Suicide prevention: a public health priority

Suicide prevention: a public health priority

In 2013, the 66th World Health Assembly adopted the first Mental Health Action Plan (2013–2020) of the World Health Organization (WHO) (WHO 2013). In its subsequent report, Preventing suicide: a global imperative (WHO 2014), the WHO provided actionable steps for the implementation of effective national suicide prevention strategies. Key components were the strengthening of suicide surveillance by improving the quality and timeliness of national data on deaths by suicide and suicide attempts, and establishing an integrated data collection system to help identify specific groups, vulnerable individuals and high-risk situations.

The policy context for suicide monitoring in Australia

Suicide has long been a significant health issue in Australia. More recently, there has been an increased emphasis on suicide prevention by Australian governments. In 2017, actions to address suicide as a priority area were included in the Fifth National Mental Health and Suicide Prevention Plan (COAG Health Council 2017). This plan committed all Australian governments to a collaborative national approach to mental health planning and service delivery, including improving the quality and timeliness of data collection on suicide; suicide attempts; and intentional self-harm in Australia. These data would provide much needed information to those responsible for the planning, funding, delivery and evaluation of suicide prevention strategies.

Suicide prevention in Australia is a complex area of policy with interconnected responsibilities. Governments, policy makers and service providers all have a role in reducing deaths by suicide as well as cases of intentional self-harm. The reasons for suicide are often complex and different for each individual. Research has shown that a range of factors are commonly present in the histories of those who died by suicide, including mental and behavioural disorders; physical illness; and psychosocial factors (including alcohol and/or other drug problems; relationship or legal issues; bereavement; impacts of chronic health conditions; disability; unemployment; homelessness; and bullying) (ABS 2019b; Clapperton et al. 2019). Therefore, effective suicide prevention requires action from a correspondingly broad range of government agencies, including those responsible for health, education, employment, urban planning, welfare and law enforcement agencies.

In acknowledgement of the devastating effects of suicide and the pivotal role governments have to play in addressing it, the Australian Government has made suicide prevention a national whole-of-government priority, and indicated a commitment to the aspirational goal of working ‘Towards Zero’ deaths by suicide (Department of Health 2019c). To this end, the first National Suicide Prevention Adviser reporting directly to the Prime Minister has been appointed and a National Suicide Prevention Taskforce has been established to coordinate collaboration between government agencies and across different levels of government.

The AIHW is currently actively involved in data improvement activities to expand the collection or availability of data on deaths by suicide and on the occurrence of intentional self-harm in 2 specific populations: Indigenous Australians and current serving, reserve and contemporary ex-serving ADF personnel. Rates of suicide among Indigenous Australians are higher than those of the non-Indigenous population. For the 5 years from 2014 to 2018, the age-standardised suicide rate for Indigenous Australians was almost twice that of the non-Indigenous population (23.7 vs 12.3 per 100,000 population) (ABS 2019a). In the 2019–20 Budget, the Australian Government committed a further $15 million to suicide prevention for Indigenous Australians (Department of Health 2019b). Also, on 5 February 2020, the Australian Government announced the appointment of a National Commissioner for Defence and Veteran Suicide Prevention. The AIHW and the Australian Commission on Safety and Quality in Health Care, along with coronial and legal experts, will provide technical expertise to support the Commissioner’s work.

Improving the evidence base

In Australia, the extent of intentional self-harm and suicidal behaviours in the broader community is largely unknown, as those presenting to emergency departments (EDs) or primary health care—or not seeking treatment—are not captured by clinical data. The ‘iceberg’ model has been used to represent the relative incidence of suicide and intentional self-harm and the difficulty of monitoring the incidence of suicide, intentional self-harm and suicidal behaviours (Arensman et al. 2017; Geulayov et al. 2018; McMahon et al. 2014; Pollock et al. 2018). In this model (Figure 9.1), the extent to which suicide, intentional self-harm and suicidal ideation are currently captured by administrative data sets can be represented as an ‘iceberg’ for which only the tip—representing suicide or intentional self-harm that results in hospital admission—is visible. Intentional self-harm that results in presentation to other health services (such as EDs, primary health care or ambulance services) or does not result in medical treatment are more common but largely hidden from view because they cannot be identified with clinical data—thus forming the ‘submerged’ part of the iceberg.

Figure 9.1: Iceberg model illustrating the extent to which suicide, intentional self-harm and suicidal behaviours are currently captured by clinical data in Australia

This figure is a representation of the extent to which national data are available to monitor the incidence of suicide, intentional self-harm and suicidal behaviours in Australia. This is often compared to an iceberg, where only the tip (representing deaths by suicide and cases of intentional self-harm admitted to hospitals) is visible, while the majority of self-harming and suicidal behaviours remain hidden under the surface and cannot currently be identified by clinical data.

Note: Adapted from: Pollock et al. 2018.

Finally, suicidal behaviours, such as making a suicide plan or having suicidal thoughts, are even more common; however, information about these behaviours may not be captured by clinical data. Instead, an indication of the prevalence of these behaviours in the community may be derived from surveys of representative samples of the population.

In recognition of the fact that data are critical to the development of effective suicide and intentional self-harm prevention policies and services, the National Suicide and Self-harm Monitoring System was announced as a component of the Australian Government’s Prioritising Mental Health Package (Department of Health 2019a). The aim of the monitoring system is to collate and coordinate data and information on suicide, intentional self-harm and suicidal behaviours in Australia to improve their coherence, accessibility, quality and timeliness. The AIHW has been funded $5 million per year for 3 years (2019–20 to 2021–22) to deliver the monitoring system. Data improvement activities to enhance the comprehensiveness of data, and the creation of a monitoring system to support the accessibility and useability of data for stakeholders, will make it a key resource to assist governments, services and communities to improve suicide and intentional self-harm prevention strategies. The National Suicide and Self-harm Monitoring System will draw on the expertise of the National Injury Surveillance Unit (NISU) at Flinders University, a collaborating unit of the AIHW, and on other subject matter experts as required.

Currently available national sources of data on suicide, intentional self-harm and suicidal behaviours include mortality data; data on the provision of hospital services; and population-based mental health surveys. These sources, and potential new sources of data to fill gaps in our understanding of suicide, intentional self-harm and suicidal behaviours in Australia, are summarised in Table 9.1 at the end of the chapter. How the data are obtained, and the limitations of each data set, are also discussed below. Examples of initiatives under the National Suicide and Self-harm Monitoring system project are also described, including those that aim to make greater use of existing data sources to identify populations at risk and to allow more timely, localised responses.

Deaths by suicide

The collation of national data on deaths by suicide in Australia requires the collaboration of multiple state and national government bodies (Box 9.1). Despite the fact that Australia has strong systems in place for the collection of death statistics, accurate reporting of deaths by suicide is particularly challenging for a number of reasons.

Generally, deaths due to external causes, including suspected suicides, are referred to a coroner for investigation of the cause and, if applicable, the intention of the deceased. Some deaths by suicide may be misclassified as ‘accidental’ or ‘undetermined’ due to the difficulty in determining the true intent of the deceased and, as a result, may lead to under-reporting (Senate Community Affairs References Committee 2010). Additionally, the international medical coding system used to classify causes of death (the WHO International Statistical Classification of Diseases and Related Health Problems, 10th revision: ICD-10), does not distinguish between suicidal and non-suicidal intent (WHO 2016). While a decision has not yet been made on whether or when ICD-11 will be adopted for coding deaths in Australia, the new classification developed by the WHO has made provisions for a new dimension to capture the ‘intended result’ (suicidal intent) of intentional self-harm (intent pending; suicidal/non-suicidal intentional self-harm) (WHO 2019).

The quality of cause of death coding can be affected by the length of time required for coronial processes to be finalised (ABS 2019a). To improve the quality of ICD coding, all coroner-certified deaths are now subject to a revisions process (Box 9.1) (ABS 2019a).

National data can mask regional variation due to significant demographic, economic or cultural differences between different regions of Australia. Therefore, more granular, regional-level mortality data can be useful in developing and monitoring local suicide prevention strategies. However, a major challenge with suicide mortality data is that suicide deaths are statistically rare events, meaning that it is difficult to achieve the statistical power that is necessary to identify patterns and causation, or to draw conclusions about reductions in the suicide rate (AHA 2014; AIHW 2011; Morris et al. 2018). These methodological challenges exist at the national level, where around 3,000 deaths per year are suicides, but for the reporting of suicide deaths at lower regional levels (for example, state/territory, local government area) or by demographic variables (for example, age and sex) these issues are compounded, as data are broken down into even smaller population groups. To date, issues relating to data volatility and robustness for measurements of rates and trends have had limited statistical analysis.

Box 9.1: Suicide mortality data

The registration of deaths in Australia is required by law and is the responsibility of state and territory Registries of Births, Deaths and Marriages. As part of the registration process, information about the cause of death is supplied by the medical practitioner certifying the death, or by a coroner. The information is provided to the Australian Bureau of Statistics (ABS) for coding of causes of death (according to the WHO International Statistical Classification of Diseases and Related Health Problems, 10th revision; ICD-10) and compilation into aggregate statistics on an annual basis (ABS 2019a). In addition, the ABS supplements these data with information from the National Coronial Information System (NCIS), a national database of coronial findings, post-mortem, toxicology and police reports (ABS 2019a). These data are sent to the Australian Coordinating Registry (ACR) who acts on behalf of all the Registries to coordinate release of the data. The AIHW receives the data from the ACR and maintains these data in the National Mortality Database (NMD), a historical (1964–2018), coherent and accurate database for analysis, linkage and reporting purposes (AIHW 2019a).

Deaths that are referred to a coroner can take time to be finalised and the coroner’s case closed. To account for this, the ABS undertakes a revisions process for those deaths where coronial investigations remained open at the time an initial cause of death was assigned. Usually, data are deemed preliminary when first published, revised when published the following year and final when published after a second year (ABS 2019a).

Problems with the reporting of small numbers in population groups—such as regional areas (AIHW 2019d), certain demographic characteristics (AIHW 2019c), or in specific populations including Indigenous Australians (AIHW: Kreisfeld & Harrison 2020) or current serving, reserve and contemporary ex-serving ADF personnel (AIHW 2019f)—have been avoided by aggregating multiple years of data to ensure confidentialisation and privacy.

State-based suicide registers

While the quality of Australian mortality data is high by world standards and historical data are available, the annually available mortality data sets have limited usefulness in informing time-sensitive responses. Delays between a death by suicide and its reporting to policy makers and service providers can be an impediment to the early detection of systematic trends and appropriate intervention responses aimed at preventing further suicides.

Coronial suicide registers have been established by some jurisdictions in Australia (including Victoria, Queensland, Western Australia and Tasmania). Information from these registers can be used to assist coroners formulate evidence-based recommendations to prevent suicide and may also be shared with local governments and service providers to better target and inform suicide prevention activities (Leske et al. 2019; Sutherland et al. 2018; Tasmanian Department of Health and Human Services 2016).

These registers have the potential to provide timely data on deaths suspected to have been by suicide and may be useful for identifying trends in locations or in the methods used for suicide. For example, the Victorian suicide register enables basic information on a suspected death by suicide (including cause of death, location of death, usual place of residence, age, sex and occupational information) to be coded within 24–48 hours of the coroner being notified of a death suspected to be by suicide (Sutherland et al. 2018). More comprehensive contextual information about modifiable risk factors (for example, the deceased’s physical and mental health history, interpersonal stressors, psychosocial factors, and other circumstances surrounding their death) are also collected from a variety of sources, such as toxicology autopsy and police reports; however, this information can take longer to be made available. The establishment of such registers does not replace the collection of nationally consistent deaths data but may enhance the development of timely interventions and appropriate localised suicide prevention strategies.

Embedding psychosocial risk factors in national mortality data

Making greater use of existing data sets removes duplication of effort, reduces costs and minimises reporting burden. In 2019, the ABS published results of a pilot study to enhance the national Causes of Death data set, by coding psychosocial risk factors for all coroner-referred deaths (including deaths by suicide) registered in 2017, via a comprehensive manual review of reports included in the NCIS (ABS 2019b). Psychosocial factors (for example, a past history of self-harm; relationship problems; legal issues; bereavement; unemployment; homelessness; and disability) were identified in 63% of all deaths by suicide. The findings of this pilot study were limited by the amount of information that was voluntarily captured in post-mortem, toxicology or police reports and coronial findings included in the NCIS. However, there is potential to enhance what the NCIS captures by requiring the recording of risk-factor information in suspected cases of suicide. The ABS is currently working to embed risk factors into the national mortality data set to provide comprehensive information on the combination of factors contributing to deaths by suicide. This data initiative is being funded as part of the Suicide and Self-harm Monitoring System project to enable monitoring of emerging trends and improve evaluation of the effectiveness of suicide prevention strategies.

Data linkage with mortality data

Data integration can enrich the value and maximise the use, and re-use, of nationally collected information, while preserving individual privacy and the security of sensitive data. It also has the potential to expand the evidence base to better support research and policy development. By combining information from existing surveys, administrative data collections and censuses, a more complete picture of the circumstances of individuals and households can emerge. Integrated data sets can also be combined with additional point-in-time and/or longitudinal information to help assess the effectiveness of policies and programs. A pilot project has shown that data integration has the potential to provide valuable insights into contextual factors (for example, employment and marital status) associated with deaths, including those by suicide (ABS 2016c). This project combined national mortality data with data from the 2011 Census through a process of probabilistic record linkage.

The integration of multiple, cross-agency data sets for use as a resource has evolved more recently. The Multi-Agency Data Integration Project (MADIP) is a cross-agency partnership between the Departments of Social Services, Health, Human Services, Australian Taxation Office and the ABS. Its purpose is to create enduring, linked, research data sets. Improvements in linkage and the ability to combine and repurpose data will provide improved measurement of outcomes for population groups of interest as well as richer statistics (ABS 2016b). As part of the Suicide and Self-harm Monitoring System project interrogation of MADIP data will be used to better understand the social determinants of death by suicide, such as educational attainment and housing tenure, and to identify population groups at increased risk.

The AIHW has also been working with the Department of Health along with state and territory health authorities to develop the National Integrated Health Services Information Analysis Asset (NIHSI AA) which includes mortality data together with hospital admissions, Medicare Benefits Schedule (MBS), Pharmaceutical Benefits Scheme (PBS) and residential aged care data. This analysis asset will enable examination of service-use patterns and the demographic profiles of those using (and, by inference, those not using) services. The AIHW will analyse the NIHSI AA to report on the service use of people in their last 12 months of life, including those who died by suicide. The potential insights from this project and analysis of other integrated data assets will greatly enhance our understanding of people-centred service use and modifiable risk factors for suicide.

Suicide attempts and intentional self-harm

The most common predictor of death by suicide or premature mortality of any kind, including accidental drug overdose, is a personal history of a previous suicide attempt or act of self-harm (ABS 2019b; Carr et al. 2017). As a consequence, monitoring of the incidence, demographic patterns and methods used in suicide attempts or instances of self-harm have the potential to improve the development of suicide prevention strategies.

Hospitalisations for intentional self-harm

Currently in Australia, the national source of intentional self-harm data is the AIHW National Hospital Morbidity Database (NHMD), a compilation of administrative data supplied by state and territory health authorities for patients admitted to public and private hospitals (AIHW 2019e). The NHMD includes demographic, length of stay and diagnosis data including external causes of injury and poisoning, as well as the procedures patients underwent in hospital, the place where the injury occurred and the type of activity being undertaken by the person when injured (AIHW 2019b). Diagnosis, intervention and external cause data are reported to the NHMD by all states and territories using the International statistical classification of diseases and related health problems, 10th revision, Australian modification (ICD-10-AM) and the Australian Classification of Health Interventions (ACHI). Information from this database can be useful to monitor trends in intentional self-harm over time and to provide, albeit limited, insights into those at risk of further self-harm. Indigenous status is included in the NHMD (AIHW 2019b) and the AIHW is investigating the feasibility of several options on reporting intentional self-harm in ADF personnel.

The ICD-10 AM coding system does not distinguish between suicidal and non-suicidal intentional self-harm (AIHW: Harrison & Henley 2014). However, suicidal ideation in the absence of a mental health condition is captured and assigned using ICD-10-AM (ACCD 2018). Although a decision has not yet been made on whether or when ICD-11 will be adopted for coding hospital admitted cases in Australia, the new classification developed by the WHO has made provisions for a new dimension to capture the suicidal/non-suicidal intention of intentional self-harm injuries (WHO 2019).

Hospitals data on intentional self-harm can only provide a partial picture of those self-harming (Figure 9.1). Other sources of data on intentional self-harm and suicidal behaviours, such as ED data, ambulance and police attendances, crisis line calls and treatment provided by mental health or primary health care services, have the potential to provide a more complete picture of these behaviours in Australia and identify opportunities for improved intervention or postvention.

Hospital emergency department data

The hospital ED is often the first point of contact with the health system for people who have harmed themselves or who have suicidal thoughts (Perera et al. 2018). State and territory health authorities provide ED data to the AIHW for collation into the National Non-admitted Patient Emergency Department Care Database (NNAPEDCD) (AIHW 2019g). The NNAPEDCD captures information on the patient’s principal diagnosis (the diagnosis mainly responsible for the attendance) and coexisting additional diagnoses. Unlike the NHMD, information on the external cause of injury, the place of occurrence or the type of activity are not captured in the NNAPEDCD (AIHW 2019g). This means presentations to the ED relating to suicide attempts or intentional self-harm cannot be identified in the current national ED data collection. Also, this data set may not capture the complexity of mental health presentations to the ED (AIHW 2019g). Therefore, information on suicidal behaviours is not coded or may only be captured in clinical notes. A technical report on the use of ED data to improve the routine surveillance of all types of injuries concluded the utility of the data source would be enhanced by including external cause data in the NNAPEDCD (AIHW: Henley & Harrison 2018).

In recognition of the key need for better data around suicide attempts and self-harm in ED data, the Australian Health Ministers’ Advisory Council has funded the Mental Health Information Strategy Standing Committee to undertake 2 projects in 2019–20 to support improvements in identification of suicide and self-harm-related presentations to EDs. These include the development of a methodology to identify presentations relating to suicide attempts within jurisdictional data and a scoping paper outlining opportunities and barriers to developing a nationally consistent data collection on suicide-related ED presentations, including recommendations for data improvements. The AIHW has been funded to develop the scoping paper, which will require engagement with a range of stakeholders, including the Mental Health Information Strategy Standing Committee and Emergency Department data custodians.

Police and ambulance attendance data

State- and territory-held police and ambulance attendance data may provide insights into self-harming and suicidal behaviours in Australia at a stage when intervention to prevent further harm or subsequent suicide may be possible. Currently, the clinical coding of ambulance presentations data is limited, with data collected for some states over a short period of time each year (Turning Point 2019) while police reports are not standardised across Australia making national analysis difficult.

A study by Turning Point and Monash University coded and analysed clinical patient records from ambulance call-outs to men presenting with acute mental health issues, intentional self-harm, suicidal behaviours, or alcohol or drug intoxication in the Australian Capital Territory, the Northern Territory, Queensland, Tasmania, Victoria and New South Wales between July 2015 and June 2016 (Turning Point 2019). The study found that coded ambulance data are an important source of information to

establish the number and characteristics of presentations for mental illness, intentional self-harm or suicidal behaviours—and may address evidence gaps by capturing information that is not currently identified by other health morbidity data sets such as ED or hospital admissions, including:

  • details of the nature and background to the attendance, including information about what was observed ‘on scene’ (such as bystander accounts, evidence of drug use and suicide intent)
  • types of people most at risk (for example, those who call an ambulance multiple times or those with increasingly harmful behaviour)
  • the location of the event (which may allow for geographic and temporal mapping)
  • the clinical outcome.

Following this initial study, the National Suicide and Self-harm Monitoring System will include the collection, improvement and dissemination of national coded ambulance data for intentional self-harm and suicidal behaviours, as well as:

  • alcohol and other drug-related ambulance attendances, including intentional alcohol and other drug poisoning and type of drug involved
  • mental health-related ambulance attendances, including the presence of symptoms, history of mental illness and risk indicators both at the time of presentation and through the life course.

This data set will address a significant gap in service-level data for populations at risk of suicide or intentional self-harm. Such a data set may also provide opportunities for data linkage to allow insights, for example, into service use patterns or cohort analysis.

Community health care data

Community health care data on intentional self-harm and suicidal behaviours are limited. Self-harming and suicidal behaviours occurring in the community may be treated by general practitioners or community and residential mental health services; however, data collections from these sources do not routinely capture this information.

Also, although MBS, PBS or Repatriation Schedule of Pharmaceutical Benefits (RPBS) data can provide information about medical services provided or prescriptions processed, by themselves these sources cannot provide information about those at risk of suicide, intentional self-harm or suicidal behaviours. Nevertheless, these sources of data may be useful in terms of adding to data linkage projects.

Suicidal ideation

Measuring the incidence of suicidal ideation with routinely collected clinical data is also limited because the majority of people with suicidal thoughts do not tend to seek medical treatment (Geulayov et al. 2018). Self-reports of suicide attempts, plans or thoughts in health surveys of representative samples of the population provide information about suicidal behaviours in the community. On the basis of survey data, suicidal behaviours are far more common than deaths by suicide or intentional self-harm (Slade et al. 2009). People who experience suicidal ideation and make suicide plans are at increased risk of suicide attempts, and people who experience all forms of suicidal thoughts and behaviours are at greater risk of death by suicide (Slade et al. 2009).

Survey data

The National Survey of Mental Health and Wellbeing (2007) indicated that, at some point in their lives, 13% of Australians aged 16–85 years had had serious thoughts about taking their own life (an estimated 2.1 million Australians), 4% (over 600,000) made a suicide plan and 3% (over 500,000) had attempted suicide (Slade et al. 2009). In 2007, the number of registered deaths by suicide in Australia was 2,229 (ABS 2016a).

The second Australian Child and Adolescent Survey of Mental Health and Wellbeing, conducted between 2013 and 2014, captured information from Australian young people aged 12–17 years about self-harming activity and suicidal behaviours (Lawrence et al. 2015). Around 1 in 10 surveyed 12–17-year-olds (10.9%, equivalent to an estimated 186,000 young people) reported having ever self-harmed and about three quarters of these had harmed themselves in the previous 12 months (8%, equivalent to an estimated 137,000 young people). Around 1 in 13 (7.5%, equivalent to an estimated 128,000 young people) 12–17-year-olds had seriously considered attempting suicide in the previous 12 months and of these, one third (or 2.4% of all 12–17-year-olds) reported having attempted suicide in the previous 12 months. Both of these national surveys relied on self-reported responses, and therefore should be interpreted with caution, as respondents may not feel comfortable reporting on intentional self-harm or suicidal behaviours.

In 2019, the Government announced funding for an Intergenerational Health and Mental Health Study (Hunt 2019). This study will include components on general and mental health, including lived experiences of suicide and related services, and will provide updated results to compare with the 1997 (ABS 1998) and 2007 National Survey of Mental Health and Wellbeing (Slade et al. 2009).

Crisis line calls and help-seeking websites

Several organisations in Australia provide tele-counselling for people in crisis, for example, Lifeline, MensLine Australia, Kids Helpline, Beyond Blue and the Suicide Call Back Service. Each helpline has its own data capture system; however, data are not standardised and governance agreements are not currently in place to allow de-identified data to be shared and analysed. Better use of these data may provide useful information on help-seeking behaviours and identify populations at risk of suicide, intentional self-harm and suicidal behaviours.