The challenge of suicide prevention in Ireland

One of the primary challenges for our health services is to identify effective means of preventing suicide. The World Health Organization (WHO, 2010) has estimated that ~1 million people complete suicide (i.e. end their own life) each year. Global suicide rates have been increasing over the last number of years across age and gender demographics (Tarrier et al. Reference Tarrier, Taylor and Gooding2008). Likewise, there was a marked increase in rates of suicide between 2006 and 2011 in Ireland (Kelly & Doherty, Reference Kelly and Doherty2013). In 2009, suicide figures in Ireland reached their highest level at 552 deaths, with males represented disproportionately, and the 2011 figures indicate 525 deaths by suicide in Ireland, with males again over represented (National Suicide Research Foundation, 2013). Ireland also holds the fifth highest rate of suicide in Europe amongst males aged between 15 and 29 in 2013 (MacSharry, Reference MacSharry2013). Young females also represent an at-risk group. Scowcroft (Reference Scowcroft2013) indicates an increase in suicide rates for Irish females between 10–14 years and 15–19 years age groups in 2011.

Along with gender and age, a number of other cohorts have been identified as being at risk of suicide. These include those with mental ill-health; those in professional occupations; those in prison [Department of Health (DoH) 2012]; rural dwellers (Statistics Canada, 2002; Hill et al. Reference Hill, Pritchard, Laugharne and Gunnell2005; Hirsch, Reference Hirsch2006); those with physical ill-health; and those undergoing stressful life events (e.g. unemployment, marital separation: WHO, 2012). Mental ill-health is estimated to be present in approximately one in four individuals, with suicide behaviour being strongly correlated with presentations such as depression, anxiety, personality disorder, and bipolar disorder (Tarrier et al. Reference Tarrier, Taylor and Gooding2008).

The WHO (2012) point to a variety of barriers to accessing intervention services for at risk individuals including: stigma; lack of detection/referral of risk by GPs; and difficulty accessing secondary care mental health services, at least in part due to excessively high thresholds for acceptance of referrals. Moreover, findings consistently indicate that there are higher rates of contact by suicide completers in primary care services (Luoma et al. Reference Luoma, Martin and Pearson2002; Owens et al. Reference Owens, Loyd and Campbell2004). A potential strategy to reduce suicide would be to reduce stigma; improve access to services and referral onto higher intensity services; and improve risk management collaboration between frontline and secondary care mental health services.

The continuum of suicide behaviour

Suicide behaviour can be conceptualised as a continuum of either internal or external acts that ranges from suicidal ideation; to serious thoughts of death; to suicide plans; to suicide attempts; and through to suicide completion. ‘Suicidal ideation’ is defined as ‘suicidal ideas in the general population’ (Casey et al. Reference Casey, Dunn, Kelly, Birkbeck, Dalgad, Lehtinen, Britta, Ayuso-Mateos and Dowrick2006: 410). Those with ‘serious thoughts of death’ are defined as those who have ‘thought seriously about trying to kill themselves at any time in the past 12 months’ [Centers for Disease Control and Prevention (CDCP) 2011: 4]. Those with ‘suicide plans’ are defined as persons who have made plans to take their own life in the past year and ‘suicide attempts’ pertain to any incident where a person has taken steps to end their life in the past year but has not completed suicide (CDCP, 2011). In contrast, deliberate self-harm (DSH) refers to purposeful acts of harming oneself, including, but not limited to, self-mutilation, cutting and burning. While such behaviours may have multiple functions, they often serve as a means to regulate affect (e.g. to express emotion and to achieve a sense of control over it; Suyemoto, Reference Suyemoto1998).

What is APSI?

Given the significant unmet clinical needs of adults with mental health presentations in primary care (Tedstone-Doherty et al. Reference Tedstone-Doherty, Moran and Kartalova-O’Doherty2008), and in the context of increasing budgetary constraints, it was necessary to develop a high throughput but high quality primary care service. Funded by the heretofore Office of the Assistant National Director for Mental Health Services, APSI was developed to provide stepped care low-intensity, high-throughput interventions in a primary care setting for adults presenting with mild-to-moderate mental health presentations. The stepped care approach utilised by APSI aims to provide a central referral point for immediate risk assessment; a continuum of interventions congruent with the continuum of suicide behaviour; and collaborative working with secondary care and other primary care services.

Building upon a 3-year pilot of a predominantly one-to-one cognitive behaviour therapy (CBT)-based service in a rural town, APSI was piloted in October 2012 and now provides services in six primary care team areas across one particular county. It is staffed by six Mental Health Practitioners, each assigned to a particular primary care team area, who to date have been graduate psychologists. In addition to an in-house training programme of clinical workshops that focus on how to work with common mental health presentations, the practitioners are trained in Applied Suicide Intervention Skills Training; and/or Skills-Based Training on Risk identification and Management (STORM); and/or cross-care suicide prevention for the travelling community. While managed by a Principal Psychologist Manager, the practitioners receive weekly clinical supervision from a Senior and a Staff grade Clinical Psychologist.

As agreed with the local secondary care mental health services, Table 1 indicates the inclusion criteria for primary and secondary care mental health services, respectively.

Table 1 Inclusion criteria for APSI and secondary care services

^a The Counselling in Primary Care (CIPC) service also provides one-to-one intervention to General Medical Service (GMS)-only clients.

^b Moderate presentations may be initially referred to Access to Psychological Services Ireland (APSI). However, if such presentations have not responded to low intensity APSI interventions (e.g. guided self-help, brief cognitive behaviour therapy), or have a level of complexity that suggests APSI interventions may be inappropriate, referral should be made to secondary care services.

As evident from Table 1, APSI may also accept referrals for service users whose on-going treatment in secondary care requires an additional time-limited intervention for a mild-to-moderate mental health presentation(s).

APSI’s model for risk referral and management

The ‘traffic-light’ system, developed by APSI (see Fig. 1), sets out how risk identification and referral can be managed between services. Low risk presentations can be appropriately managed by APSI and moderate risk cases (highlighted by amber arrows) can be managed by APSI in collaboration with local secondary care and associated services. In contrast, high-risk cases (highlighted by red arrows) are referred directly by GPs to the local secondary care services, and once risk decreases, these cases can be referred to APSI.

Fig. 1 ‘Traffic-light’ system for risk referral and management. *One-to-one intervention also provided by CIPC to General Medical Service (GMS) patients. **Stepped Care intervention provided by the Health Service Executive (HSE) in collaboration with local voluntary agencies.

Table 2 profiles the multiple factors implicated in low, medium and high risk of suicide behaviour. However, the level of risk will also be influenced by several person-centred factors (e.g. repertoire of coping skills; access to natural supports). Hence, an appropriate assessment tool to assess for other complicating factors (e.g. hopelessness; intent; social-isolation, etc.) also needs to be utilised.

Table 2 Indicators of low, medium and high riskFootnote ^a

^a Information adapted from references (Bryan & Rudd, 2006; Substance Abuse and Mental Health Services Administration, 2009; Craig & Rudd, Reference Craig and Rudd2006).

What an effective suicide prevention service needs to provide

Effective suicide prevention and management requires recovery-based principles to permeate all aspects of support and service provision. More specifically, a recovery model of mental health represents a holistic approach to the concerns of service users and their carers that recognises the need to develop a ‘tight bundle of relevant responses’ (Heginbotham, Reference Heginbotham1999: 258) that directly correspond to their needs and wishes. The recovery model also refutes the episodic model of health care provision that provides intensive intervention only in acute periods by seeking to provide ‘on-going therapeutic input and the need for a significant degree of mental health team-working and collaboration between different agencies so that the totality of service users’ needs is addressed’ (Byrne & Onyett, Reference Byrne and Onyett2010: 14).

Suicide prevention measures recommended by WHO include primary and secondary mental health services working together in an integrated manner to provide brief interventions to those at risk (WHO, 2012: 18). Indeed, effective communication between primary and secondary care services is vital for suicide prevention (DoH, 2012). The National Institute for Health and Clinical Excellence (NICE, 2011b: 18) recommends that ‘if a person who self-harms is receiving treatment or care in primary care as well as secondary care, primary and secondary health and social care professionals should ensure they work cooperatively, routinely sharing up-to-date care and risk management plans’. Meta-analyses have indicated collaborative care to be both more effective and cost efficient than treatment as usual (TAU) for depression (Gilbody et al. Reference Gilbody, Bower, Fletcher and Sutton2006; Glied et al. Reference Glied, Herzog and Frank2010).

Figure 2 indicates how the APSI steps map onto the continuum of suicide behaviour.

Fig. 2 Estimated prevalence rates of suicidal ideation, serious thoughts of death, suicidal plans and suicide attempts. **Based on Irish population figures (Casey et al. Reference Casey, Dunn, Kelly, Birkbeck, Dalgad, Lehtinen, Britta, Ayuso-Mateos and Dowrick2006; CDCP, 2011; CSO, 2011) estimates as above.

Subsumed within an inter-service shared-care recovery model, any suicide prevention service needs to reflect what the literature has been found to be effective in preventing suicide. A systematic review indicated that several key factors point towards reduced suicide mortality, including education of key stakeholders; increased access to services and assessment; use of CBT and service user choice; a focus on depression; follow-up care; and case co-ordination.

Education of key stakeholders such as health professionals, at risk cohorts and the general public has a strong supportive evidence base. Large population studies indicate a reduction in suicide rates through provision of mental health-related information and media communication [Hegerl et al. Reference Hegerl, Althaus, Niklewski and Schmidtke2005; National Office for Suicide Prevention (NOSP) 2011]. Provision of information to at-risk groups, such as males (NOSP, 2007), and in raising general awareness of services (Mann et al. Reference Mann, Apter, Bertolote, Beautrais, Currier and Haas2005) have shown some promising results in suicide prevention (Hegerl et al. Reference Hegerl, Althaus, Niklewski and Schmidtke2005; Jorm et al. Reference Jorm, Christensen and Griffiths2005). Longitudinal data indicates levels of mental health staff above the median within Veterans Integrated Service Networks correlated with decreases in suicide rates amongst veterans of 11.2–12.6% (Katz et al. Reference Katz, Kemp, Blow, McCarthy and Bossarte2013). As mental health is a primary predictor of suicide behaviour (Cheng, Reference Cheng1995), education of frontline workers in suicide prevention has had a positive impact on suicide reduction (Mann et al. Reference Mann, Apter, Bertolote, Beautrais, Currier and Haas2005). Education campaigns aimed at improving physician recognition of risk and depression have also correlated with a reduction in rates of suicide (Mann et al. Reference Mann, Apter, Bertolote, Beautrais, Currier and Haas2005).

Research indicates statistically significant increases, post suicide intervention training, in the use of suicide prevention practice behaviours, knowledge and in self-efficacy in substance abuse treatment providers (Connor et al. Reference Connor, Wood, Pisani and Kemp2013).

Increased access to services and assessment also represents a key preventative strategy. Up to 40% of suicide completers will consult with their GP within 1 month before death but the perceived barriers to accessing secondary care services may mean that such individuals are often managed solely by their GP (Owens et al. Reference Owens, Loyd and Campbell2004). Significant differences exist in urban-rural Irish suicide rates (Kirwan, Reference Kirwan1991; Clarke et al. Reference Clarke, Bannon and Denihan2003), with suicide rates in rural males rising by 50% between 1980 and 1990 compared with no corresponding rise in urban males (Hirsch, Reference Hirsch2006). Research also indicates a steady rise in suicide rates in rural females (Kelleher et al. Reference Kelleher, Taylor and Rickert1992, Reference Kelleher, Keeley and Corcoran1997; Rygnestad, Reference Rygnestad1992). A significant number of people who engage in self-harming behaviour have not been provided with an assessment, and approximately one in every 100 of such individuals will die by suicide during the following year (DoH, 2012). Hence, information about past suicide attempts, history of DSH (Crowley et al. Reference Crowley, Kilroe and Bourke2004), and directly asking about suicidal ideation (Schulberg et al. Reference Schulberg, Hyg, Bruce, Lee, Williams and Dietrich2004: 341; Vannoy et al. Reference Vannoy, Fancher, Meltvedt, Unützer, Duberstein and Kravitz2010) are important factors in identification of risk.

Use of CBT and service user choice are indicated as highly effective intervention strategies. Depression and anxiety are strongly associated with suicide behaviour (Tarrier et al. Reference Tarrier, Taylor and Gooding2008). Although no single intervention has been shown to be effective (Gunnell & Frankel, Reference Gunnell and Frankel1994), CBT-based interventions have shown promising results, particularly in the treatment of depression, anxiety and reducing DSH (Knesper, Reference Knesper2010). An RCT investigating self-harm found that CBT was an effective treatment intervention (Slee et al. Reference Slee, Garnefski, van der Leeden, Arensman and Spinhoven2008). Research also indicates that providing CBT results in an ~50% reduction in repeat suicide attempts (Mann et al. Reference Mann, Apter, Bertolote, Beautrais, Currier and Haas2005). Service user choice regarding psychological services can improve both attendance and clinical outcomes (Swift & Callahan, Reference Swift and Callahan2009). NICE (2004, 2009, 2011a) recommends CBT-based bibliotherapy for a number of mild-to-moderate mental health presentations. Meta-analysis indicates such an approach is as effective as other brief therapy for anxious and depressive presentations (Den Boer et al. Reference Den Boer, Wiersma and van den Bosch2004). NICE (2011a) also recommends cCBT and bibliotherapy for managing mild-to-moderate anxiety.

Timely recognition of depression (Goldney, Reference Goldney2005; Hepner et al. Reference Hepner, Rowe, Rost, Hickey, Sherbourne, Ford, Meredith and Rubenstein2007; IMO, 2008) is required to prevent suicidal behaviour. A primary care-based RCT (n=297) indicated that therapist-guided internet therapy was more effective than GP interventions for depression. Education of frontline workers regarding treating depression has correlated with a decrease in suicide rates (Rutz et al. Reference Rutz, Walinder, Von Knorring, Rihmer and Pihlgren1997). Brown et al. (Reference Brown, Ten Have, Henriques, Xie, Hollander and Beck2005) indicates a sample group of those at ‘high risk’ of suicidal attempts who were provided CBT resulted in a reduction in self-harm, suicidal ideation and an improvement in symptoms of depression (Knesper, Reference Knesper2010).

Follow-up care and case co-ordination including a systematic programme of contact with at-risk individuals has been shown to have a preventative effect for up to 2 years (Appleby et al. Reference Appleby, Dennehy, Thomas, Farragher and Lewis1999; King et al. Reference King, Baldwin, Sinclair, Baker, Campbell and Thompson2001). Studies suggest that co-ordinated care (Knesper, Reference Knesper2010) is particularly important for those who have been recently discharged from inpatient units (Large et al. Reference Large, Sharma, Cannon, Ryan and Nielsson2011). Dhossche et al. (Reference Dhossche, Ulusarac and Syed2001) point to insufficient aftercare as a primary factor in elevating risk for discharged secondary care mental health patients. Gilbody et al. (Reference Gilbody, Bower, Fletcher and Sutton2006) meta-analysis results indicated that collaborative care, including follow-up phone-calls and utilisation of a ‘depression care manager’, to be more effective than TAU regardless of length of intervention. The ‘Improving Mood-Promoting Access to Collaborative Treatment’ programme has shown promising results in the United States, evidenced by better quality of life; fewer symptoms of depression; higher rates of physical functioning (Hunkeler et al. Reference Hunkeler, Katon, Tang, Williams, Kroenke, Lin, Harpole and Unützer2006); and improvements to both mental and physical health (Unützer et al. Reference Unützer, Katon, Callahan, Williams, Hunkeler, Harpole, Marc Hoffing, Della Penna, Hitchcock Noël, Lin, Areán, Hegel, Tang, Belin, Sabine Oishi and Langston2002).

Table 3 describes the stepped-care model utilised in APSI; the goals of each step for identifying and managing risk; and how APSI adheres to the evidence base listed above. It is evident from this table that APSI provides a number of stepped-care interventions aimed to constitute a ‘tight bundle’ of treatment options for service users. Table 3 indicates that APSI also provides a continuum of care aimed at seamless transition between higher-intensity secondary care services and/or other primary care intervention services.

Table 3 How the APSI model of service delivery targets the continuum of suicide behaviour.

APSI, Access to Psychological Services Ireland; DSH, deliberate self-harm; SCAN, Suicide Crisis Assessment Nurse; cCBT=computerised cognitive behavioural therapy.

^a CIPC also provides one-to-one therapy for General Medical Service (GMS) users.

How APSI can integrate with existing services?

In the current Irish service provision context, those at risk of suicide may self-refer or be referred via their GP/primary care health worker for observance in an acute hospital unit. Typically discharged from the acute unit after a short period of time, these service users may then receive intervention from secondary care mental health services, dependent on their level of assessed risk/severity of mental health presentation. While this system can initially manage those at ‘high risk’ of suicide, insufficient after-care elevates risk for discharged individuals (Dhossche et al. Reference Dhossche, Ulusarac and Syed2001).

Those at risk for suicide may also have access to a mental health liaison nurse, typically in a general hospital setting. In a study conducted by the Office of the Nursing and Midwifery Director, respondents indicated the need for increased psychosocial interventions and service user facilitation in using available community supports for those at risk of suicide (HSE, 2012a). Research indicates that introduction of a nursing role in consultation with mental health services was viewed as beneficial in terms of practical care-orientated interventions and in negotiating access to mental health services (Sharrock et al. Reference Sharrock, Grigg, Happell and Keeble-Devlin2006).

The Suicide Crisis Assessment Nurse (SCAN) service was piloted in rural Irish settings between 2007 and 2010 and had since been rolled out in many other areas. It aims to provide increased access to risk assessment by nurses, and it also provides informal information and support to GPs on client management. There is also the recently rolled out medical card holder-only Counselling in Primary Care Service (or CIPC).

However, SCAN and other services offer what are effectively fragmented and low throughput models of care with service users and their carers having to potentially engage with several different services. The Mental Health Commission recommends that services be underpinned by a number of principles including accessibility, comprehensiveness, continuity, co-ordination, recovery-based and timeliness (Byrne & Onyett, Reference Byrne and Onyett2010). In providing an integrated suite of interventions including rapid access to a comprehensive risk assessment in each primary care team area, APSI provides the improved continuity and co-ordination of care that is required by those service users on the suicide behaviour continuum.

Looking to the immediate future, given the indicated low baseline of shared care activity in our services (McHugh & Byrne, Reference McHugh and Byrne2013), it is important that multi-agency shared care protocols are developed with a view to establishing after-care services for those post-high risk phase; increasing the efficiency with which referrals that involve the continuum of suicide behaviour are processed; and providing access to mental health interventions (e.g. APSI) for those at risk who do not meet the criteria for secondary care mental health intervention. More robust inter-agency links may also act to reduce the stigma associated with suicide behaviour (WHO, 2012); and remove potential barriers in accessing services (Owens et al. Reference Owens, Loyd and Campbell2004) such as making it easier for service users and their carers to navigate between the appropriate services.