Introduction

A global health concern is self-harm, a behavior that is highly prevalent in young people. Compared to young people who do not self-harm, those who self-harm are more likely to experience adverse outcomes, including non-fatal outcomes such as financial hardship and poor mental health, and fatal outcomes such as suicide (Borschmann et al., 2017; Daukantaitė et al., 2021; Hawton et al., 2015). Despite the high prevalence of self-harm in young people and the serious consequences of self-harm, only a small proportion of young people seek help for their self-harm (e.g., Morey et al., 2008; Ystgaard et al., 2009), emphasizing a need to better understand the factors that influence help-seeking for self-harm in this population. The current study addresses this need, by taking a systems thinking perspective in the synthesis of the existing literature on the facilitators and barriers to help-seeking for self-harm in young people.

It has been estimated that each year, more than 700,000 people worldwide die by suicide (World Health Organization, 2021). For every death by suicide, many more people engage in self-harm, with or without suicidal intent. There is evidence indicating complex differences between people who self-harm with suicidal intent and those who self-harm without suicidal intent. For example, Mars et al. (2014) reported that people who self-harm with suicidal intent tend to experience higher levels of depressive and anxiety symptoms compared to those who self-harm without suicidal intent. Cleare et al. (2021), more recently, reported that people who self-harm with suicidal intent experience greater suicidal ideation, feelings of defeat and internal entrapment, and perceived burdensomeness, compared to those who self-harm without suicidal intent. The added complexity is that suicidal intent is temporally dynamic, such that suicidal intent fluctuates from time to time (Bryan, 2020). Despite these complexities, the overarching issue of self-harm warrants attention because self-harm remains an important risk factor for suicide (Carroll et al., 2014). There is evidence indicating that one in 25 people who present to the hospital for self-harm, irrespective of suicidal intent, die by suicide within five years of their presentation to the hospital (Carroll et al., 2014). Thus, the current study adopts a broad definition of self-harm; that is, any non-fatal act of injuring oneself deliberately, irrespective of the degree of suicidal intent (Knipe et al., 2022). Self-harm is viewed as a coping mechanism for managing emotional distress (Royal College of Psychiatrists, 2010).

Self-harm is prevalent in many populations, including people experiencing homelessness (e.g., Barrett et al., 2018), people from low- and middle-income countries (e.g., Knipe et al., 2019), and older people (e.g., Morgan et al., 2018). Self-harm is also highly prevalent in adolescents or young people (e.g., Gillies et al., 2018). In self-harm and suicide research, the period of adolescence is often defined by the chronological age of the young person (as opposed to their physical or cognitive development status) but there is little consensus on the chronological age definition (Hawton et al., 2012; see also Sawyer et al., 2018). The age range used to define adolescents or young people in self-harm and suicide research is broad, with a lower age limit of 10 years (e.g., Beckman et al., 2018) to an upper limit of 30 years (e.g., Cleary et al., 2017). Thus, the current study adopts a broad definition of adolescence, considering studies that described their sample as young people (e.g., youth, adolescents, teens, teenagers). Self-harm in young people has elicited increased attention worldwide because the prevalence of self-harm is higher in young people than in the general community (e.g., Australian Institute of Health & Welfare, 2024; McManus et al., 2019) and self-harm tends to first occur and peak in adolescence (e.g., Nock et al., 2013). It has been estimated that 20% of young people engage in self-harm and the average age of the first self-harm incident is 13 years (Gillies et al., 2018; Lucena et al., 2022; Nock et al., 2013). The prevalence of self-harm has continued to increase over the last few decades (Griffin et al., 2018; McManus et al., 2019; Sara et al., 2022). The study by Griffin et al. (2018) indicated a 22% increase in hospital emergency department presentations for self-harm between the 10-year period of 2007 and 2016, with more pronounced increases in females and young people aged 10 to 14 years. The study by Sara et al. (2022) also indicated an increase in the prevalence of self-harm in young people since COVID-19, with more pronounced increases in young people from socio-economically advantaged and urban areas. Importantly, self-harm in adolescence increases the likelihood of adverse experiences later in life (Borschmann et al., 2017; Daukantaitė et al., 2021; Hawton et al., 2015). For example, the population-based cohort study by Borschmann et al. (2017) indicated that people who self-harm in their adolescence (at 15 years of age) were more likely to encounter social disadvantages (e.g., financial hardship) and poorer mental health (e.g., substance dependence) that persisted into adulthood (at 35 years) when compared to people who did not self-harm in their adolescence. Self-harm in adolescence also increases the risk of fatal outcomes such as suicide (Hawton et al., 2015).

The lasting impacts of self-harm in adolescence on personal and social functioning and quality of life underscores the need to address self-harm in this population. Young people who self-harm would benefit from support and resources to not only manage and/or reduce their self-harm but also identify and confront the causes of this behavior (e.g., underlying emotional distress). A necessary step in accessing the support and resources required is that the young person seek help for their self-harm and this involves the disclosure of their self-harm. However, young people who self-harm are often reluctant to disclose their self-harm, keeping it a secret (Chandler, 2018). Help-seeking is described as “any action or activity carried by a [young person] who perceives [themselves] as needing personal, psychological, affective assistance of health or social services with the purposes of meeting this need in a positive way” (Barker, 2007). Help can be sought from formal, professional sources (e.g., psychologists, counselors) or informal sources (e.g., family, friends). There is evidence indicating that young people prefer seeking help from informal sources over formal sources (Rickwood & Braithwaite, 1994; Rickwood et al., 2007). It has been estimated that around 30% of young people sought help from informal sources following their self-harm (Ystgaard et al., 2009) while less than 20% of young people sought help from formal sources (Morey et al., 2008; Ystgaard et al., 2009). Irrespective of the type of help-seeking sources, the low rates of help-seeking for self-harm in young people are a concern as it suggests that young people who self-harm are not receiving the support and resources they need. These unmet needs will likely contribute to the rising prevalence of self-harm in this population.

Several reviews have been conducted to synthesize the factors that influence help-seeking, with a few of these reviews focused only on the facilitators and barriers to help-seeking for self-harm in young people (Michelmore & Hindley, 2012; Rowe et al., 2014) and other reviews on help-seeking for mental health problems in young people more generally (Aguirre Velasco et al., 2020; Barrow & Thomas, 2022; Gulliver et al., 2010; Radez et al., 2021). Across these reviews, there were facilitators and barriers common to both help-seeking for self-harm and mental health problems more generally. These common facilitators include assurance of confidentiality, being treated with respect and with no judgment, and previous positive experiences while common barriers include stigma, negative responses from others, and poor mental health literacy (Aguirre Velasco et al., 2020; Barrow & Thomas, 2022; Gulliver et al., 2010; Michelmore & Hindley, 2012; Radez et al., 2021; Rowe et al., 2014). However, across these reviews, it was apparent that some barriers relate to only help-seeking for self-harm and not mental health problems. Some notable barriers include the fear of being perceived as “attention-seeking”, or the fear that others would prevent them from future self-harm (i.e., fear around interventions) (Rowe et al., 2014). Thus, it would be necessary to refrain from assuming that the facilitators and barriers to help-seeking for mental health problems would also apply to help-seeking for self-harm. The latest review that focused only on help-seeking for self-harm in young people was published by Rowe et al. (2014), and this review synthesized the known facilitators and barriers across 11 studies, reporting that there are many barriers to help-seeking for self-harm in young people and comparatively, only a few facilitators. Given that nearly a decade has passed since the publication by Rowe et al. (2014), an updated review of the literature is needed. In addition to an updated review, it is proposed that a systems thinking perspective be taken to this review, to generate new insights and contribute to the knowledge base of help-seeking for self-harm in young people.

The discipline of systems thinking has had a long history in understanding and responding to complex problems (Leveson, 2004; Rasmussen, 1997). Proponents of systems thinking argue that, when attempting to understand and respond to complex problems, breaking the system (e.g., “help-seeking for self-harm” system) down into individual components and analyzing the components is limited (Leveson, 2004; Rasmussen, 1997). Instead, they argue that the system needs to be represented as a single unit of analysis, allowing the study of how its components interact to influence behavior (e.g., help-seeking for self-harm) (Leveson, 2004; Rasmussen, 1997). In complex systems, such as healthcare (Kannampallil et al., 2011; Raben et al., 2018), the interactions among components are non-linear and difficult to predict, leading to the potential for unintended consequences. For example, a decision to re-direct funding from face-to-face mental health services to online services, with the intent to increase help-seeking, can impact the accessibility and quality of face-to-face services. This, in turn, can have adverse impacts on young people who prefer and benefit from face-to-face services.

A well-established systems thinking framework is Rasmussen’s Risk Management Framework (Rasmussen, 1997), which was developed initially to understand accident causation but has since been applied more broadly to understand system performance. Rasmussen’s Risk Management Framework posits that systems comprise several hierarchical levels (e.g., government, regulators, local government, service delivery, individuals, and infrastructure) (Rasmussen, 1997). Stakeholders at each of these levels share the responsibility for the system’s performance and safety. This is because the decisions and actions of all system stakeholders, from those proximal in space and time to the behavior (e.g., a young person who self-harm, their friends and family), to those more distal in space and time (e.g., government, policymakers), contribute to the emergent behavior of the system. Rasmussen’s Risk Management Framework has been used to understand, holistically, the network of factors that influence behavior by representing where the factors reside across the system hierarchy (Rasmussen, 1997; Svedung & Rasmussen, 2002). The use of this systems thinking framework can support the identification of leverage points in the system, where interventions can result in effective behavioral change across the system (Meadows, 1999).

A handful of studies have applied systems thinking frameworks and methods to public health domains (e.g., Austin et al., 2002; Koorts et al., 2022), including in mental health (e.g., Lane et al., 2020; Occhipinti et al., 2021; Vacher et al., 2023). Of most relevance across these studies, is the study conducted by Lane et al. (2020). Lane et al. applied Rasmussen’s Risk Management Framework to understand barriers to treatment access for people with eating disorders. In their study, they extracted the known barriers to treatment access for people with eating disorders from the existing literature and mapped these barriers across the “eating disorder treatment” system hierarchy. In doing so, Lane et al. reported that many of the known barriers related to lower levels of the system hierarchy—individuals and social processes (e.g., belief in eating disorder seriousness, stigma and shame, attitudes towards professional help). Comparatively, there were fewer known barriers related to higher levels of the system hierarchy (e.g., government policy and budgeting, service delivery). Their study informed the current understanding of treatment access for people with eating disorders, by highlighting the gaps in knowledge and the importance of studying system-level barriers to improve treatment access. In addition to the insights generated from the use of Rasmussen’s Risk Management Framework in the study by Lane et al., important insights have also been generated from the use of other systems thinking frameworks and methods. For example, the study by Occhipinti et al. (2021), which applied system dynamics modeling to develop a simulation model of mental health system behavior. Using system dynamics modeling, they simulated the effects of different programs and initiatives in the mental health care system on suicidal behavior in a few regional Australian populations. Their study informed the optimal mix of programs and initiatives that can be implemented within the mental health system to reduce deaths by suicide (by up to 16% over 10 years) across the region. The point here is that systems thinking frameworks and methods (e.g., Rasmussen’s Risk Management Framework, system dynamics modeling) can generate new, important insights that can enhance the current knowledge base and inform policy decision-making. These insights would not have been observed if reductionistic approaches were used to address complex problems.

Current Study

Self-harm is highly prevalent in young people, representing a major health concern in this population. Young people who self-harm would benefit from support and resources to address their self-harm., but only a small proportion of young people seek help for their self-harm. Reviews have been conducted previously to understand the factors that influence help-seeking for self-harm in this population, with the last review published in 2014. An updated systematic review of the factors that influence help-seeking for self-harm in young people is, therefore, timely. In particular, a systematic review that adopts a new perspective on this topic, to generate new insights and further contribute to the knowledge base. The current study was designed to address this gap. More specifically, the current study aimed to synthesize the existing literature on the facilitators and barriers to help-seeking for self-harm in young people, and analyze and map these facilitators and barriers across the help-seeking system hierarchy using a systems thinking framework.

Method

Protocol

This systematic review was prospectively registered in the International Prospective Register of Systematic Reviews (PROSPERO; registration number CRD42023384541) (Booth et al., 2011). This review was conducted in line with the Preferred Reporting Items for Systematic Review and Meta-analyses (PRISMA) guidelines (Page et al., 2021). Ethics approval was not required for this systematic review as no human recruitment or participant was involved.

Electronic Search

The literature search for the systematic review was carried out between the period of 9 December 2022 and 5 January 2023 (inclusive), on four databases: PubMed, Medline, PsycINFO, and Scopus. Covidence (Veritas Health Innovation, n.d.), a web-based collaborative software platform, was used to facilitate the literature searching and screening process.

The aim and scope of the systematic review determined the Boolean search terms relating to “self-harm”, “youth”, and “help-seeking”. All of the terms were searched by title, abstract, and keywords, to ensure that the initial search results contained only relevant records and that they were manageable in further rounds of screening. The final search strategy was consistent across all four databases and was as follows: ((“self harm” OR “self-harm”) OR (suicide*) OR (suicide* AND (attempt OR ideation OR behavio*)) OR (“self injur*” OR “self-injur*) OR (“non-suicidal” OR “nonsuicidal” AND ((“self harm” OR “self-harm”) OR (“self injur*” OR “self-injur*”))) OR “deliberate” AND ((“self harm” OR “self-harm”) OR (“self injur*” OR “self-injur*”)))) AND (young OR youth OR adolescen* OR teen OR children) AND (((seek* OR sought) AND (help OR service OR support OR treat*)) OR disclos*).

Although this systematic review provides an updated review to the one published by Rowe et al. (2014), no restrictions on publication dates were applied to the search. A reason for this is that this systematic review represents the first in the field of self-harm research to apply a systems thinking perspective in the analysis of factors that influence help-seeking for self-harm, and as such, restrictions on publication dates will result in an incomplete understanding of these factors across the help-seeking system hierarchy.

Eligibility Criteria

A set of inclusion and exclusion criteria was applied during the screening process to identify records relevant to this review.

Inclusion Criteria

Records were included in the review if they met all of the following criteria:

  1. (i)

    The primary study sample involved young people with a history of self-harm. The definition of self-harm has been a point of controversy among researchers regarding the inclusion of suicidal intent in its definition (Butler & Malone, 2013; Zareian & Klonsky, 2019). Given the review is focused on help-seeking and not on the aetiology of self-harm, a broad definition of self-harm was adopted in this review, and studies where self-harm was described as an act of self-injury, irrespective of suicidal intent, were included. The adoption of this broad definition also ensures that relevant records were not excluded.

  2. (ii)

    The definition of “young people” was not restricted by age. Given there is little consensus on the age definition for adolescence (Sawyer, 2018), studies in which the sample was described to consist of young people (e.g., youth, adolescents, teens, teenagers) were included.

  3. (iii)

    The outcome of interest in the study was help-seeking for self-harm, from either formal or informal sources.

  4. (iv)

    The record was published in English language and the full-text version of the record was available.

Exclusion Criteria

Records were excluded from the review if they did not meet the inclusion criteria or any of the following criteria:

  1. (i)

    The outcome of interest in the study was help-seeking for mental health problems more generally.

  2. (ii)

    The study was not peer-reviewed (e.g., pre-prints, conference articles, student theses).

  3. (iii)

    The record was a review article (e.g., narrative review, systematic review), a commentary, or a personal retrospective article.

The second author conducted the initial search, removed record duplicates, and reviewed the retrieved records against the eligibility criteria, first by title and abstract, and then by the full-text version. Discussions were held between the first and second authors if there were insufficient details reported in the title, abstract, or full-text version of the record. A consensus decision was then made to include or exclude the record from this review. The screening process resulted in the final set of records, of which all were articles, in the systematic review.

Quality Assessment

The quality of the articles included in the systematic review was assessed using the Mixed Methods Appraisal Tool (Hong et al., 2018). The Mixed Methods Appraisal Tool allowed for the quality assessment of studies with diverse designs (e.g., qualitative, quantitative descriptive, and mixed-methods studies) (Pace et al., 2012; Pluye et al., 2009). The Mixed Methods Appraisal Tool checklist includes two screening questions relating to the research question of the study and the data collected to address the research question, and five questions relating to the methodological quality of the study. Each question required a response of “Yes” (scored as “1”), “No” or “Can’t tell” (both scored as “0”). For articles with qualitative and quantitative study designs, the score on each of the five questions relating to the methodological quality was summed and the total score reflected the quality of the study. For articles with a mixed-methods study design, the score on each of the questions relating to methodological quality (total of 15 questions) was summed, divided by 15, and then multiplied by five. A total score of zero indicates the lowest quality and five indicates the highest quality. The two screening questions were not scored.

The second author conducted the quality assessment of the studies in the included articles and the first author conducted an independent quality assessment for 25% of these articles (randomly allocated) as an inter-rater reliability check for this assessment. The inter-rater reliability score (Cohen’s Kappa) was 0.90, indicating almost perfect agreement (Landis & Koch, 1977). Discussions were held between the first and second authors to resolve the disagreements.

Data Extraction and Presentation

Relevant information was extracted from each article. Publication information extracted included the author names and publication year. Study characteristics extracted included the geographical context of the study, sample size, participant demographics (age, gender or sex), study design, and definition of self-harm used in the study.

Factors that influence help-seeking for self-harm were also extracted, where they were reported by the author(s) of the included articles to have influenced help-seeking for self-harm. The factors were coded and reported in this systematic review as facilitators of help-seeking, barriers to help-seeking, and other factors associated with help-seeking. For this review, facilitators were defined as factors that promote or enable help-seeking for self-harm (e.g., assurance with trust, privacy, and confidentiality). Barriers were defined as factors that prevent or obstruct help-seeking for self-harm (e.g., fear of worrying others or being burdensome). Other factors were defined as factors that describe the demographics of young people (e.g., gender identity) or characteristics of their help-seeking (e.g., before or after their self-harm) and were reported to differ in their likelihood of help-seeking for self-harm.

The facilitators and barriers to help-seeking for self-harm were mapped onto a “help-seeking system” hierarchy (underpinned by Rasmussen’s Risk Management Framework; Rasmussen, 1997), of which there were six levels.

  1. 1.

    Government: Federal and state government agencies and policymakers.

  2. 2.

    Regulatory bodies and associations: Organizations that facilitate the regulation of the domain (e.g., Australian Health Practitioner Regulation Agency), lobby groups, and mental health advocacy bodies.

  3. 3.

    Local area government and organization management: Service management and local organizations (e.g., schools, hospitals, institutions).

  4. 4.

    Service delivery and social environment: Health professionals and the social support networks (e.g., friends, family, carers) of young people who self-harm.

  5. 5.

    Individuals and processes: Young people who self-harm.

  6. 6.

    Equipment, infrastructure, and surroundings: The equipment, infrastructure, and surroundings of the local environment with which young people interact when seeking help.

For example, if a study identified that having a previous positive experience encouraged young people to seek help for self-harm, the facilitator “Previous positive experience” was placed at the Individuals and processes level of the system hierarchy, as this facilitator relates to young people’s perceptions of their last help-seeking attempt. If a study identified that the lack of support or advocacy prevented them from seeking help for self-harm, the barrier “Lack of support/advocacy” was placed at the Service delivery and social environment level of the system hierarchy, as this barrier relates to the inaction by others around young people. In addition to the extraction of factors that influence help-seeking for self-harm, relationships between the factors (causal or correlational) were extracted, where they were reported by the author(s) of the included articles. For example, if a study identified that long wait times for a mental health support service led to young people feeling ignored or abandoned, and these feelings prevented future help-seeking for self-harm, the relationship between “Long wait times” and “Feeling ignored or abandoned” was documented.

The first and second authors extracted the data from the articles included in this systematic review and the fourth author conducted an independent data extraction for 25% of these articles (randomly allocated) as an inter-rater reliability check for this data extraction. The inter-rater reliability score (Cohen’s Kappa) was 0.75, indicating substantial agreement (Landis & Koch, 1977). Discussions were held between the first, second, and fourth authors to resolve the disagreements.

Results

Full-Text Selection

The search across the four databases (PubMed, Medline, PsycINFO, Scopus) retrieved a total of 4953 records. Of these records, 2803 duplicate records were identified and removed from the systematic review. Following the removal of these duplicate records, the title and abstract of the remaining 2132 records were screened based on the eligibility criteria. This resulted in a total of 1918 records excluded. The full-text version of the remaining 214 records was assessed for their eligibility and this resulted in a further 181 records being excluded. The screening process resulted in the inclusion of 33 records, all of which were articles, in this systematic review (Fig. 1).

Fig. 1
figure 1

PRISMA flowchart of the systematic review process (Color figure online)

Full size image

Quality Assessment

The articles included in the systematic review were assessed for their study methodological quality using the Mixed Methods Appraisal Tool (Hong et al., 2018). The quality scores for articles that adopted a qualitative study design (n = 13) ranged from four to five, quantitative study design (n = 13) ranged from two to four, and mixed-methods study design (n = 7) ranged from two to four. The quality score for each of the articles included in the systematic review is presented in Fig. 2 (refer to Supplementary Material for raw scores of each item in the Mixed Methods Appraisal Tool checklist).

Fig. 2
figure 2

Quality assessment of the articles included in the systematic review. Note. Quality assessment was conducted using the Mixed Methods Appraisal Tool (Hong et al., 2018). The three panels in this figure reflect the study design (i.e., qualitative, quantitative, mixed-methods) adopted in the respective articles. The Y-axis indicates the article citation and the X-axis indicates the quality score (Color figure online)

Full size image

Publication Information

The articles included in this systematic review were published between the years 2003 and 2023, with 2017 and 2021 having the highest number of publications (n = 4 in each year).

Study Characteristics

Study characteristics regarding the geographical context of the study, sample size, and participant demographics (age, gender or sex) from each of the articles included in this systematic review can be found in Table 1.

Table 1 Publication information and study characteristics of the articles included in the systematic review
Full size table

The largest proportion of studies were conducted in the United Kingdom (UK; n = 11, 35.48%). This was followed by Australia (n = 6, 18.18%) and the United States of America (USA; n = 5, 15.15%). The sample sizes reported across the studies varied extensively, from seven participants (McAndrew & Warne, 2014) to 11406 participants (Fadum et al., 2013). The age distribution of participants across the studies was not consistently reported, with some studies reporting detailed information (mean, standard deviation, range) and others reporting minimal information (range only). The youngest age for young people (lower limit) across the studies was 11 years (Lustig et al., 2021; Mojtabai & Olfson, 2008) and the oldest age (upper limit) was 30 years (Cleary, 2017). The gender or sex distribution of participants across the studies was also not consistently reported. Some studies reported gender distribution while others reported sex distribution. Further, some studies reported detailed information on the gender or sex distribution while others reported minimal information (e.g., the number or proportion of participants for only one gender or sex, even if their study included participants of other genders or sex).

Study characteristics regarding the study design and the self-harm definition adopted in the study from each of the articles included in this review can also be found in Table 1. There was an equal number of articles that adopted a qualitative (n = 13, 39.39%) or quantitative study design (n = 13, 39.39%), and the remaining articles adopted a mixed-methods study design (n = 7, 21.21%). The definition of self-harm differed across the articles included in this review. In many articles, self-harm was defined as a deliberate act of injuring oneself, but definitions differed in whether there was suicidal intent behind this behavior. A handful of articles included in this review focused on suicide attempts, which could be considered as self-harm with suicidal intent (i.e., where the intended outcome is fatal). Some of the articles focused on suicidality, defined “the risk of suicide, usually indicated by suicidal ideation or intent, especially as evident in the presence of a well-elaborated suicidal plan” (American Psychological Association, 2018), which can also include self-harm behaviors (Keefner & Stenvig, 2021).

Factors that Influence Help-Seeking for Self-Harm

The factors that influence help-seeking for self-harm extracted from the articles can be found in Table 2, and the descriptions for these factors can be found in Table 3.

Table 2 Factors that influence help-seeking for self-harm extracted from the articles included in the systematic review
Full size table
Table 3 Definitions of the facilitators and barriers to help-seeking for self-harm extracted from the articles included in the systematic review
Full size table

The known facilitators of help-seeking for self-harm and where they reside across the help-seeking system hierarchy can be found in Fig. 3. A total of 17 distinct facilitators for self-harm in young people were extracted from the articles. Across the system, there was one facilitator relating to the regulatory bodies and associations, eight facilitators relating to the service delivery and environment, seven relating to the individual and processes, and one relating to equipment, infrastructure, and surroundings. There were no facilitators relating to the government, or local area government and organisation management. The facilitators that were most frequently reported across the studies included non-judgment or acceptance of self-harm by others (n = 7), communication by others (n = 5), online service delivery (n = 4), and assurance of trust, privacy, and confidentiality from others (n = 3).

Fig. 3
figure 3

The known facilitators of help-seeking for self-harm in young people, across the help-seeking system. Note. The number in parenthesis indicates the number of studies from which the facilitator was extracted (Color figure online)

Full size image

The known barriers to help-seeking for self-harm and where they reside across the system can be found in Fig. 4. A total of 34 distinct barriers to help-seeking for self-harm in young people were extracted from the studies. Across the system, there were two barriers relating to regulatory bodies and associations, two relating to the local area government and organizational management, four relating to service delivery and environment, and 26 relating to the individual and processes. There were no barriers relating to the government, or equipment, infrastructure, and surroundings. The barriers that were most frequently reported across the studies included young people feeling ashamed or embarrassed of their self-harm (n = 12), their perception that they can manage their self-harm on their own (n = 11), their fear of worrying others or being a burden (n = 8), their fear of being judged or stigmatized (n = 8), and trust, privacy, and confidentiality concerns (n = 7).

Fig. 4
figure 4

The known barriers of help-seeking for self-harm in young people, across the help-seeking system. Note. The number in parenthesis indicates the number of studies from which the barrier was extracted (Color figure online)

Full size image

The relationships between the factors reported in the articles were also extracted and are illustrated in Fig. 5. The relationships extracted were only identified once or twice across the studies, and most of these relationships were between a facilitator and an other factor, or between a barrier and an other factor. Gender identity was most frequently identified as having a relationship with a barrier. For example, gender identity was associated with young people being discriminated against or victimized (Burke et al., 2021), feeling ignored or abandoned (Fortune et al., 2008b), and their fear of worrying others or being burdensome (Fortune et al., 2008b). Only a handful of relationships were found between facilitators or barriers. One example was that long wait times for a support service were associated with young people feeling ignored or abandoned (Byrne et al., 2021).

Fig. 5
figure 5

Relationships between the known facilitators, barriers, and other factors, across the help-seeking system. Note. The green-colored boxes indicate the known facilitators, the red-colored boxes indicate the known barriers, and the white-colored boxes indicate other factors extracted from the studies. The number in parenthesis indicates the number of studies from which the factor was extracted. Relationships are indicated by the lines linking the factors, and the number of studies from which the relationships were extracted are indicated on the lines. The green-colored lines indicate relationships between known facilitators, the red-colored lines indicate relationships between known barriers, the yellow-colored lines indicate relationships between a known faciliator and an other factor, the purple-colored lines indicate relationships between a known barrier and an other factor, and the blue-colored lines indicate relationships between other factors (Color figure online)

Full size image

Discussion

Self-harm in young people represents a major health concern, with these young people being at risk of adverse social and health outcomes later in their lives. Young people who self-harm are likely to benefit from support and resources to address their self-harm but they often do not seek help. There is a need to understand the factors that influence help-seeking for self-harm in young people, yet the most recent review on this topic was published in 2014 (Rowe et al., 2014). This current study was conducted to provide an updated review of the factors that influence help-seeking for self-harm in young people. More specifically, this systematic review (i) synthesized the existing literature on the known facilitators and barriers to help-seeking for self-harm in young people, and (ii) used a nuanced approach, a systems thinking framework (Rasmussen’s Risk Management Framework; Rasmussen, 1997) to analyze these factors across the help-seeking system hierarchy. This review is the first in the field to have taken a systems thinking perspective to understand the factors that influence help-seeking for self-harm in young people.

Facilitators and Barriers to Help-Seeking for Self-Harm

In recent years, researchers have proposed the use of systems thinking frameworks and methods to study mental health, with some having implemented these frameworks and methods to better understand areas of mental health (e.g., Lane et al., 2020) or to inform priority-setting (e.g., Occhipinti et al., 2021). The use of a systems thinking framework, Rasmussen’s Risk Management Framework (Rasmussen, 1997), in this systematic review, provided insights into the key focus areas in research on help-seeking for self-harm in young people and demonstrated knowledge gaps in the existing literature. Five key findings emerged from this systems thinking-based analysis.

The first finding was that the majority of the known facilitators and barriers to help-seeking for self-harm were related to lower levels of the system hierarchy. More specifically, these factors were related to the decisions and actions of the professionals who provide care (e.g., counselors, psychologists) and social support networks (e.g., friends, family) (Service delivery and social environment level) and those of the young people who self-harm (Individuals and processes level). Comparatively, there were fewer known facilitators and barriers related to higher levels of the system, which include the Local area government and organization management, Regulatory bodies and associations, and the Government levels. This disproportionate representation of factors in the system is a common observation across different domains of research, as evident in studies that have used systems thinking frameworks and methods. For example, studies have sought to identify what is known regarding barriers to treatment access for individuals with eating disorders (Lane et al., 2020), the factors that enable child sexual abuse in sport (Dodd et al., 2023), the factors that contribute to cycling incidents (Salmon et al., 2022), and the factors influencing risk at rail level crossings (Read et al., 2021), have all concluded that more is known regarding the lower level factors.

The disproportionate focus on the health professionals who provide care, social support networks, and young people identified in this systematic review is likely to reflect a knowledge gap, resulting from the paucity of research on how decisions and actions of those in the higher levels of the system hierarchy, especially the Government level, influence help-seeking for self-harm in young people. This knowledge gap suggests that research to date has overlooked key systems thinking tenets; namely, that outcomes (e.g., help-seeking or a lack of help-seeking) are created by the interactions between stakeholders at all system levels and are the responsibility of all system stakeholders (Rasmussen, 1997). This first finding, therefore, suggests that more expansive research is required to better understand help-seeking for self-harm, with a sound starting point being to investigate the impact of policy initiatives and changes to these initiatives on help-seeking for self-harm. An example policy initiative in Australia is the Better Access initiative that provides rebates to eligible individuals so they can access up to 10 individual and up to 10 group allied mental health services (Australian Government Department of Health & Aged Care, 2022). One of the eligibility requirements to use Better Access is that the individuals must have a mental health diagnosis. However, there is some evidence suggesting that this requirement is a barrier to help-seeking (Pirkis et al., 2022). Future research investigating how factors relating to the higher levels of the help-seeking for self-harm system influence help-seeking for self-harm in young people is recommended, as it is highly likely that these factors would have impacts on help-seeking for self-harm in young people.

The second finding was that there were 17 distinct facilitators of help-seeking for self-harm in young people. Among these known facilitators, many were related to the actions and decisions of professionals who provide care and the social support networks of young people who self-harm (Service delivery and social environment level). More specifically, these facilitators included non-judgment and acceptance of self-harm by others, communication by others, and assurance of trust, privacy, and confidentiality from others. Non-judgment or acceptance of self-harm by others was the most frequently reported facilitator across the studies. This finding is important because it suggests that a potential avenue to encourage help-seeking behaviour is through the interactions of young people who self-harm with professionals who provide care and their social networks. These professional and social support structures play a critical role in reflecting positive values and beliefs that can facilitate help-seeking behaviors (Lauriks et al., 2014; Schenk et al., 2018). Additionally, future research is needed to better understand the drivers of these facilitators; that is, factors that promote these facilitators of help-seeking for self-harm.

The third finding was that there were 34 distinct barriers to help-seeking for self-harm in young people. Among these known barriers, many were related to the young people who self-harm themselves (Individuals and processes level). More specifically, these barriers included disbelief in the seriousness of self-harm, previous negative experiences, their personal feelings (e.g., feeling ashamed or embarrassed about their self-harm) and personal fears (e.g., fear of being judged or stigmatized for their self-harm). The most frequently reported barriers included young people’s personal feelings of shame and embarrassment, their belief that they can self-manage their self-harm, and their fear of worrying others or being burdensome. Notably, many barriers related to young people’s personal feelings and personal fears about how others may respond to their self-harm. These feelings and fears may not be validated—others may not respond to their self-harm the way they had expected—but they remain significant barriers to help-seeking for self-harm in this population. These barriers are exceptionally difficult to address, as they can be deeply entrenched in a young person’s values and long-held beliefs, or how they think of and perceive their self-harm. However, one could argue that while these barriers are related to the young people who self-harm themselves, these are driven by societal beliefs and norms (e.g., the societal stigma associated with self-harm may result in a young person to self-stigmatize their own self-harm). This means that a possible leverage point to break down the barriers to help-seeking is through addressing societal attitudes and treatment towards one another.

The fourth finding, which was alluded to earlier in the discussion of our second and third findings, was that there were twice as many barriers to help-seeking for self-harm than there were facilitators. Our systems thinking-based analysis of the facilitators and barriers to help-seeking for self-harm indicated that more is known about the barriers to help-seeking for self-harm compared to facilitators. It may be the case that there are simply more barriers than there are facilitators, or that the research focus has been on barriers. The latter would suggest that the field does not have an in-depth understanding of the facilitators of help-seeking for self-harm. Future research could explore the use of methods such as appreciative inquiry (Bushe, 2011) to better understand the facilitators of help-seeking for self-harm and how these factors may be integrated into practice to enhance help-seeking. An additional benefit could be gained by integrating such methods with systems thinking frameworks and methods (e.g., Rasmussen’s Risk Management Framework), encouraging the consideration of all levels of the help-seeking for self-harm system hierarchy and how the decisions and actions of actors who reside in each level influence the young person’s help-seeking behaviors.

The fifth and final finding was that there is a lack of research that considers the relationships between the known facilitators or barriers to help-seeking for self-harm in young people. This is perhaps most concerning, as understanding the interactions that drive behavior is a central tenet of systems thinking. Further, it is arguably not possible to optimize behavior without understanding such interactions. Only a few studies have identified relationships, and of the relationships identified, the most frequently reported relationships were associated with the gender identity of the young person, a demographic factor, rather than another facilitator or a barrier. For example, gender identity was associated with the fear of being judged or stigmatized, resulting in a barrier to help-seeking for self-harm (Fortune et al., 2008b) or with trust, privacy, and confidentiality concerns (Fortune et al., 2008a). One of the characteristics of a complex system (Cilliers, 1998), and a principle underlying systems thinking (Rasmussen, 1997), is that behavioral outcomes (e.g., help-seeking for self-harm) result from interactions between decisions and actions across the entire system. Thus, the lack of research that considers relationships between the facilitators or barriers suggests that researchers have typically taken a reductionist view of this complex issue, by analyzing these facilitators or barriers in isolation. If help-seeking for self-harm in young people is viewed as a complex problem, then it is critical that the interactions between facilitators and barriers that influence young people’s help-seeking behaviors are well-understood.

A Note on the Quality of the Articles

In discussing the key findings from this systematic review, it is worth noting the outcomes of the quality assessment of the articles included in the review. Articles were assessed for their methodological quality using the Mixed Methods Appraisal Tool (Hong et al., 2018) but were not excluded based on low methodological quality. Overall, the articles included in this review demonstrated reasonable levels of methodological quality. However, interestingly, articles that adopted a qualitative study design scored higher than those that adopted a quantitative or mixed-methods study design. This does not necessarily suggest that qualitative study designs are superior methodologically compared to quantitative study designs or mixed-methods study designs, primarily because they were assessed on different criteria. Common methodological issues observed in the quantitative study design or quantitative components of the mixed-methods study design included insufficient, or the lack of, information reported around the representativeness of the sample and risk of nonresponse bias (see Supplementary Material). These methodological issues have also been noted in other systematic reviews (e.g., Thomason & Moghaddam, 2021; Zhuang et al., 2023).

Limitations

A limitation of this systematic review relates to the types of records included in the review; namely, only articles published in peer-reviewed journals were included. This means that findings from studies reported in other publication types; namely, gray literature such as conference proceedings, government reports, and student theses were not considered in this synthesis and analysis of this review. The inclusion of gray literature may pose methodological challenges in the review (e.g., lack of scientific rigor, additional time and resources in literature searching) but the benefits of including gray literature (e.g., reduced publication bias) could outweigh these challenges (Paez, 2017).

A second limitation relates to the broad definition of self-harm adopted in this review. The current, fifth edition Diagnostic and Statistical Manual of Mental Disorders (DSM-5) recognizes non-suicidal self-injury (i.e., self-harm without suicidal intent) as clinically distinct from suicidal self-harm (i.e., self-harm with suicidal intent) (American Psychiatric Association, 2013) but the diagnostic criteria for non-suicidal self-injury require further study and it remains controversial as to whether the definition of self-harm should include suicidal intent. Studies have demonstrated differences between self-harm with and without suicidal intent in their genetic etiology (Campos et al., 2020) and psychological risk profiles (Cleare et al., 2021; Mars et al., 2014). However, people who self-harm, engage in self-harm with and without suicidal intent (e.g., Wilkinson et al., 2011) and their suicidal intent can fluctuate temporally (e.g., Bryan, 2020). In light of the ongoing controversy and the inconsistent findings, this review adopted a broad definition of self-harm, including articles that focus on help-seeking for self-harm, irrespective of the suicidal intent behind the self-harm. By adopting this broad definition, it is assumed that the facilitators and barriers to help-seeking are similar across help-seeking for self-harm with suicidal intent and those without suicidal intent, when they may differ. Future research should consider whether the facilitators or barriers for help-seeking might differ when seeking help for self-harm with and without suicidal intent.

A third limitation relates to the variability in the age definition of young people across the articles included in this systematic review. The period of adolescence is not well defined, with some defining this period as ages between 10 to 19 years, and others 15 to 24 years or more inclusively, 10 to 24 years (Sawyer et al., 2018). Thus, this review included articles that described their study sample as young people, youth, adolescents, teens, or teenagers. How adolescence is conceptualized and defined has implications on our understanding of help-seeking behaviors, such that the developmental transition from being a child to being an adolescent can influence the way help is sought and/or the types of help-seeking sources they prefer. This is particularly important because many current health systems are limited in their ability to provide the comprehensive care young people require (Sawyer et al., 2018), which may consequently affect help-seeking behaviors.

The fourth limitation relates to the lack of consideration of gender or sex differences in this systematic review. This review did not consider gender or sex differences in the facilitators and barriers to help-seeking for self-harm; however, it was apparent that gender identity was associated with several factors (e.g., barrier: being discriminated against or victimized). Therefore, it would be important for these gender or sex differences in help-seeking behaviors to be accounted for when developing interventions or strategies to improve help-seeking in young people. It is worth noting that across the articles included in this systematic review, there was inconsistent reporting on gender or sex distribution of the study sample. Some studies reported gender distribution while others reported sex distribution, and some studies reported detailed information on gender or sex distribution while others reported minimal information. This inconsistent reporting of gender or sex distribution makes it difficult to assess whether there are gender or sex differences in the facilitators and barriers to help-seeking for self-harm. Future research should ensure that all relevant study sample characteristics are collected and reported.

The final limitation relates to the research field more broadly, and that is the conceptualization of help-seeking behaviors. Studies typically focus on assessing help-seeking behaviors at a single point in time and this focus means that only a fraction of one’s help-seeking process is considered. However, the help-seeking process is often prolonged, non-linear, and dynamic. In the context of young people who self-harm, a young person might “move in and out” of help-seeking for their self-harm, depending on their experiences of insecurity and unfamiliarity with mental health support, structural obstacles, and the process of finding and seeking help (e.g., Lost in Space model; Westberg et al., 2020). Future research should account for the dynamic nature of help-seeking for self-harm in young people, as not doing so will undermine the effectiveness of strategies implemented to improve help-seeking in this population.

Conclusion

Self-harm in adolescence has lasting impacts on their personal and social functioning and quality of life. Young people who self-harm would benefit from support and resources, but the rates of help-seeking in this population remain low. Thus, it is necessary to understand the facilitators and barriers to help-seeking for self-harm in young people, from a novel, systems thinking perspective. This systematic review synthesized the existing literature on the known facilitators and barriers to help-seeking for self-harm in young people from a systems thinking perspective. The systems thinking-based analysis revealed five key findings. First, there is a biased representation of research focus on the lower levels of the help-seeking for self-harm system. Second, many of the known facilitators related to the service delivery and social environment while, third, many of the known barriers related to young people who self-harm and their cognitions and feelings. Fourth, there were twice as many known barriers than there were known facilitators. Fifth and finally, there is a lack of research that considers the relationships between the known facilitators or barriers to help-seeking for self-harm in young people. In addition to contributing to the knowledge base on help-seeking for self-harm in young people, the review also identified key gaps in the knowledge base. Namely, a lack of research on how higher levels of the system influence help-seeking for self-harm in young people and how facilitators and barriers interact to influence help-seeking for self-harm in this population. This systematic review demonstrates the benefits of using systems thinking frameworks and methods in understanding the facilitators and barriers to help-seeking for self-harm in young people.