Social influences on NSSI and suicidality

There is growing recognition of the importance of social factors in NSSI and suicidality-related outcomes. The clustering of incidences of NSSI and a number of deaths by suicide in specific locations and amongst specific social groups has led to the investigation of possible social influences on NSSI and suicidal behaviours [1, 20, 21], particularly peer influence amongst younger adolescents and young adults [22, 23]. Theories and concepts detailing the possible social origins of NSSI and suicidality include contagion [20], the Werther effect (relating to increases in suicide rates via media coverage) [24, 25], and suicide diffusion [26]. Although social factors and pressures may be implicated in NSSI and suicidality, purely focusing on group-level social factors may ignore important individual-level factors which exacerbate or protect against such external pressures.

Common to many psychological theories and approaches to understanding suicidality and NSSI is the influence of an individual’s perception of their social environment, such as their connectedness to others, their community, and social group [12, 27], and their status or rank within social groups [28]. For example, there is evidence of an increased risk of NSSI and suicidality amongst individuals who perceive that they have limited social support [29–33], who perceive themselves to be of low social rank [34], and who have been exposed to NSSI or a death by suicide in their peer group or family [21, 35]. Whilst theoretical models and empirical studies have highlighted the potential key role of individual perceptions of their social environment and relationships to others in suicidality, how perceptions of social norms are implicated in the pathways to NSSI and suicidality remains poorly understood.

What are social norms?

Social norms are a key influence on personal behaviours and are broadly defined as “rules and standards that are understood by members of a group, and that guide and/or constrain social behaviour without the force of law” (p. 152) [36]. Social norms are proposed to influence intention formation for a range of behaviours and are featured in a number of prominent health-related behaviour models, most notably the Theory of Planned Behavior [4, 37]. Social norms are implicated in various health behaviours, including but not limited to: alcohol and other substance use [38, 39]; dietary behaviours [40, 41]; infection control behaviours [42, 43]; the use of contraceptives [44]; and use of sun protection [45]. As both NSSI and suicidality can be understood as similar health behaviours influenced by the formation of intentions, perceived social norms may have a similar influence on NSSI and suicidality-related outcomes.

Social norms have been defined, operationalised, and measured in multiple ways. Common operationalisations of social norms include the perceived rates of a behaviour (descriptive norms), perceived group attitudes or approval (injunctive norms) [46], subjective norms of an important other (the perceived social pressure to engage or not engage in a behaviour) [37], collective norms (“prevailing codes of conduct that either prescribe or proscribe behaviours that members of a group can enact”, p.129 [5]), moral norms (characterised by feelings of shame when violated [47]), to self-other discrepancies in terms of over or underestimating social behaviours and attitudes compared to actual norms [3, 4, 39, 48]. Social norms may also inform explicit legal frameworks and legislation based on agreed standards of permissible and unacceptable behaviour to more implicit and unwritten standards, such as widely shared beliefs about how individuals are expected to behave [49]. Social norms may operate at a societal or group level, based on social interaction and communication of expected standards amongst group members, or at an individual psychological level in the form of interpretations or perceptions of broader collective norms [5, 50]. There is, however, no universally agreed typology of social norms, and there are disagreements in the theoretical literature about the distinctiveness between different social norms types [47].

Perceived social norms may also serve different social influence functions. For example, perceived norms may indicate what to do in a given environment or context, especially novel situations, and thus encouraging conformity to match how others behave (“informational social influence”), to indicating what an individual ought or should do in terms of accepted standards or behaviours (“normative group pressure”) [51]. In relation to NSSI and suicidality, these different influences may be more nuanced compared to other behaviours. NSSI/suicidality are typically not public or majority behaviours, which may not be typically viewed as socially acceptable, and so they may be viewed differently depending on the social context and the individual’s group affiliations.

Social norms, NSSI and suicidality

There is an acknowledged lack of understanding on the relationship between social norms with NSSI and suicidality, with calls for further research and theory development on this possible relationship [22]. There have been specific calls for more work understanding the norms-suicidality/NSSI association amongst key high-risk groups who may be particularly influenced by perceived peer behaviours and attitudes (e.g., young people who identify with goth subcultures [52]). The differing conceptualisations of social norms presents a particular challenge for understanding their role in NSSI and suicidality, and health-related behaviours more broadly [3, 4, 50]. These different forms of social norms may potentially influence the development of NSSI and suicidal thoughts, feelings, and behaviours, in different ways. Social norms may act as a subtle social influence pressure to conform to a perceived set of standards [37] which may have the potential for penalties or social sanctions for non-conformity [49]. Deviation from actual collective norms, or a perception that one is deviating from social norms, could be important in precipitating and/or exacerbating suicidal thoughts and feelings, such as feelings of shame, humiliation, defeat and entrapment which have already been implicated in the psychological pathways to suicide [28, 53, 54]. Perceived social norms may have unique roles in terms of exacerbating NSSI and suicidality, as well as potentially being protective against poor outcomes depending on the type of social norms.

Although recent theoretical developments have focused on outlining the processes which may take an individual from thinking about NSSI and suicide towards planning and/or engaging in related behaviours [28, 55, 56], few theories explicitly describe the role of social norms in this process. Considering that social norms have been implicated in the formation of intentions to engage in a variety of protective and risky health behaviours, it is not unfeasible that social norms have an influence on NSSI and suicidality, but where, when, and how social norms are implicated is unclear. To date, only one psychological theory of suicide (O’Connor’s Integrated Motivational Volitional, IMV, model [28, 57]) has outlined the possible role of social norms as one of a number of motivational moderators of the relationship between entrapment and suicidal ideation [28]. This proposed role of social norms has yet to be tested in the context of the IMV and there remains some lack of clarity of how social norms function as a moderator (as a buffering or exacerbating factor) or which types of norms are implicated in suicidal ideation or other outcomes. Whilst other psychologically-focused theoretical models of suicide and NSSI exist [55], these typically do not discuss the potential role of social norms in NSSI/suicidality outcomes.

This lack of clarity about the role of perceived social norms in NSSI and suicidality is problematic for the research literature and for the development of effective intervention approaches [22]. Social norms are a key component of many health behaviour change interventions and are typically featured in normative feedback comparing individual or group perceptions versus actual reported norms as a means to reduce the perceived social pressure to engage or not engage in range of behaviours (e.g., Social Norms Approach interventions reducing substance use) [3]. Due to the current lack of clarity about the types of social norms associated with NSSI and suicide-related outcomes, it is unclear whether normative feedback could be effectively used in harm prevention or harm reduction approaches or how such normative feedback should be phrased (i.e., which norms or reference groups to feature in feedback). Care is especially needed as normative feedback used with other health behaviours may not easily translate given the complexity of NSSI and suicide-related behaviours. For example, Social Norms Approach interventions typically highlight the actual, lower, rates of negative behaviours and peer approval versus the commonly held (mis)perceptions in order to promote behaviour change [3, 58, 59]. Given their relative infrequency, such feedback for NSSI/suicide-related behaviours might imply that such behaviours are more commonplace than the reality, thereby maintaining or increasing NSSI/suicidality. Alternatively, such normative feedback could further promote a sense of shame or difference amongst those who are suicidal or who have engaged in NSSI who deviate from the norm, leading to worse outcomes. Understanding the types of perceived social norms associated with NSSI and suicidality, and the directions of such relationships, would be important for understanding whether normative feedback could and/or should be used for NSSI/suicidality, and in developing appropriately targeted interventions.

The present review

To date, there have been no prior reviews of the literature on the role of social norms in NSSI and suicidality-related outcomes, and no prior scoping of the types of perceived social norms studied in relation to NSSI/suicidality and the role of perceived social norms in these outcomes. Whilst NSSI and suicide have their differences as phenomena, both can be viewed as forms of health-behaviour influenced by the formation of intentions [55, 60], and such intentions may in turn be influenced by perceived social norms as a guide for how to behave and what others view as acceptable behaviours. There remains a lack of clarity in the research literature on how conformity or non-conformity to some form of perceived social norms is associated with NSSI and/or suicidality outcomes. The different, multiple, and often contrasting ways in which social norms have been defined presents a particular challenge for understanding their influence in health-related behaviours and intention formation more generally [3, 4], and in the development of appropriate behaviour change interventions.

The present review is the first attempt to scope the extent of the existing literature and identify the types of perceived social norms associated with NSSI and/or suicidality (i.e., suicidal thoughts, feelings, and behaviours). A scoping review approach [61] was adopted given the lack of prior reviews of social norms and NSSI/suicidality. A scoping review was appropriate considering existing acknowledgements of the heterogeneity of the social norms literature [3], specifically the inconsistent and often contradictory ways in which social norms have been operationalised in the broader literature (e.g., see Shulman and colleagues’ content analysis of studies employing social norms [4]). The present review is the first attempt to scope the types of social norms implicated in NSSI and suicidality outcomes and determine the extent of the existing literature and its methodological approaches, upon which future systematic literature approaches and syntheses may be conducted.

Design

A systematic scoping review [61] of the published research literature was conducted to identify the types of perceived social norms studied in relation to NSSI and suicidality, and the role of perceived social norms in these outcomes (see S1 File for the PRISMA Scoping Review Checklist). A scoping review was deemed appropriate based on a pilot review exercise which indicated the existence of a heterogenous literature on NSSI, suicidality, and social norms. In contrast to full systematic reviews, which tend to focus on specific research questions and specific study designs based on a narrow range of quality-assessed studies [61], scoping reviews are particularly useful where the extent of a literature is unknown or unexplored and can assist in mapping a literature and providing a broader overview of an evidence base regardless of its quality [62]. In line with recommendations for scoping reviews [61, 63], the present review represents an initial exploration of the literature, accommodates an identification of the types of perceived social norms implicated in NSSI and suicidality, as well as an understanding of the extent of the literature on this topic. The review protocol was registered on the Open Science Framework (see https://osf.io/btpzc/?view_only=11a12b31c7854225a022139b4228eed1).

Eligibility criteria

Published studies were selected for inclusion based on the following criteria: (1) they investigated the role of perceived social norms in the individual experience of suicide-related thoughts, feelings and/or behaviours including self-harm or non-suicidal self-injury (e.g. as an outcome variable in quantitative studies or part of a research question or an identified theme for qualitative studies); (2) they constituted original empirical research; (3) were published in the English language; and (4) had been peer-reviewed. Original qualitative and quantitative studies were eligible for inclusion, whilst review articles or commentaries were ineligible. No restrictions in the ages or the nature of the study samples were applied to ensure the maximum inclusion of studies (e.g., adolescent, general population, and clinical samples). Studies on suicide attacks (e.g., suicide bombings, suicide terrorism), murder-suicides, or assisted suicide (i.e., euthanasia) were ineligible.

Literature search strategy

Initial searches were conducted on the following databases in November 2020: Scopus, Web of Science, CINAHL, PubMed, PsychInfo, and ASSIA. Search terms related to ‘social norms’ and ‘suicidality’ or ‘NSSI’ focusing on the publication titles, abstract, keywords, and main text were used, for example: SOCIAL NORMS, NORMATIVE PERCEPTIONS, SUBJECTIVE NORMS, PERCEIVED NORMS, SUICID*, PARASUICIDE, and SELF-HARM. Broad search terms were used to capture a range of suicidality-related experiences (e.g., thoughts, behaviours, attempts) and the different operationalisations of “social norms” (e.g., norms, normative influence, normative misperceptions). For example, the following search string was used on the Web Of Science database with the results limited to English-language sources: TS = (suicid* AND “social norm*”). Searches were limited to English language articles where possible, but no other limits were applied to ensure the maximum inclusivity of returned articles for screening (i.e., no limit on date of publication or other filters were applied).

Studies which focused on reporting “normative scores” for psychometric assessments, or which analysed national trends data, were excluded as these did not specifically focus on individual perceptions of social norms and their influence on suicidality-related outcomes. Initial literature searches were restricted to articles published until the end of October 2020, with a top-up search conducted in June 2022 for articles published between November 2020 and May 2022. The lead author conducted the searches and screenings, with an independent check of the screenings and extracted data conducted by the second author, with disagreements between the authors and ambiguities in the reviewed studies discussed between the two authors. Hand searches of key suicidology journals which have published psychologically focused research (e.g., Crisis, Archives of Suicide Research, Suicide and Life-Threatening Behavior, Journal of Affective Disorders), together with searches of the publication records of prominent NSSI and suicidology researchers (e.g., Thomas Joiner, David Klonsky, Matthew Nock, Rory O’Connor, Ellen Townsend), were conducted to identify additional studies of relevance to the review.

Data extraction and charting

Data extracted from eligible articles included: author name(s), date of publication and a full academic reference, sample characteristics (type/population, setting, sample sizes, screenings), methodology, analysis strategy, suicidality/NSSI measures, testing of suicide and/or social norms theory, primary and secondary findings (including indirect quantitative effects), evaluation and limitations of the study, and the conclusion(s) of the study.

Summary of identified literature

Initial literature searches identified a sample of 31 eligible studies (4 qualitative, 26 quantitative, 1 mixed methods; see Fig 1), with an additional five quantitative studies identified through the top-up searches, making a final sample of 36 eligible studies (4 qualitative, 31 quantitative, 1 mixed methods).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 1. PRISMA scoping review flowchart detailing the screening process.

* Note that studies could be excluded for not meeting both the suicidality/self-harm outcome and the social norms predictor/focus screening criteria.

https://doi.org/10.1371/journal.pone.0286118.g001

Table 1 provides a summary of the study designs, social norms investigated or explored, and the suicidality and/or NSSI outcomes featured in the reviewed studies (S2 File includes additional details on study designs and methods). In terms of the characteristics of the reviewed studies, most (n = 28) were conducted in non-clinical settings (e.g., the community, schools, universities, and/or online participation panels), with six studies conducted in hospitals [64–69], one in a prison [70], and one study conducted at a US armed forces training base [71]. There was substantial variability in the sample demographics, particularly across participant ages, with studies sampling adolescents and high school students to older adults (see Table 1). Three studies were conducted in non-Western countries [72–74], with the majority of studies based in the UK, USA, Canada, Australia and New Zealand. Most studies used cross-sectional survey designs (n = 27), with five prospective self-report survey studies, and four studies featuring interview or focus groups (note that E. McDermott et al., 2018, featured both a cross-sectional survey and interviews [69]).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Summary of the reviewed studies’ design, social norms investigated/measured, and the relationships between social norms and NSSI/suicidality.

https://doi.org/10.1371/journal.pone.0286118.t001

Seven out of the thirty-one quantitative studies explicitly tested a theoretical model of suicide in their aims and hypotheses. Six of the reviewed studies [67, 71, 75–78] tested hypotheses that were based on the Interpersonal Theory of Suicide (IPTS) [12, 27]. One study tested more general assumptions of Nock and Prinstein’s Four-Factor Model of self-harm in the context of conformity to masculine social norms [79, 80]. None of the IPTS studies fully tested the three main components of the model at the same time (i.e., thwarted belonginess, perceived burdensomeness, acquired capability); for example, Daruwala et al. (2021) measured all three constructs but only the ‘fearlessness about death’ aspect of acquired capability, not pain tolerance [71]. Four studies only measured the acquired capability for suicide in terms of the ‘fearlessness about death’ construct [67, 75, 77], with one study measuring acquired capability as a suicide-related outcome [78], and a further study measured the thwarted belongingness and perceived burdensomeness constructs but not acquired capability [76]. Ten studies tested theories featuring social norms, including: the Theory of Planned Behaviour [64, 65, 70, 73, 74]; the Theory of Reasoned Action [81]; the Theory of Normative Social Behavior [82]; the Social Norms Approach [83, 84]; with a further study testing predictions of overestimations of peer depressed mood norms in the context of Beck’s Cognitive Theory of Depression [85]. No studies tested both a theory of suicide and a theory of social norms at the same time.

Types of social norms studied in relation to NSSI/suicidality

There was substantial variability in the types of perceived social norms investigated or explored across studies, and in their measurement. All of the quantitative studies (n = 31, plus E. McDermott et al.’s 2018 mixed methods study [69]) used some form of self-report questionnaire measure to assess perceived social norms. For the qualitative studies, perceived social norms were discussed as themes in their own right or as part of broader themes, but none of the reviewed qualitative studies included a standardised assessment of perceived norms.

Not all of the reviewed studies focused on measuring specific suicide-related norms in relation to suicide-related outcomes. Nine studies reported assessing explicit suicide-related norms, such as the perceived prevalence of suicidal behaviour or others’ attitudes towards self-harm behaviours or suicidality [64, 65, 73, 74, 82–85, 95]. Sixteen studies focused on the conformity to some form of masculine social norm (e.g., masculine gender roles, masculine sense of honour, traditional or hegemonic norms) [67, 68, 71, 75–79, 90–94, 97–99], with one study focusing on female gender roles in relation to cultural norms associated with childbirth during marriage [72], and one study focusing on the non-conformity to broader traditional gender and sexuality norms amongst LGBT youth [69]. Two studies measured norms relating to the experience of depression and anxiety-related symptoms [82, 85]. Five studies focused on help-seeking norms, including perceptions of the acceptability of seeking help for suicidality and depressed mood [66, 70, 81, 89, 91]. Two studies reported the role of perceived parental norms and beliefs on suicidality amongst adolescents, including parents’ prosocial behaviours [86] and Asian American parents’ beliefs conforming to traditional sociocultural norms about gender [96]. Finally, two studies measured the influence of substance-use related norms in relation to suicidality, including alcohol and cannabis use [88, 100]. The next section provides a summary of these studies’ findings by the type of social norms investigated, grouped by male and masculine norms, female and feminine norms, gender and sexuality norms, descriptive and injunctive norms, normative misperceptions, and other forms of perceived norms.

Summary of the relationship between social norms and NSSI/suicidality outcomes

Twenty-seven quantitative studies reported significant associations between perceived norms and self-harm/suicidality [64, 67, 70, 71, 73–79, 81–89, 92–95, 97, 99, 100]. Three studies reported significant relationships only in correlation analyses or univariate models but not in multivariate models when other predictors and clinical variables were included [73, 89, 94]. Four studies reported no significant norms and self-harm/suicidality relationship [65, 66, 90, 96].

Twenty-two quantitative reported that perceived social norms were associated with worsened suicidality and NSSI [64, 67, 71, 73–79, 81–85, 87, 88, 92–95, 97, 100]. Negative outcomes included a history of self-harm/NSSI [84], an increased risk of self-harm [79], increased suicidal ideation [76, 85], suicide intentions [74] and suicide attempts [75], and lower likelihoods of seeking support for self-harm/suicidality [93]. Specific norms implicated in suicidality and self-harm included a greater conformity to self-reliance norms associated with masculinity [71, 92, 99]. Closer, more proximal reference groups for social norms also tended to have stronger relationships with self-harm and suicidality compared to more distal groups [83]. There were some potential sex-differences amongst younger age groups, with boys appearing to be more influenced by family and parental norms than girls (e.g. [86, 88]), and girls appearing more influenced by peers and others their age [87, 88].

Eight quantitative studies reported a protective effect of conformity to perceived social norms on suicidality and self-harm outcomes, including: a lower risk of suicide [88]; reduced suicide intentions [73] (correlation analyses only); a less severe history of suicide attempts [86]; and increased intentions to seek help or support others [70, 82, 89]. Two studies reported protective effects of conformity to specific forms of masculine norms on suicidal ideation amongst male participants, including norms relating to heterosexuality [99], and risk-taking and emotional control [92].

A number of quantitative studies reported indirect relationships between social norms and self-harm/suicidality, in the form of mediated and/or moderated effects [64, 67, 74, 77, 78, 82, 88]. Although, two studies reported non-significant mediated/moderated effects involving social norms in suicidality and self-harm [73, 81]. Example indirect effects reported in the reviewed studies include Swahn and colleagues (2010) who reported a moderation of the relationship between teenagers’ personal alcohol use and history of attempted suicide by perceived peer and adult approval norms, with lower suicide risk amongst teenagers who did not use alcohol and thought family/peers disapproved of alcohol use, and higher risk amongst teenagers who used alcohol but thought friends disapproved of alcohol use [88]. Other examples of indirect effects involving social norms include perceived pro-suicide injunctive norms mediating the relationship between communicating about suicide with friends, family and co-workers with prospective suicide intentions [74], suggesting that increased exposure to discussing suicide with close social contacts leads to stronger perceptions that suicide is a socially acceptable act thereby increasing intentions. Elsewhere, the relationship between conformity to perceived masculine honour norms and increased suicide capability was mediated by the experience of painful/provocative life events [78]. Investigating indirect relationships between perceived social norms with NSSI and suicidality was not common across the quantitative studies.

In terms of the qualitative studies, a sense of violating perceived norms in terms of gender and sexuality was discussed in relation to shame and an increased risk of suicidal thoughts and behaviours amongst Sri Lankan woman who gave birth/were pregnant outside of marriage [72] and LGBT youth in England [69]. For men especially, traditional masculine norms relating to independence, self-control, and stoicism, appear to be particularly restrictive and increased suicidality [68, 91, 98]. Age also appeared to be important for how men viewed suicide as an act, with older men viewing suicide as a means of maintaining one’s independence and sense of masculinity later in life whilst also viewing suicide as a sign of weakness amongst younger men [98]. Other work suggested that challenging traditional perceptions of masculinity in younger men can lead to increased help-seeking and increased feelings of masculinity [68]. Across the qualitative studies, not conforming to perceived ideals and social norms was typically associated with suicidality through a sense of shame or failure for not meeting some perceived set standard (e.g., in terms of expectations of gender, age, and/or sexuality).

The operationalisation of social norms

How perceived social norms are measured and defined is a challenge for the literature [108], not just in the context of suicidality but in relation to health-behaviours more broadly [3]. As highlighted in this review, there were a variety of measures and approaches to defining perceived social norms which presents a challenge for understanding the role of perceived norms in suicidality-related outcomes and in conducting meta-analyses. Many studies fail to provide adequate details of the wording of their social norms measures, such as the behaviour or attitude and/or the referent group which the norm items measure. One study in the review used a composite social norm score across different normative referent groups, which may not account for the different effects on suicidality associated with different referents (e.g. close family members versus religious leaders [73]). Other studies in the wider literature use measures which appear to combine perceived social norms with other constructs, including attitudes and intentions [104, 105].

There is still some lack of clarity about whether NSSI and suicidality is more strongly associated with more specific perceived normative behaviours or attitudes, or which reference groups are more influential (i.e., more proximal versus more distal groups). Some, but not all, of the reviewed studies reported weak-to-moderate effects of perceived social norms on suicidality and self-harm outcomes [67, 71, 78, 81, 94]. These effect sizes may reflect that perceived social norms are only a modest predictor of outcomes in themselves (although this might vary depending on the sample and norms studied), but also the complex and multifaceted causal factors implicated in suicidality and NSSI, as well as the general difficulty in predicting these outcomes [55, 109, 110].

It was notable that few studies in the review measured or discussed the role of social identification, only one study reporting a significant interaction between stronger group identification and greater social norms of self-harm in the intention to engage in deliberate self-harm amongst a clinical sample [64]. Social identification is an important moderator of the effect of perceived norms on health behaviours and may strengthen the influence of the perceived behaviours or attitudes of a social group on the individual [111, 112], particularly for behaviourally relevant social referent groups [113]. Understanding the specific social identification processes implicated in the social norms and self-harm/suicidality relationship may be important for suicide prevention efforts, particularly for informing personalised normative feedback intervention approaches featuring social norms messages (e.g., those based on the Social Norms Approach) [3].

Conformity to perceived traditional hegemonic masculinity norms in relation to suicidality was a clear focus of the reviewed studies and it was often discussed how men appear to be at a greater risk of suicide because of restrictive male gender norms. There are various issues with psychometric measures of masculine norm conformity, in particular the widely used CMNI scale [114] which has been criticised for being less of a measure of masculine norms (i.e. what others do) and more a measure of an individual’s personal behaviours, attitudes and feelings about masculinity [115]. There is also a danger of the literature assuming that gender norms have unique influences on the sexes (i.e. perceived masculine norms being only relevant to those who identify as men), in assuming the binary nature of gender (i.e. male vs female), as well as ignoring the heterogeneity of masculine identities [99]. There was some evidence in the literature reviewed here that conformity to masculine norms and the capability for suicide or intentions to seek help is not consistently moderated by biological sex, and conformity to perceived masculine social norms can influence women’s experiences of suicidality as much as men [67, 77, 93]. It might be that men are more likely to be socialised to conform to these restrictive masculine norms, but these perceived norms and standards may be influential on individuals regardless of their sex and/or gender identity [93]. There is an ongoing discussion of the complex gender politics in the suicidology field, particularly the implicit assumptions of suicide as a male behaviour and reliance on simplistic dualistic approaches to gender [116]. Greater diversity in the sampling of participants by sex and gender in studies of masculine norms and suicidality is needed, as is the move away from binary assumptions of gender in theoretical models of suicide.

Additional research implications

Studies also varied by their use of validated psychometric assessments of ideation and behaviour, with a number of studies appearing to use single-item measures which may not be sufficiently sensitive to distinguish between different aspects of NSSI/suicidal thoughts or behaviours [66, 75, 83, 85–89, 96, 97, 99]. Other studies (e.g. [95]) included detailed assessments of passive and active suicidal ideation, thoughts about death and dying, and thoughts more specifically related to suicidal intent respectively. Whilst the present review included studies focusing on both self-harm and suicide, these two behaviours are not necessarily synonymous as many individuals may engage in self-harming behaviours without thinking about suicide. Although, the engagement in NSSI and other forms of self-injury is a potential risk factor for suicide by increasing one’s capability for suicide [117], considering evidence of heightened rates of attempted suicide amongst individuals with histories of NSSI [69, 84, 109].

There is also a need to understand the role of other social influence factors that may mediate or moderate any potential association between perceived social norms and self-harm/suicidality. As highlighted earlier, few studies reported conducting mediation or moderation analyses, and due to the variance in the types of social norms measured across studies, different forms of perceived norms appeared to have different roles in the pathways to self-harm and suicidality. More fine-tuned analyses focusing on indirect relationships would improve the understanding of the ‘how’ and ‘when’ perceived social norms are implicated in outcomes relating to self-harm and suicidality. A key factor to consider may be the exposure to self-harm or suicide in the individual’s social environment and their relationships with friends, family, and/or peers [118–120], which may influence the capability for suicide [12, 27], and inflate the perceived social norms of others’ self-harm/suicide-related behaviours and attitudes [121]. Exposure to suicide amongst social groups was not commonly measured or featured in the reviewed studies, although Reyes-Portillo and colleagues (2019) reported that perceived social norms remained a predictor of suicidal ideation whilst controlling for past exposure to suicide [95]. There may be potential interactive effects between exposure to self-harm and/or suicide in one’s social group and normative perceptions of these behaviours in relation to self-harm/suicidality outcomes. Understanding the more complex relationships between perceived social norms and self-harm/suicidality outcomes is important for developing effective intervention strategies, particularly in understanding the types of norms and mediating or moderating factors that may worsen or protect against poorer outcomes.

There is a final consideration for the broader social norms and NSSI/suicidality research literature in terms of the speculative nature of social norms’ involvement in these outcomes. As previously outlined, our literature searches identified a large body of research literature for screening, however the majority of this (then excluded) literature did not actually study ‘social norms’ as a causal factor or predictor of NSSI or suicidality. A common issue in this literature was the tendency for ‘social norms’ to offered as a general speculative causal factor for authors’ findings despite the absence of measures of social norms in the respective studies. An observation from our scoping of the existent literature is the presence of some likely common-sense assumptions of the importance and role of social norms in NSSI and suicidality, and a tendency by authors to speculate that some (unmeasured) ‘social norms’ may explain certain, potentially unexplainable, patterns of results. Greater caution is required in the literature when speculating on the role of ‘social norms’, however defined, in the experience of NSSI and suicidality.

Implications for practice

Whilst caution should be exercised in developing interventions based on the heterogenous literature identified here, challenging perceived social norms has been a focus of a number of health interventions for other behaviours [3] and there is the potential for normative feedback to be used in suicide prevention efforts. Considering that socio-cognitive variables, such as perceived norms, may influence suicide-related outcomes over and above clinical symptoms (e.g., depression and anxiety [64]), focusing on symptom reduction alone may not be sufficient for suicide prevention. Given that perceived social norms are a key influence on intention formation, normative feedback could be an effective means of reducing intentions to engage in self-harm and suicidal behaviours and in promoting help-seeking behaviours, although how effective such feedback would be with those in acute distress is not clear.

There have been a number of suicide prevention interventions using social norms feedback. Several studies have attempted to change perceived norms, or have directly measured changes in perceived social norms, in relation to intentions to seek mental health help [122–124] or in intervening and referring others for support, including for suicidality [125–127]. There is also some evidence to suggest that individuals overestimate rates of suicidal thoughts amongst peers [124, 128], which may indicate that challenging perceived descriptive norms amongst the wider population in terms of the frequency of suicidal ideation or self-harm behaviour could be fruitful for prevention efforts. There may be opportunities to better understand the role of perceived social norms amongst those without current or past personal histories of self-harm or suicidality, given that social norms can be powerful in terms of communicating acceptable standards and expectations for individuals in social groups. This is a complex phenomenon to study as there is evidence of overestimation of self-harm and suicidality behaviours, yet still self-harm and suicidality are viewed as socially unacceptable (e.g. [84]), which may indicate complex interactions between perceived descriptive norms (behaviour frequencies) and injunctive norms (e.g. perceived acceptability). For example, there is evidence that individuals exposed to a suicide in their social networks may perceive suicide as more common, yet still somewhat incomprehensible as an act, than those not exposed [121, 129]. Such (mis)perceived social norms may increase risk amongst those exposed to self-harm and/or suicide by making suicide appear to be a more frequent behaviour than the actual social norm but lead to feelings of shame after engaging in the same behaviours which are perceived to be socially undesirable. Such complexities represent a challenge for potential interventions, especially when ensuring that social normative feedback does not have unintended negative consequences or boomerang effects, such as increasing symptoms or distress, increasing feelings of shame by violating perceived norms, or making suicidal behaviours or planning appear more frequent than the reality [130]. Although, there is evidence that boomerang effects do not occur after normative feedback for other health behaviours, such as alcohol use [131].

Strengths and limitations

There are several strengths and limitations to consider with the present review. A systematic search of the published literature was conducted, across a range of databases and using a broad search strategy to ensure the maximum return of eligible studies. It should be noted that only published English-language studies were included in this review which may limit the findings’ generalisability. Due to the nature of the present scoping review, which aimed to map the extent of the literature, caution is needing in drawing definitive conclusions about the strength of the relationship between perceived social norms with NSSI and suicidality outcomes. In addition, considering the heterogenous nature of the reviewed studies, particularly in the operationalisation and measurement of perceived social norms, it would not appear to be appropriate to conduct a meta-analysis or qualitative meta-synthesis without a narrower, more selective, systematic review of the literature. The variety of the types of norms featured in the reviewed studies, as well as differences in the social norms measures used across studies and lack of theory-informed studies, also represents a challenge for ascertaining the role of perceived social norms in self-harm and suicidality. However, the present review has identified a number of quantitative and qualitative studies investigating the role of perceived social norms in NSSI and suicide-relevant outcomes, and has identified several key types of social norms that have been associated with these outcomes.