Abstract
Over 2500 people die of suicide every year in Australia with the majority of victims under the age of 40 years [1]. This paper describes three primary prevention initiatives that were undertaken in Esperance, Western Australia. Esperance is a geographically isolated community of 17 000 individuals. In the year prior to the project, it had eight documented suicides translating into a rate of 47 per 100 000 (95% Confidence Intervals of 25–69 per 100 000). This compared with a rate in Australia of 13 per 100 000.
The aim of this project was to help the Esperance Suicide Prevention and Support Taskforce implement research findings to the prevention of suicide. The taskforce included representatives from health services, social services, police, primary care, voluntary agencies, sporting clubs and local business. We wanted to evaluate three initiatives for the primary prevention of suicide: (i) providing suicide awareness sessions for staff members in health, education and social services; (ii) limiting the sale of over the counter analgesics (aspirin and paracetamol) to packets containing less than the minimum lethal dose; (iii) implementing Commonwealth media guidelines in the reporting of suicides.
As regards the first initiative, training increases the ability, confidence and willingness of general practitioners (GPs) and community health staff to help a person at risk of suicide [2–4]. Training sessions also improve the ability of GPs to detect young people with psychological disturbance, suicidal ideation and depression, with the frequency of detection being related to the amount and quality of the training received [4].
Regarding the two other initiatives, admissions in Western Australia for deliberate self-harm with paracetamol and aspirin have increased for both genders between 1981 and 1998 [5], particularly females aged between 15 and 19 years [6]. Studies in Britain and France have shown that paracetamol overdoses have fallen in number and severity after paracetamol became less available [7–10], although the results of a New South Wales study were inconclusive [11]. Perhaps this was because restrictions on access in Australia only lasted 4 months. Restricting the reporting of a particular method of suicide also leads to a reduction in the number of persons who use this means [12–14], and there are Commonwealth Government guidelines on the reporting of suicide in Australia [15].
We assessed changes in knowledge, awareness, attitudes and behaviour in each of these three areas after our intervention.
Method
Process measures were used for the evaluation of the project as numbers of attempted and completed suicides in Esperance were too small to have sufficient statistical power. In the case of suicide awareness, a survey questionnaire was derived from previous work [2, 16–21]. The questionnaire was designed to obtain qualitative and quantitative measures of the following: (i) knowledge of suicide related issues on a three-point scale with a maximum score of 10 (Section A) [16–20]; (ii) willingness to intervene with an individual at risk of suicidal behaviour and raise the issue of suicide on a five-point scale derived from the Suicide Intervention Beliefs Scale (SIBS) [2] (Section B); (iii) knowledge of risk factors and behavioural indicators identified in the literature as being associated with suicide on a five-point scale with a maximum score of 30 (Section C) [16–21]; (iv) perceived level of knowledge of professional and ethical responsibilities in responding to suicide risk on a five-point scale (Section D); and (v) perceived comfort, competence and confidence in responding to a person at risk of suicidal behaviours developed by Turley and Tanny on a four-point scale across three items (Section E) [2].
Differences in awareness, knowledge and attitudes between occupational groups were examined using descriptive statistics, Students t-test and analysis of variance as appropriate. Equivalent non-parametric tests (Mann–Whitney & Kruskal–Wallis tests) were also used to minimize type 1 error.
Process measures on the effect of the two other initiatives included a change in the percentage of retail outlets, such as chemists and supermarkets, selling paracetamol or aspirin packets in less than potentially lethal quantities (8 g), and changes in the awareness, knowledge and use of Commonwealth Government guidelines on the reporting of suicides among printed and broadcast media. This was assessed by a telephone interview with media at local, regional and statewide levels using a standardized pro forma.
Changes in awareness, knowledge, attitudes and behaviour following each of the three initiatives were examined using descriptive statistics, Paired Sample t-test and the Wilcoxon Signed Ranks Test as appropriate.
Results
Suicide awareness
Respondent characteristics
Two hundred questionnaires were distributed to staff members in 13 local services covering health, education and social services. One staff member in each service distributed questionnaires to workers who were in regular contact with individuals at potential risk of suicidal behaviour. Seventy-five per cent responded (50 males and 99 females). Hospital and school staff were the two largest categories each comprising a quarter of the sample (n = 42 and 39, respectively). Other groups included mental health (n = 12), community health (n = 11), employment services (n = 11), police (n = 9), GPs (n = 8), counselling services (n = 6), crisis service (n = 6), and family support services (n = 5). Although 70% had either an undergraduate or postgraduate degree (n = 104), only 40% had received any training on suicide prevention (n = 90). Ninety-three participants (62.4%) reported that they had some contact with individuals at risk of suicide in the previous 6 months. The majority of respondents (59.7%) were aware of their own organization's guidelines for responding to suicidal risk.
Baseline
Intervention and outcome assessment
All respondents were invited to attend a training day. This included training related to the way in which attitudes can influence responses to individuals at risk of suicide, risk factors for suicide, communication skills and role play, the association with deliberate self-harm and depression, and identifying level of suicide risk through scenarios and discussion.
The training package was derived from the Youth suicide advisory committee gatekeeper training manual and Lives worth living, a training manual designed for rural and remote workers. Information on risk factors for suicide and suicide prevention was taken from Life: a framework for prevention of suicide and self-harm in Australia developed by the Commonwealth Department of Health and Aged Care and Setting the evidenced-based research agenda for Australia a literature review of suicide prevention funded by the National Youth Suicide Prevention Strategy [16–21].
Separate sessions were organized for hospital and community staff. Thirty-three persons participated in the training (16 nurses and 17 community health workers), of whom 21 (66%) returned questionnaires covering the same areas as the baseline assessments. The results showed a significant increase in participants’ knowledge of suiciderelated issues and risk factors for suicide (Table 1). In terms of beliefs about suicide intervention there were increases in both the willingness to intervene and to raise the issue of suicide (Table 1). The participants’ reported level of knowledge regarding professional and ethical responses in suicide prevention also increased significantly as did their comfort, competence and confidence levels when assisting a person at risk of suicidal behaviour (Table 1). In both the baseline and outcome assessments, there were identical results using the equivalent nonparametric tests.
Changes in participant's knowledge of suicide-related issues (Section A), beliefs about suicide intervention (Section B), risk factors for suicide (Section C), knowledge on professional and ethical responses in suicide prevention (Section D), and comfort, competence and confidence levels when dealing with suicidal behaviour (Section E)
Limiting availability of over the counter analgesics
All retail outlets in Esperance selling aspirin and paracetamol were contacted (three supermarkets, one delicatessen and three pharmacies) and given details of the benefit of limiting the quantity of over the counter analgesics. All outlets sold unrestricted amounts of paracetamol and aspirin. They were asked if they would limit sales of either medication to packets containing less than 8 g, the minimum lethal dose. Following the intervention of the research worker (JS), one supermarket owner agreed to implement the strategy (pending agreement from all other retailers), a second claimed that the intervention had increased his awareness of the danger of analgesics but was unwilling to implement the strategy stating it was impractical, and a third claimed to be restricted by central office policy. Two pharmacists stated that they were hampered in implementing the strategy by prepackaged analgesics carrying the pharmacy brand name that were produced in only one size (over the lethal amount). However, both stated that the intervention had increased their concern and willingness to monitor the sale of analgesics in large quantities to any one customer. Overall, the strategy was successful in increasing the awareness of four retailers (57%) on the danger of analgesics and in obtaining an agreement from three (43%) to monitor the sale of large quantities to individual customers.
Media guidelines
We interviewed representatives from local and statewide media about their policy when reporting suicide, and their awareness and use of the Commonwealth Health Department Guidelines. These were the Esperance Express newspaper, the Kalgoorlie Miner newspaper, Radio West Esperance, ABC Local Radio (Kalgoorlie), WIN television (Channel 10), Golden West Network (GWN) Television, the West Australian newspaper and ABC TV.
Of the eight media representatives interviewed, only three were aware of the Health Department Guidelines for the reporting of suicide (the Esperance Express newspaper, the West Australian newspaper and the Golden West Network (GWN) television station). Only one believed that the guidelines would influence the way in which they report (or refrain from reporting) the occurrence of suicidal behaviour [22]. Seven out of eight accepted that incidents of suicide should not be reported unless the circumstances were compelling enough to be in the ‘public interest’ such as murder suicides, celebrity suicides or a person facing criminal charges; they also agreed that the use of appropriate terminology was necessary [22]. Although the West Australian newspaper acknowledged the importance of terminology when reporting suicides, it also supported the view of the Press Council that there is insufficient evidence of a clear link between media portrayal and imitation suicides [23]. Of the two local newspapers (the Esperance Express and the Kalgoorlie Miner), the one that did not have a copy of the guidelines asked for one [22]. Both subsequently contacted the researchers to check that coverage met the guidelines, which it did. They had not done this before the intervention. Radio West Esperance also asked for a copy of the guidelines but did not make further contact [22].
Discussion
The high level of community concern about suicide and the geographical isolation of Esperance meant that it was a suitable location for the population-based initiatives discussed in this paper. To our knowledge, this is the first study in Australia to evaluate the impact of three evidence-based interventions in a discrete community. Limitations of this study include the fact that assessments were not carried out blind to intervention status, and that the small population meant that there was insufficient power to measure the effect on rates of deliberate self-harm or suicide, necessitating the use of process measures as proxy indicators of outcome.
Of the three initiatives, the intervention to increase suicide awareness had the most success. The two-thirds of subjects who returned questionnaires reported significant increases in awareness of suicide-related issues and risk factors for suicide, as well as reported levels of knowledge of professional and ethical responses and comfort, competence and confidence levels when assisting a suicidal person. Other studies have shown similar improvements following suicide prevention training [2–4], although there is no evidence as yet to suggest that it reduces suicidal behaviour [1].
Of the other initiatives, there was only evidence of a change in knowledge, attitudes and behaviour in local as opposed to statewide media. This suggests that local initiatives can only succeed if complemented by measures at statewide and national level. Similarly, it proved very difficult to secure the agreement of local pharmacies to limit the sale of over the counter (OTC) analgesics, although Esperance was an ideal location for such an initiative. The town is geographically isolated, making access to alternative outlets difficult, and there were only seven places where OTC medications were sold. Pharmacies were either limited by central policy or the concern of being at a competitive disadvantage. This suggests that implementing voluntary codes of conduct among commercial competitors is difficult, and that legislative measures such as those implemented in the UK and France are required.
Footnotes
Acknowledgements
This project was funded by the Quality Improvement Programme of the Mental Health Division of the Health Department of Western Australia