Socio-Demographic Factors and Co-Morbidities in a Sample of Australian Young People with High Risk of Suicide

Abstract

Introduction: Increasing rates of youth suicide are a growing concern worldwide. The single highest risk factor for suicide is a prior suicide attempt; however, knowledge around the broader factors that may increase the likelihood of the initial suicide attempt is limited. Understanding these risk factors is critical for clinicians to inform the development of early intervention strategies. This study examines the characteristics of a specific cohort of young people attending a tertiary mental health service in Melbourne, Victoria, following a suicide attempt or with persistent suicidal ideation. Method: A file audit of all young people attending the service was conducted from the commencement of the new service in 2022 through to the end of 2024. One hundred and seventy-one files were examined for specific data, including items such as age, gender, gender diversity, socio-economic factors and neurodivergence. Results: One-sample z proportion tests showed significantly higher proportions of young people who were LGBTIQIA+, transgender, gender diverse or non-binary (TGDNB), neurodivergent, not engaged in employment, education or training (NEET), and international students in the clinical cohort compared with the Australian population. Almost one-fifth of the cohort reported a history of trauma. Discussion: The study provides insight into characteristics observed among a cohort of suicidal young people. Findings highlight that young people from several marginalised groups were more frequently observed in this clinical cohort. The findings have implications for youth mental health services, supporting earlier identification of potential factors associated with suicidality so that prevention strategies can be implemented in a timely manner.

1. Introduction

Youth suicide is a preventable public health concern, representing the third leading cause of death for young people around the world [1]. The definition of what constitutes suicide, suicide attempts, and suicidal ideation varies within the literature; however, a useful definition is a self-injurious act with some evidence of an intent to die [2]. For the purpose of this study, young people were referred to the included service with either an observed or expressed intention to die by suicide. In Australia, over 3000 people die by suicide each year and youth suicide is the number one cause of death in young people aged 15 to 24 years. In 2023, 392 young people under the age of 24 died by suicide, with rates trending upwards over the past two decades [3,4]. For each death, there are approximately ten times as many hospitalisations each year for intentional self-harm [3]. In Australia, suicide rates appear to be increasing and remain higher than those reported in several comparable high-income countries, including the UK, Ireland, and Sweden [5].

Moreover, the high reported national rates conceal important differences in the young people most affected, with higher rates observed in rural and remote regions, as well as socio-economic disadvantage and barriers to accessing appropriate mental health care in some communities [6]. Between 16 and 19% of people who have sought hospital care after self-harm and suicide attempts may re-present to services, highlighting the presence of persistent risk [7,8]. A previous suicide attempt is the strongest predictor for future suicide [9]. Not everyone presents to mental health services following a suicide attempt; however, for those who do, suicide rates within the first three months after discharge are reported to be one hundred times greater than global suicide rates [10]. These statistics highlight the importance of understanding the characteristics of young people who present to mental health services with suicidality in order to support earlier identification and clinical management of risk.

There are relatively few naturalistic descriptions of the demographic and social characteristics of young people attending Australian community mental health services [11]. Considering that the literature suggests a third of adolescents with suicidal thoughts will go on to make a suicide attempt [12], improving understanding of how established co-morbidities and risk factors manifest among young people presenting to services may help inform earlier recognition of risk and improved clinical decision-making. Some risk factors in the literature are well-known, including substance use, non-suicidal self-harm, and adverse childhood experiences [5,7]. These factors rarely occur in isolation and often interact across developmental, psychological and social domains, contributing to cumulative vulnerability to suicidal behaviour [2].

A recent umbrella review by Prades-Caballero et al. [13] examined risk factors associated with suicidal behaviour in adolescents and identified four broad categories: individual, interpersonal, community, and societal factors. Individual factors include demographic and personal characteristics that may increase vulnerability to suicidal behaviour, such as age, gender, sexual orientation, emotional regulation difficulties, social isolation, and physical health conditions [13,14]. Childhood trauma, including sexual abuse and gender-based violence, has been identified as a risk factor for females, while emotional factors such as hopelessness were particularly associated with suicide risk among males. Several studies also reported that young people identifying as LGBTIQIA+ were at significantly higher risk of suicide attempts, which has been attributed to stigma, victimisation, and poor social acceptance.

Interpersonal factors were also found to be important contributors to suicidal ideation. These included adverse life events such as bereavement and childhood maltreatment, as well as relational difficulties including relationship breakdown, loneliness, and social isolation [14,15,16].

The review [13] conceptualised community factors as those centred around public institutions such as educational opportunities, while societal factors included broader social influences such as economic and political conditions. However, their review noted a lack of evidence examining community or societal issues as risk or protective factors for youth suicidality [13]. Despite this limited evidence, increasing attention has been given to the relationship between youth mental health and disengagement from education, employment, or training (NEET) [17], which is examined in the present study. A systematic review and meta-analysis examining the association between NEET status and mental health found that NEET status was a consistent predictor of suicidal behaviour [18]. While disengagement from school or work may further disadvantage young people experiencing mental health difficulties, it remains unclear whether educational or workplace stressors themselves may contribute to disengagement [19]. The limited evidence examining community and societal influences highlights an important gap in the literature and suggests that future research incorporating broader structural and environmental data may provide a more comprehensive understanding of youth suicidality.

Despite the awareness of such risk factors, suicide rates are increasing. Consequently, it has been established that improving the care received by people after a suicide attempt is a high priority [5,9]. Many people are discharged from the hospital without proper aftercare or may not proactively seek help post-discharge [7]. Initial assessments may prioritise safety planning rather than exploring and addressing the factors that have caused someone to contemplate suicide. Young people may then feel that their longer-term needs are unmet, leaving them unmotivated to seek further care [20]. Follow-up rates post-presentation to a health service with suicidality have been noted to be as low as 27% in some studies [21]. Intersecting risk factors demonstrate the complexity of suicide vulnerability. Understanding how these factors manifest in young people presenting to services is critical for informing early identification and triage practices when young people present at mental health services. A clearer understanding of the characteristics of this population may assist clinicians in identifying vulnerability earlier and tailoring support to improve engagement with at-risk young people.

This study therefore aims to describe the demographic and social characteristics of a cohort of young people receiving support from a post-suicidal engagement program following their presentation to a tertiary mental health service due to persistent suicidal ideation or following a suicide attempt [22,23]. This cohort represents a real-world clinical sample drawn from a tertiary mental health service and may not be representative of all suicidal young people in the community.

1.1. Background

The current study was undertaken to examine a cohort of young people admitted to a tertiary mental health service, which supports children and young people from birth to 25 years following a suicide attempt or with persistent suicidal ideation, in South East Melbourne. The post-suicide mental health program was one of four government-funded, assertive, outreach aftercare programs for young people aimed primarily at those aged 12 to 25 years, known as CY Hope (Child and Youth Hospital Outreach Post-Suicide Engagement). The CY Hope service is a clinical/non-clinical, psychosocial outreach program that aims to support children and young people at risk of a suicide attempt, along with their families. The CY Hope services were established in response to the 2021 Royal Commission into Victoria’s Mental Health System. This study examines data which were collected clinically for each admission at one of these services.

1.2. Aims

This project examined the data, via a file audit, of a cohort of young people attending the new post-suicide outreach initiative over 3 years from the commencement of the program in January 2022 until the end of 2024. The aim was to understand the demographics and characteristics of the young people entering the service to provide an understanding of the presence of potential risk factors. Specifically, the file audit looked for particular demographic and personal characteristics of the cohort based on the prior literature, and included gender identity, LGBTIQIA+ status, age, cultural characteristics, neurodivergence, employment or education status, and any mention of trauma, for example, sexual abuse. A further aim was to provide descriptive context regarding how the characteristics of this clinical cohort compare with those reported in the broader Australian population. We hypothesised that there would be significant differences between the socio-demographic characteristics of the cohort compared to the general population.

2. Method

2.1. Participants—Young People Attending CY Hope

Those included in the study were all young people who were accepted onto the CY Hope program during the study period. Young people with more complex psychiatric presentations requiring intensive medical management, such as psychosis or mania, or those expected to require longer-term specialist care, were referred to alternative services and were therefore not admitted into the CY Hope program. The program primarily accepts young people whose suicidality is considered to be associated with psychosocial stressors and who may benefit from short-term case management and psychosocial support. The study therefore focuses on a specific clinical sample of young people with active suicidal intentions presenting for care during the study period.

The total number of children and young people accessing the service during this time period was 171. Those who were referred to the CY Hope service but not admitted or who declined service were excluded. Consent for the data included in the file audit to be used for research purposes was obtained during admission. No one declined to consent for this purpose. Therefore, all data were included over the period of time detailed.

2.2. Population Comparisons

To examine the presenting cohort characteristics with those of a similar demographic within the Australian population, data was obtained from the most recent sources available; primarily the Australian Bureau of Statistics (ABS) [18], the Australian Institute for Health and Welfare (AIHW) [3], and a large study by Higgins et al. [24]. Both the ABS and AIHW are national government statistical agencies. Although population-level statistics were available for some of the demographic characteristics examined in this study, comparable data from similar real-world clinical cohorts of young people presenting with suicidality are limited. National population statistics were available for several demographic variables examined in this study. However, comparable population data were not available for some intersecting characteristics (e.g., neurodivergence and LGBTIQIA+ identity), which limited direct statistical comparison. As such, national population statistics were used to provide descriptive context for the characteristics observed in this clinical sample.

2.3. Procedure

A file audit was conducted on the records of all young people who accessed the CY Hope service from the commencement of the service in early 2022 until the end of 2024. The audit methodology was chosen for its utility in understanding real-world data and its contribution to ongoing service improvement [25]. All data examined in the study were routinely collected, and the research team determined the specific variables for inclusion in the file audit based on the prior literature. The following characteristics were included: age, assigned sex at birth, gender, First Nations identity, cultural diversity background, LGBTIQIA+ status, prior diagnosis of neurodivergence, employment and education status, and other potential sources of situational stress such as relationship, financial, and physical health concerns, as well as any historical issues, such as childhood trauma. As part of the admission process, participants provided written consent for their data to be used in an aggregate form for the purpose of service improvement and research. Once extracted from hospital files, the data was transferred onto a de-identified spreadsheet for reporting of demographic information and analysis of frequencies. Ethics approval was obtained from Alfred Hospital Human Research Ethics Committee, project number 794/24. As the study relied on retrospective clinical documentation recorded by multiple clinicians, the availability and accuracy of some variables depended on the quality of routine clinical records.

2.4. Data Analysis

To determine whether several of the demographic risk factors of the present sample were statistically higher or lower than that in the Australian population, a series of nonparametric one-sample z proportion tests were conducted (see [26]). To determine whether the proportions differed significantly between the sample and the Australian population, the Score (continuity corrected) confidence intervals and Score (continuity corrected) z proportion tests were chosen as they have shown to be more accurate when comparing proportions than equivalent Wald tests (see [27,28]).

3. Results

3.1. Participant Characteristics

Data was extracted for 171 children and young people. Ages ranged from 7 (one child aged seven years was an outlier in this initiative which was aimed at 12–25-year olds, but the tertiary service at the centre of the study treats age 0 (newborn infants) to 25 years inclusive) to 25 years (mean = 20, SD = 3.6 years). From the 171 participants, 59% identified as female, 39% as male, and 2% as nonbinary or not specified. Characteristics of the cohort are presented in

Table 1. Cohort and population characteristics.

Variable	Cohort %	Australian Population %	One-Sample z Proportion Tests (Two-Tailed)
			z	p	95% CI
LGBTIQIA+	47%	9.5% [14]	16.50	<0.001	0.392, 0.545
Transgender, gender diverse, and non-binary (TGDNB)	12.9%	2.3% [20]	8.96	<0.001	0.084, 0.190
Neurodivergent	24.6%	7.45% [14]	8.38	<0.001	0.185, 0.318
LGBTIQIA+ and neurodivergent	14%	Not known
Not employed, in education or training (NEET)	22%	12% (15–24 years) [14]	3.76	<0.001	0.159, 0.287
International student	10.5%	3.48% [2]	4.81	<0.001	0.065, 0.164
Aboriginal or Torres Strait Islander	4.1%	3.8% [2]	0.001	0.999	0.018, 0.086

.

The file audit also examined records for evidence of trauma; for example, physical or sexual abuse, childhood adversity, and family violence. The total number who met one or more trauma criteria was 18%. Other key aspects included a recent relationship breakdown (15%), physical health issue (8.5%), unemployment (11%), financial stress (30%), recent death (including recent suicide of a known person) (3.5%), and other family related issues which were not due to trauma (6%).

3.2. Comparison to the Australian Population

The results of the one-sample z proportion tests revealed that the proportion of neurodivergent individuals in the present sample was significantly higher than the proportion of those under 25 with a neurodivergence diagnosis in Australia. Tests also revealed that the proportion of LGBTIQIA+ people in the present sample was a significantly higher proportion than in the Australian population. Similarly, the proportion of those identifying as transgender, gender diverse, and non-binary (TGDNB) in the present sample was significantly higher than in the Australian population. The proportion of international students in the present sample was also significantly higher than the proportion of international students under 25 living in Australia. Tests also revealed that the proportion of young people in the present sample who were not engaged in employment, education or training was significantly higher than in the Australian population. There was no significant proportional difference between the present sample and the Australian population when considering Aboriginal and Torres Strait Islander status.

4. Discussion

The current study aimed to examine the socio-demographic characteristics and possible risk factors of young people who were part of a post-suicidal engagement program at a tertiary community child and adolescent mental health service in Melbourne, Victoria. We report on the patterns observed within a real-world tertiary service cohort rather than prevalence of characteristics among suicidal young people in the community. Consistent with previous research, females were over-represented in the cohort. However, the findings suggest that there were other key non-modifiable characteristics that were higher in this cohort, such as being non-heterosexual, TGDNB, and/or neurodivergent, compared to their peers in the Australian general population. Other characteristics of the cohort that appeared to be higher than in the general population included being an international student, and not being in employment, education or training (NEET). These findings align with previous research [13] which highlighted how individual and interpersonal factors such as gender, sexual orientation, and adverse childhood experiences (ACEs) can be risk factors for future suicidality. In the current study, the findings highlight that suicidal young people from some marginalised groups may be significantly more likely to present at mental health services and engage with post-suicidal support. These findings highlight characteristics that may be important considerations for youth mental health services when assessing and supporting young people presenting for care.

Young people who identified as non-heterosexual or transgender/gender diverse/non-binary (TGDNB) were significantly higher in this sample than in the general population. This finding aligns with prior literature, which suggests that young people who identify as non-heterosexual or TGDNB are three to four times more likely to have suicidal ideation and behaviour than the general population [29]. Furthermore, young people who are non-heterosexual tend to experience higher incidences of childhood trauma, sexual and physical abuse, and family victimisation compared to their heterosexual peers [24]. In addition, higher rates of substance misuse among non-heterosexual youth and continuing monosexism faced by bisexual and pansexual people may also contribute towards poor mental health [30]. Stigma, prejudice, and discrimination can also create a hostile social environment for non-heterosexual people that causes minority stress and consequent mental health problems [19]. With regard to TGDNB people, it has been suggested that global ongoing discrimination and government policy, which includes restrictions on access to gender-affirming healthcare, can lead to a greater sense of hopelessness, further perpetuating negative experiences leading to suicidality [18].

The prevalence of neurodivergence in the sample was also significantly higher than in the general population. This aligns with prior research showing that autistic youth are six times more likely to attempt suicide compared to the general population [31]. There may be several reasons for this. Several studies indicate an increased likelihood for autistic people to experience other mental health disorders, including being at an increased risk during adolescence and early adulthood [32,33]. Rigid thinking patterns, a propensity to impulsiveness, being victimised and difficulties in developing relationships and support networks, may also predispose these young people to suicidality [24], as may the effects of social isolation and disruption caused by the coronavirus pandemic in the preceding years. Young adults with autism can also experience higher instances of internalising problems due to bullying and factors such as social isolation, which can increase suicide risk [34]. Similarly, masking behaviours, sensory overload, and consequent autistic burnout may also contribute [35]. Some research also suggests that differences in emotional communication and expression among some autistic people, particularly within clinical contexts that rely on neurotypical communication norms, may contribute to difficulties recognising or assessing mental health concerns prior to suicidal crises [31]. Additionally, those with ADHD may experience challenges with emotional regulation and increased impulsivity that may increase suicidal behaviour [36].

The study showed that 14% of neurodivergent young people in this sample also identified as LGBTIQIA+. There are currently no Australian population figures to draw upon as a comparison; however, a recent meta-analysis by Kallitsounaki and Williams [37] indicted that ASD diagnosis among transgender people was 11%, although the review was not youth specific. Autistic and TGDNB populations can face similar social barriers, such as discrimination, compared to the general population, which can lead to higher rates of mental health issues [38]. Future research looking at this cohort may help to build further understanding about the intersections between neurodiversity, sexuality, gender, and suicidality.

In the current study, 22% of young people were not employed in education or training, which is significantly higher than general population figures [39]. In addition, 30% of the cohort indicated experiencing financial stress. Prolonged economic inactivity for young people, such as lack of education and employment, can increase the odds of a suicide attempt by 3.6 or 260% [17]. Young people not engaged in education or employment will likely experience greater financial hardship, which can also negatively impact their mental health and lead to increases in suicidal ideation [40]. These findings highlight the importance of considering educational and employment disengagement for young people as these are factors which may reflect broader social and economic vulnerabilities.

A further characteristic in the sample was the high proportion of international students. International students represented 11% of the total in the current study compared to 3.35% of all students studying in Australia at the time [41]. International students can face unique risks for suicidality due to discrimination, social isolation, economic pressure, and cultural alienation [41]. Limited research exists on effective suicide prevention for international students in Australia [42]. Many lack access to appropriate healthcare, including psychological support, and suffer financial stressors and stigma, which may prevent them from accessing mental health services [42,43]. Future research looking at international students, including the unique factors that play a role in their suicidality and/or mental health more generally, may help to elucidate the problem and provide greater insight into how best to support this cohort.

Young people identifying as Aboriginal or Torres Strait Islanders were not over-represented in this sample; however, it is not clear whether this represents an actual lower rate of suicidality or lower rates of help-seeking due to factors which may include historical and institutional racism. The lower representation of Aboriginal and Torres Strait Islander young people in the present sample may also reflect the urban location of the service, where the overall number of Indigenous presentations was relatively small. However, this reflects other published literature which suggests that despite being more likely to experience suicidality, only a minority of Aboriginal youth end up receiving treatment [44].

4.1. Limitations

This study has several limitations that suggest clear directions for future research. First, referral volumes and reasons for non-admission were not systematically captured in the clinical information system. As a result, the number of young people referred to the service but not accepted into the program due to complexity (e.g., requiring more than 12 weeks of case management), severity (requiring hospitalisation), living outside the catchment zone, or being outside the recommended age range (12–25 years) is not known. The study therefore represents a selected clinical cohort of young people accepted into the service rather than all young people with suicidality who were initially referred for care. Thus, when comparing the current cohort with the wider Australian population, we can only compare those who seek support for suicidality and who do not require more intensive mental health support. Future work in this area should make attempts at capturing a wider range of referrals and outcomes across all referral pathways, which may potentially better comparisons with population figures.

Another important limitation is that as this is a single-site tertiary clinical cohort, the findings cannot be generalised to all Australian young people experiencing suicidality. Differences between this cohort and national population averages may reflect referral pathways, admission criteria, and service selection processes in addition to suicidality-related vulnerability. Despite this limitation, the present results support prior research about potential risk factors for suicidal behaviour. Future research involving multi-site studies and comparison groups drawn from community, emergency department, or primary care settings may help to determine whether the patterns observed are specific to this service or reflect broader trends.

The study also utilised a retrospective file audit based on routinely collected clinical data recorded by multiple clinicians. Despite the strength of using real-world clinical data, the availability and accuracy of some variables depended on the quality and completeness of clinical documentation, which may introduce information bias.

It is also important to consider that data collection occurred at the end of the COVID-19 pandemic, during which Melbourne experienced extended lockdown periods. These circumstances may have influenced patterns of mental health presentations and service utilisation among young people; however, this was not examined within the scope of the present study, since the service commenced in 2022, and therefore no pre-pandemic service data were available for comparison.

Finally, comparisons with national population statistics were intended to provide approximate comparisons rather than establish population-level risk factors. The absence of comparable real-world clinical cohorts of young people presenting with suicidality limits the ability to directly compare these findings with similar service-based populations.

Future studies should include multi-site designs, prospective recruitment, and systematic recording of referral and admission outcomes. The inclusion of more comparable clinical or community cohorts would improve interpretation of demographic differences. In addition, longitudinal follow-up may help clarify which subgroups benefit most from post-suicide aftercare programs.

4.2. Implications

The findings of this cohort study highlight the complexity of underlying issues that impact the mental health of young people accessing the CY Hope service. Rather than a single factor, many young people presented with a cluster of overlapping vulnerabilities (e.g., neurodivergent, non-heterosexual, and financial stress). Consistent with previous research, higher proportions of non-heterosexual, TGDNB, and neurodivergent youth were observed in this clinical cohort presenting with suicidality compared to the general population. Financial stress, relationship stress, and previous trauma were also commonly reported within the cohort. International students were also over-represented, possibly because they face additional barriers to accessing care, while Aboriginal and Torres Strait Islanders were not over-represented, possibly for the same reason. These patterns highlight the importance of considering intersecting social and demographic vulnerabilities when assessing young people presenting with suicidality in clinical settings.

These findings have implications for mental health services that manage the care of young people. Many of those presenting at the service had multiple social and demographic factors which were associated with vulnerability to suicidality, particularly those that placed them in a minority population. Clinicians working with suicidal young people should seek to adapt assessments and safety planning to account for factors such as minority stress, sensory needs, and communication preferences which are present in neurodivergent populations. For example, the high amount of neurodiversity in the cohort, as well as in prior studies, indicate that improving neuro-affirming approaches could be an important way in which services can adapt care. Similarly, the increased proportion of non-heterosexual young people experiencing suicidality [29] suggest support services should be respectful and safe spaces for those from across the spectrum of sexualities. These findings highlight the importance of services recognising intersecting social and demographic vulnerabilities when working with young people.

The high level of reported trauma indicates the need for embedded trauma-informed care in clinical practice. The findings also have implications for service design and delivery, such as a need for targeted outreach for over-represented young people as outlined in this study. Additional steps could include partnerships with, for example, LGBTIQIA+ community organisations, youth housing, and education and employment services to ensure wrap-around support for these young people.

Given disengagement from education and employment were common in this study, prevention efforts could include school- and community-based programs that support engagement with education and vocational pathways. Public health messaging should continue to address stigma and promote help-seeking for young people in these minority groups.

Understanding causal mechanisms should be a focus for future research; in particular, how each of the key factors intersects with suicide and suicidal ideation. A longitudinal follow-up could potentially clarify which subgroups benefit the most from outreach programs and which factors predict re-presentation, recovery or disengagement. A multi-site approach would offer greater generalisability.

5. Conclusions

This study provides important insights into the socio-demographic and clinical characteristics of the young people presenting at a tertiary post-suicide outreach service. The findings highlight the over-representation of young people from minority groups, e.g., non-heterosexual, TGDNB, NEET, and international students. Many young people also reported additional stressors such as financial stress, relationship difficulties, and trauma which highlighted the complex, layered challenges associated with suicidality in this cohort. These findings reinforce growing evidence suggesting that social marginalisation, stress, and unmet support may contribute to vulnerability among young people experiencing suicidality. Although Aboriginal and Torres Strait Islander people were not over-represented in this cohort, this finding needs to be interpreted with caution and continued efforts to embed culturally grounded practice should remain at the forefront of youth suicide intervention programs.

Overall, the findings reinforce that suicide risk in young people is multidimensional. As a psychosocial intervention, it is critical that suicide aftercare services identify and tailor their approaches to the social and demographic factors identified in the study. Tailoring services to address these intersecting needs will continue to be essential in the future to reduce youth suicide and improve long-term mental health outcomes.