurious behaviours among college students in India. International Journal of Social Psychiatry, 61, 39–49.

26.

Kola

Petrela

Zaimi

(2013). Trends of attempted suicide in Albanian children and adolescents. Archives of Psychiatry and Psychotherapy, 4, 39–44.

27.

Konradsen

Van Der Hoek

Peiris

(2006). Reaching for the bottle of pesticide – A cry for help. Self-inflicted poisonings in Sri Lanka. Social Science & Medicine, 62, 1710–1719.

28.

Lai

N. M. S.

(2011). Transition to financial independence of youth in Asian countries: Does furthering one’s education pay? (Expert Paper No. 2011/15). New York, NY: United Nations Department of Economic and Social Affairs.

29.

Law

B. M.

Shek

D. T.

(2013). Self-harm and suicide attempts among young Chinese adolescents in Hong Kong: Prevalence, correlates, and changes. Journal of Pediatric & Adolescent Gynecology, 26, S26–S32.

30.

Liang

Yan

Zhang

Zhu

Situ

… Huang

(2014). Differences between non-suicidal self injury and suicide attempt in Chinese adolescents. Asian journal of psychiatry, 8, 76–83.

31.

Liu

Sun

Yang

(2008). Parent-reported suicidal behavior and correlates among adolescents in China. Journal of Affective Disorders, 105, 73–80.

32.

McKinnon

Gariepy

Sentenac

Elgar

F. J.

(2016). Adolescent suicidal behaviours in 32 low- and middle-income countries. Bulletin of the World Health Organization, 94, 340–350f.

33.

Modesti

P. A.

Reboldi

Cappuccio

F. P.

Agyemang

Remuzzi

Rapi

Parati

(2016). Panethnic differences in blood pressure in Europe: A systematic review and meta-analysis. PLoS ONE, 11, e0147601.

34.

Moran

Coffey

Romaniuk

Olsson

Borschmann

Carlin

J. B.

Patton

G. C.

(2012). The natural history of self-harm from adolescence to young adulthood: A population-based cohort study. The Lancet, 379, 236–243.

35.

Muehlenkamp

J. J.

Claes

Havertape

Plener

P. L.

(2012). International prevalence of adolescent non-suicidal self-injury and deliberate self-harm. Child and Adolescent Psychiatry and Mental Health, 6, 10.

36.

Muula

A. S.

Kazembe

Rudatsikira

Siziya

(2008). Suicidal ideation and associated factors among in-school adolescents in Zambia. Tanzania Journal of Health Research, 9, 202–206.

37.

National Collaborating Centre for Mental Health. (2012). National Institute for Health and Clinical Excellence: Guidance. Self-harm: longer-term management. Leicester, UK: The British Psychological Society & The Royal College of Psychiatrists.

38.

Oksuz

Malhan

(2005). Socioeconomic factors and health risk behaviors among university students in Turkey: Questionnaire study. Croatian Medical Journal, 46, 66–73.

39.

Owens

Horrocks

House

(2002). Fatal and non-fatal repetition of self-harm. British Journal of Psychiatry, 181, 193–199.

40.

Patel

Ramasundarahettige

Vijayakumar

Thakur

Gajalakshmi

Gururaj

… Million Death Study Collaborators. (2012). Suicide mortality in India: A nationally representative survey. The Lancet, 379, 2343–2351.

41.

Patton

G. C.

Hemphill

S. A.

Beyers

J. M.

Bond

Toumbourou

J. W.

Mcmorris

B. J.

Catalano

R. F.

(2007). Pubertal stage and deliberate self-harm in adolescents. Journal of the American Academy of Child & Adolescent Psychiatry, 46, 508–514.

42.

Patton

G. C.

Sawyer

S. M.

Santelli

J. S.

Ross

D. A.

Afifi

Allen

N. B.

… Bonell

(2016). Our future: A Lancet commission on adolescent health and wellbeing. The Lancet, 387, 2423–2478.

43.

Perkins

J. M.

Subramanian

S. V.

Christakis

N. A.

(2015). Social networks and health: A systematic review of sociocentric network studies in low- and middle-income countries. Social Science & Medicine, 125, 60–78.

44.

Pillai

Andrews

Patel

(2009). Violence, psychological distress and the risk of suicidal behaviour in young people in India. International Journal of Epidemiology, 38, 459–469.

45.

Prinstein

M. J.

Heilbron

Guerry

J. D.

Franklin

J. C.

Rancourt

Simon

Spirito

. (2010). Peer influence and nonsuicidal self injury: Longitudinal results in community and clinically-referred adolescent samples. Journal of Abnormal Child Psychology, 38, 669–682.

46.

Shek

D. T.

(2012). Self-harm and suicidal behaviors in Hong Kong adolescents: Prevalence and psychosocial correlates. Scientific World Journal, 2012, 932540.

47.

Sidhartha

Jena

(2006). Suicidal behaviors in adolescents. Indian Journal of Pediatrics, 73, 783–788.

48.

Skegg

(2005). Self-harm. The Lancet, 366, 1471–1483.

49.

Somer

Bildik

Kabukcu-Basay

Gungor

Basay

Farmer

R. F.

(2015). Prevalence of non-suicidal self-injury and distinct groups of self-injurers in a community sample of adolescents. Social Psychiatry and Psychiatric Epidemiology, 50, 1163–1171.

50.

Soylu

Taneli

(2013). Investigation of social, emotional, and cognitive factors with effect on suicidal behaviour in adolescents with depression. Archives of Neuropsychiatry, 50, 352–359.

51.

Swahn

M. H.

Ali

Bossarte

R. M.

Van Dulmen

Crosby

Jones

A. C.

Schinka

K. C.

(2012). Self-harm and suicide attempts among high-risk, urban youth in the U.S.: Shared and unique risk and protective factors. International Journal of Environmental Research and Public Health, 9, 178–191.

52.

Syed

E. U.

Khan

M. M.

(2008). Pattern of deliberate self-harm in young people in Karachi, Pakistan. Crisis, 29, 159–163.

53.

Tang

Zhu

Liu

(2011). Association between non-suicidal self-injuries and suicide attempts in Chinese adolescents and college students: A cross-section study. PLoS ONE, 6, e17977.

54.

Toprak

Cetin

Guven

Can

Demircan

(2011). Self-harm, suicidal ideation and suicide attempts among college students. Psychiatry Research, 187, 140–144.

55.

Vijayakumar

Jeyaseelan

Kumar

Mohanraj

Devika

Manikandan

(2013). A central storage facility to reduce pesticide suicides-a feasibility study from India. BMC Public Health, 13, 850.

56.

Wan

Y. H.

C. L.

Hao

J. H.

Sun

Tao

F. B.

(2011). Deliberate self-harm behaviors in Chinese adolescents and young adults. European Child & Adolescent Psychiatry, 20, 517–525.

57.

Washburn

J. J.

Richardt

S. L.

Styer

D. M.

Gebhardt

Juzwin

K. R.

Yourek

Aldridge

(2012). Psychotherapeutic approaches to non-suicidal self-injury in adolescents. Child and Adolescent Psychiatry and Mental Health, 6, 14.

58.

Wells

Shea

O’Connell

Peterson

Welch

Losos

Tugwell

(2000). The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses. www.ohri.ca/programs/clinical_epidemiology/oxford.asp

59.

Whitlock

Muehlenkamp

Eckenrode

Purington

Baral Abrams

Barreira

Kress

(2013). Nonsuicidal self-injury as a gateway to suicide in young adults. Journal of Adolescent Health, 52, 486–492.

60.

Wilson

M. L.

Dunlavy

A. C.

Viswanathan

Bovet

(2012). Suicidal expression among school-attending adolescents in a middle-income sub-Saharan country. International Journal of Environmental Research and Public Health, 9, 4122–4134.

61.

Wong

W. C.

Cheung

C. S.

Hart

G. J.

(2008). Development of a quality assessment tool for systematic reviews of observational studies (QATSO) of HIV prevalence in men having sex with men and associated risk behaviours. Emerging Themes in Epidemiology, 5, 23.

62.

You

Lin

M. P.

Leung

(2013). The best friend and friendship group influence on adolescent nonsuicidal self-injury. Journal of Abnormal Child Psychology, 41, 993–1004.

63.

Zahl

D. L.

Hawton

(2004). Repetition of deliberate self-harm and subsequent suicide risk: Long-term follow-up study of 11,583 patients. British Journal of Psychiatry, 185, 70–75.

64.

Zemaitiene

Zaborskis

(2005). Suicidal tendencies and attitude towards freedom to choose suicide among Lithuanian schoolchildren: Results from three cross-sectional studies in 1994, 1998, and 2002. BMC Public Health, 5(1), 83.

65.

Zoroglu

S. S.

Tuzun

Sar

Tutkun

Savacs

H. A.

Ozturk

Kora

M. E.

(2003). Suicide attempt and self-mutilation among Turkish high school students in relation with abuse, neglect and dissociation. Psychiatry and Clinical Neurosciences, 57, 119–126.