Appendix Athe Evidence Profile

Appendix AThe evidence profile

This appendix provides a summary of each of the papers and reports included in the review and shows their degree of conformity to the three evaluation criteria—quality and risk of bias, quantity (that is, the number of study participants), and their generalisability to the population of interest.

Table A.1Suicidal behaviour and ideation in the Australian community

Author(s) / Country, study group and period / Main measure / Summary of results / Evaluation criteria
Rates/prevalence / Risk factors/groups / Quality and risk of bias / Quantity (number of participants) / Generalisability
De Leo et al., 2005 / Brisbane and Gold Coast residents from the general population / Non-fatal suicidal behaviour in the Australian community through the WHOSUPRE-MISS study / 0.4% of participants had attempted suicide in the preceding 12 months and 4.2% had attempted suicide in their lifetime. 10.4% of the overall sample had seriously considered suicide in their lifetime and 4.4% had made a suicide plan / Individuals reporting a suicide attempt or plan often had problems with their partners (40.7%), followed by financial difficulties (24.1%), problems with parents (22.8%) and difficulty in forming and sustaining social relationships (21.7%) / Provides clearly defined population and reliable standardised data collection method / 11,572 persons aged over 18 / High generalisability to metropolitan Queensland
Page et al., 2006 / Australian suicide cases with socio-economic status obtained from censusesfrom 1979 to 2003 / Secular trends in SES differentials in suicides in Australia / The largest disparities in suicide rates for different SES were witnessed in males, particularly young males. In the most recent period suicide rates among young malesincreased in the low SES group from 44.8 per 100,000 in 1994–1998 to 48.6 per 100,000 in 1999–2003 (an 8% increase). Suicide rates in the middle SES group decreased from a peak of 37.3 per 100,000 to 33.5 per 100,000 (a 10% decrease) and in the high SES group from 33.0 per 100,000 to 27.9 per 100,000 (a 15% decrease) / Young males with low SES / Provides clearly defined population and reliable standardised data collection method / All suicides (Australian Bureau of Statistics) / High generalisability to Australian community
Taylor et al., 2007 / South Australian general population data collected using a monthly risk factor surveillance system in 1997 to 2005 / Relationship between suicidal ideation and a range of risk, social and demographic factors and related health concerns / 4.7% of South Australians aged 16 years and over had suicidal ideation / Among the predictors of suicidal ideation were being separated, divorced or never married, having problems with money, psychosocial stress (K10), a lack of physical activity or fruit consumption, regular health service use, and mental health service use within the last 4 weeks / Provides clearly defined population and reliable standardised datacollection method / 5,037 people who participated in telephone interviews / High generalisability to South Australian community
ABS, 2008; Slade et al., 2009 / Australian general population in 2007 / The 2007 National Survey of Mental Health and Wellbeing including data on non-fatal suicidal behaviour in the Australian population / 0.4% of the Australian community had attempted suicide in the preceding 12 months. 13.3% of Australian adults reported suicidal ideation in their lifetime, 4% reported a suicidal plan and 3.3% reported a suicide attempt. 2.3% of the sample reported suicidality within the preceding 12 months / Among the risk factors for non-fatal suicidal behaviour in the preceding 12 months were being a young female and having a diagnosed mental illness; 8.6% of people with a mental illness were suicidal in the preceding 12 months compared with 2.3% of those without a mental illness / Provides clearly defined population and reliable standardised data collection method (national survey) / 16,015,300 people aged 16–85 years / High generalisability to Australian community
Andersen et al., 2010 / Australian workers employed in selected occupations, identified through the Queensland Suicide Registerin 1990 to 2006 / Rate and risk of suicide in different occupations and theoverall working population / Overall suicide rates by occupation were as follows: construction workers, 18.6 per 100,000;transport workers, 19.1 per 100,000;agricultural workers, 24.1 per 100,000; cleaners, 12.8 per 100,000; nurses, 9.0 per 100,000; and overall employed population, 10.6 per 100,000 aged 15–64 years / Risk groups included agricultural, construction, transport and cleaning workers for males and artists and nurses for females / Provides clearly defined population and reliable standardised datacollection method / 7,652 suicide deaths from the Queensland Suicide Register / High generalisability to the working population in Queensland
Large & Nielssen,2010 / Australian general population in 1988 to 2007, retrieved through Australian Bureau of Statistics data / Suicide mortality rates and suicide methods for males and females in Australian states and territories from 1988–1997 and 1998–2007 / There was a decline in rates by shooting, gassing, poisoning and drowning in males and a decline in shooting, gassing, jumping from a height and drowning among females, but an increase in hanging by both males and females in 1998 to 2007 compared with 1988 to 1997 / n.a. / Provides clearly defined population and reliable standardised datacollection method / All suicides (Australian Bureau of Statistics) / High generalisability to Australian community
Martin et al., 2010 / The Australian Epidemiological Study of Self-injury / Studied self-injury in the Australian community / 2.6% of Australians aged 10–100 years had self-injured in the preceding 12 months / n.a. / Provides clearly defined population and reliable standardised data collection method / 12,006 people aged 10–100 years / High generalisability to Australian community
Sankaran-arayanan et al., 2010 / Australian suicide cases attending mental health services in Newcastle and Maitland, New South Wales, in 2003 to 2007 / Studied the differences in urban and rural suicide in mental health patients / There were 44 suicides over the study period and 62% of these resided in rural areas. The difference between rural and urban suicide age-standardised rates was statistically significant (3.1 per 100,000 per year compared with 1.1 per 100,000 per year) / Individuals living and seeking mental health treatment in rural areas were at increased risk of suicide compared with their urban counterparts / Provides clearly defined population and reliable standardised data collection method / 44 suicide deaths / Moderate generalisability to Australian community
Relatively small sample limits generalisablity; also limited to patients of mental health services
Cheung et al., 2012 / Australian general population in 2004 to 2008 / Studied the metropolitan–rural–remote differentials of suicide in all states and territories / Overall, the male age-standardised rates for metropolitan, rural and remote zones were 15.67, 18.19 and 30.0 per 100,000 person-years respectively. The rates for females were 4.6, 4.56 and 5.54 per 100,000 person-years. With the exception of the Northern Territory and New South Wales, remote and rural zones had significantly higher male suicide rates than metropolitan zones / Possible risk factors included socio-economic deprivation, compositional factors, high risks for Indigenous people, and low access to mental health services / Provides clearly defined population and reliable standardised data collection method / All suicides (Australian Bureau of Statistics) / High generalisability to Australian community
Ide et al., 2012 / First-generation immigrants in Australia and the Australian-born population from 1974 to 2006 / Compared the rates of suicide in different groups of first-generation immigrants and the Australian-born population / Male immigrants were at increased risk of suicide during 1974 to 1988, although over the years the trend reversed. In 2004 to 2006 the suicide rate in males born overseas was lower than that in Australian-born males (16.5 per 100,000 compared with 22.7 per 100,000). Female suicide rates remained similar between groups over the study period / During 2004 to 2006 the highest suicide rates were in males from Eastern Europe (35.3 per 100,000), New Zealand (24.6 per 100,000) and Western Europe (24.2 per 100,000). The Australian-born male suicide rate during the period was 22.7 per 100,000. The highest rates in females during the same time period were in migrants from Western Europe (8.1 per 100,000) and the United Kingdom and Ireland (7.7 per 100,000); this compared with 5.5 per 100,000 in Australian-born females / Provides clearly defined population and reliable standardised data collection method / All suicides (Australian Institute of Health and Welfare) / High generalisability to Australian community
Milner et al., 2014 / Australian general population using data from the National Coroner’s Information System in2001 to 2010 / Association between suicide and employment status and suicide rates during the period of the recession (2007–2009) compared with the preceding year (2006) / During 2001–2010 the overall age-adjusted annual rate of suicide in economically inactive/unemployed males was 57.7 per 100,000 and 20.2 for females. For employed people the overall suicide rate was 10.5 per 100,000 males and 2.2 per 100,000 females. Suicide rates increased among economically inactive/unemployed and employed males during the recession and among economically inactive/unemployed females but notamong employed females / Economically inactive/unemployed males and females were at increased risk of suicide compared with employed males and females / Provides clearly defined population and reliable standardised datacollection method / All suicides (Australian Bureau of Statistics) / High generalisability to Australian community
ABS, 2015 / Australian general population in 2013 / The prevalence and demographic variables of suicide in the Australian community / There were 2,522 cases of suicide in Australia in 2013 (1,885 males and 637 females). Overall suicide rate was 10.9 per 100,000 (16.3 per 100,000 for males and 5.4 per 100,000 for females) / Overall, the greatest risk was in the 85 year and over age group (17.7 per 100,000). In males the highest risk was also in the 85 years and older group (38.3 per 100,000), while in females the highest risk was in the 35–44 year age group (8.5 per 100,000) / Provides clearly defined population and reliable standardised data collection method / 2,522 suicides / High generalisability to Australian community

n.a. Not available.

Table A.2Suicidal behaviour and ideation among Australian serving military personnel

Author(s) / Country, study group and period / Main measure / Summary of results / Evaluation criteria
Rates/prevalence / Risk factors/groups / Quality and risk of bias / Quantity (number of participants) / Generalisability
Hadfield Sheffield, 2009 / Australian serving military personnel 2000 to 2007 / Suicide rate in serving military personnel / Suicide rates in the Australian Defence Force from 2000 to 2007 were lower than in the male general population (13.8 per 100,000 compared with 23.6 per 100,000) / n.a. / Provides clearly defined population and reliable standardised datacollection method / n.a. / High generalisability to Australian serving military personnel
McFarlane et al., 2010 / Australian serving military personnel in 2010 / Mental health of Australian serving military personnel, including non-fatal suicidal behaviour / 3.9% of Australian Defence Force personnel reported suicidal ideation, 1.1% reported making a plan, and 0.4% reported attempting suicide / Compared with the community,Australian Defence Force personnel had more than double the risk of suicidality (ideation and plans) but were equally likely to make an attempt. Mental illness was a major risk factor for suicide attempts in ADF personnel’ with 2.8% of personnel with a mental illness attempting suicide in the preceding year (compared with 0.1% with no disorder). This suggested that 90% of the ADF personnel who attempted suicide in the preceding 12 months had a psychiatric disorder (95% CI 75%, 100%) / Provides clearly defined population and reliable standardised datacollection method / 24,481 Australian Defence Force personnel / High generalisability to Australian serving military personnel
Kõlves et al., 2012 / Australian serving military personnel 2000 to 2010 / Suicide rate in Australian military personnel, as retrieved from Australian Defence Force suicide database / Males in the Australian Defence Force had significantly lower suicide rates than males in the general population in the 25–29 year age group (13.31 per 100,000 compared with 25.82 per 100,000), 30–34 year age group (15.08 per 100,000 compared with 28.61 per 100,000) and 35–39 year age group (2.81 per 100,000 compared with 26.78 per 100,000). Small numbers of females limited the ability to calculate female rates / The mean annual crude suicide rate was highest in the Navy (16.6 per 100,000), followed by the Army (14.3) and the Air Force (7.4) / Provides clearly defined population and reliable standardised datacollection method / 73 Australian Defence Force suicides (68 males and 5females) / Moderate generalisabilityto Australian serving military personnel
Small number of females limited analysis for this group

n.a. Not available.

Table A.3Suicide mortality among international serving military personnel

Author(s) / Country, study group and period / Main measure / Summary of results / Evaluation criteria
Rates/prevalence / Risk factors/groups / Quality and risk of bias / Quantity (number of participants) / Generalisability
Skegg et al., 2010 / New Zealand serving military personnel in 1973 to 2004 / Rate of suicide in military personnel relative to a number of other occupations, including dentists, doctors, farmers, hunters, nurses, pharmacists, police and veterinarians / Few of the occupations investigated had high risks of suicide, some were at lower risk than the overall employed population. Men in the armed forces were at low risk (male crude suicide rate 11 per 100,000 compared with 29.8 per 100,000 among all employed males). No female suicides occurred in the military personnel group / n.a. / Provides clearly defined population and reliable standardised datacollection method / 940 suicide deaths in the selected occupations. 26 deaths in male military personnel, no deaths in female military personnel / High generalisability to New Zealand serving military personnel
Moderate generalisability to Australian serving military personnel due to similarities between countries
Ramchand et al., 2011 / US serving military personnel in 2001 to 2008 / Suicide rate in US serving military personnel over time / The suicide rate among serving military personnel climbed from 10.3 per 100,000 in 2001 to 15.8 per 100,000 in 2008. Rates were highest in the Marine Corps (19.5 per 100,000) and the Army (18.5 per 100,000). Compared with the US general population, the rate in the armed forces was noticeably higher (general population suicide rate of approx. 10 per 100,000) / Army and Marine Corps personnel were at an increased risk compared with Navy and Air Force / Provides clearly defined population and reliable standardised datacollection method / n.a. / High generalisability to US serving military personnel
Limited generalisablity to Australian serving military personnel
Skopp et al., 2012 / US armed forces personnel in 2001 to 2009 / Case–control study comparing suicide cases to matched controls to analyse the impact of traumatic brain injury, mental illness, and life events on suicide / n.a. / Risk factors for suicide included mood disorders (1.6, 1.37–1.80), relationship problems (2.0, 1.51–2.63), and family circumstances problems (2.0, 1.25–3.04). While traumatic brain injury, alcohol dependence and PTSD were not risk factors individually, co-morbidity of mental health diagnosis significantly increased the odds of suicide (χ2=60.8, p <0.01) / Provides clearly defined population and reliable standardised data collection method / 1,764 serving military suicides and 7,018 matched controls / High generalisability to US serving military personnel
Limited generalisablity to Australian serving military personnel
Logan et al., 2015 / US Army personnelin 2005 to 2010 / Risk for suicide amongserving military personnel compared with suicide cases from the US general population / n.a. / The most common predictors of suicide for both groups were mental health problems, intimate partner problems, alcohol or substance abuse problems, recent crises and job problems. Army suicide cases were significantly less likely to have previous suicide attempts or to have disclosed their intentions to others (p= <0.05) / Provides clearly defined population and reliable standardised data collection method / 141 active Army suicides and 563 civilian suicides / Moderate generalisability to US serving military personnel
Limited generalisablity to Australian serving military personnel
Street et al., 2015 / US active duty regular army soldiers in 2004 to 2009 / Compared suicide rates in deployed soldiers with those for never deployed and previously deployed soldiers / The suicide rate among currently deployed women (14.0 per 100,000 person-years) was 3.1–3.5 times the rate for other (neverdeployed and previously deployed) women. The suicide rate among currently deployed men (22.6 per 100,000 person-years) was 0.9–1.2 times the rate for other men / The increased risk among female currently deployed soldiers compared with other female soldiers was greater than the increased risk among male currently deployed soldiers compared with other male soldiers.A significantly higher proportion of female than male soldiers had a history of healthcare visits with diagnoses of mental or behavioural disorders in the preceding 2–4 years (27.3% v. 19.6%; χ= 333.3, p = <0.001) / Provides clearly defined population and reliable standardised data collection method / 975,057 active duty army soldiers / High generalisability to US serving military personnel
Limited generalisablity to Australian serving military personnel

n.a. Not available.

Table A.4Non-fatal suicidal behaviour and suicidal ideation among international serving military personnel

Author(s) / Country, study group and period / Main measure / Summary of results / Evaluation criteria
Rates/prevalence / Risk factors/groups / Quality and risk of bias / Quantity (number of participants) / Generalisability
Belik et al., 2009 / Canadian military personnel who completed the Canadian Community Health Survey: Mental Health and Well-Being Canadian Forces Supplement (CCHS-CFS) in May–Dec 2002 / Traumatic events as risk factors for suicide attempts in serving military personnel / The prevalence of lifetime suicide attempts in men was 2.2% and the prevalence in women was 5.6% / Risk factors included sexual and other interpersonal traumas (for example, rape, sexual assault, spousal abuse and child abuse) for both men (adj. ORs ranging from 2.31 to 4.43) and women (adj. ORs ranging from 1.73 to 3.71). The number of traumatic events experienced was positively associated with an increased risk of suicide attempts / Provides clearly defined population and reliable standardised datacollection method / 8,441 serving military personnel aged 16 to 54 years; response rate 81.1% / High generalisability to Canadian serving military personnel
Moderate generalisability to Australian serving military personnel due to similarities in military (size and structure)
Belik et al., 2010 / Canadian military personnel who completed the Canadian Community Health Survey: Mental Health and Well-Being Canadian Forces Supplement (CCHS-CFS) in May-Dec 2002 / Prevalence of suicidal ideation and attempts in Canadian Armed Forces personnel with the general population / There were no significant differences between military personnel and the general population with regard to suicidal ideation in the preceding year (4.0% and 3.8%). Suicide attempts were significantly more prevalent in the civilian population (0.6%) compared with military personnel (0.2%) / Females in the Canadian Armed Forces were significantly more likely to report suicidal ideation and attempts than were their male counterparts. In the civilian population females were also significantly more likely to report suicide attempts than were civilian males; this was not the case for suicidal ideation.A significant interaction effect was noted between sex and suicide attempts (interaction OR = 3.86; 95% CI 1.27, 11.74), indicating that females in the military population were significantly more likely to report suicide attempts than were females in the civilian population. Most mental disorders were significantly related to suicidal ideation and suicide attempts in both populations (adj. ORs ranging from 2.23 for panic attack in the civilian population to 80.73 for depressive episode in the military). / Provides clearly defined population and reliable standardised datacollection method / 8,441 active Canadian military forces personnel and 36,984 citizens aged 15 years or older; response rate 77% / High generalisability to Canadian serving military personnel
Moderate generalisability to Australian serving military personnel due to similarities in military (size and structure)
Nock et al., 2014 / Active duty Army personnel participating in theArmy STARRS All-Army Study (AAS) in Apr–Dec 2011 / Lifetime prevalence and socio-demographic, Army career, and psychiatric predictors of suicidal behaviours / Over their lifetime 13.9% of participants had experienced suicidal ideation, 5.3% reported suicidal plans and 2.4% had attempted suicide.Most reported cases (47.0–58.2%) had pre-enlistment onsets / 38.5% of ideators developed a plan and 17.1% attempted suicide. In most cases the onset of suicidal plans or attempts in ideators occurred within a year of the onset of ideation (58.3–63.3%). Being female was associated with post-enlistment ideation (OR=2.1; 95% CI 1.4–3.1), plans (OR=2.7; 95% CI 1.5–5.1) and attempts (OR= 3.3; 95% CI 1.5–7.5); lower rank was associated with post-enlistment suicide attempts(OR = 5.8; 95% CI 1.8–18.1); and the number of deployments increased the risk of suicide attempts, with the greatest risk being in those with 3 or more deployments (OR = 3.7; 95% CI 0.9–16.1) / Provides clearly defined population and reliable standardised datacollection method / 5,428 non-deployed soldiers participating in a group self-administered survey / Moderate generalisability to US serving military personnel. Only included non-deployed personnel
Limited generalisablity to Australian serving military personnel
Vanderploeg et al., 2014 / Deployed and non-deployed US serving military personnel in 2009 and 2010 / Risk and protective factors for suicidal ideation in deployed and non-deployed military personnel / The overall percentage of suicidal ideation at the time of the survey was 4.2% / Personnel who had returned from deployment reported significantly more suicidal ideation than non-deployed personnel (5.5% compared with 3%) (χ =12.21;p =<.001). Risk factors in non-deployed personnel included prior psychological trauma, major depressive disorder, PTSD and excessive alcohol drinking were significant predictors of suicidal ideation. Risk factors among deployed personnel included a deployment-related mild traumatic brain injury (OR=2.69;p=.025) and current PTSD (OR=15.24;p =<.001).Being married or coupled compared with never being married was also associated with increased suicidal ideation (OR =3.91;p=.017). Post-deployment social support was not protective against suicidal ideation / Provides clearly defined population and reliable standardised data collection method / 3,098 Florida National Guard members / High generalisability to US serving military personnel
Limited generalisablity Australian serving military personnel
Ejdesgaard et al., 2015 / Deployed Danish soldiers 1990 to 2009 / Risk and protective factors for suicidal ideation and suicide attempts among deployed Danish soldiers / n.a. / Significant risk factors for suicidal ideation included drug abuse, a poor financial situation before deployment, a heavy workload and/or repatriation during deployment, and attending a poor athletic and recreation program after deployment.Significant protective factors against suicidal ideation were support from friends at home during deployment and appreciation by the general population after deployment. Significant risk factors for suicide attempts were an unhappy childhood and pointless tasks during deployment. There were no significant protective factors against suicide attempts / Provides clearly defined population and reliable standardised datacollection method / 1,264 Danish soldiers deployed between 1990 and 2009 / High generalisability to Danish serving military personnel
Limited generalisablity to Australian serving military personnel

n.a. Not available.