- Research article
- Open Access
- Open Peer Review
Interventions to reduce suicides at suicide hotspots: a systematic review
BMC Public Healthvolume 13, Article number: 214 (2013)
We’re sorry, something doesn't seem to be working properly.
Please try refreshing the page. If that doesn't work, please contact support so we can address the problem.
‘Suicide hotspots’ include tall structures (for example, bridges and cliffs), railway tracks, and isolated locations (for example, rural car parks) which offer direct means for suicide or seclusion that prevents intervention.
We searched Medline for studies that could inform the following question: ‘What interventions are available to reduce suicides at hotspots, and are they effective?’
There are four main approaches: (a) restricting access to means (through installation of physical barriers); (b) encouraging help-seeking (by placement of signs and telephones); (c) increasing the likelihood of intervention by a third party (through surveillance and staff training); and (d) encouraging responsible media reporting of suicide (through guidelines for journalists). There is relatively strong evidence that reducing access to means can avert suicides at hotspots without substitution effects. The evidence is weaker for the other approaches, although they show promise.
More well-designed intervention studies are needed to strengthen this evidence base.
A ‘suicide hotspot’ is a specific, accessible and usually public site which is frequently used as a location for suicide and gains a reputation as such . The most common types of suicide hotspot are bridges, tall buildings and cliffs [2–5], railway tracks [6, 7], and rural or secluded locations . The suicide methods typically used at these sites, such as jumping from a height, jumping or lying in front of a train and inhalation of car exhaust have a high probability of being lethal . Suicides at hotspots can have a distressing impact on those who witness the event, find the deceased or are involved in some other way [10, 11]. They often receive high profile media coverage , which may increase the risk of ‘suicide contagion’. There is no agreement on the number of suicides that is required to identify a site as a ‘suicide hotspot’. However, more than one suicide at a particular location suggests that the site has appeal for suicidal individuals and provides means or opportunity for suicide, and may therefore warrant intervention .
Various interventions have been implemented to reduce the risk of suicide at suicide hotspots. The current review examines the evidence for the effectiveness of these interventions. Specifically, it addresses the following research question: ‘What interventions are available to reduce suicides at hotspots, and are they effective?’
We searched Medline from its inception to April 2012 using the following search string, with words mapped onto MeSH headings: (suicid* OR hotspot) AND (cliff OR building OR high-rise OR multi-storey OR viaduct OR rail OR metro OR subway OR river OR lake OR sea OR public* OR secluded OR remote OR woods OR forest OR rural OR magnet OR location OR secluded OR bridge OR skyscraper OR car park OR underground OR road OR motorway OR highway OR reservoir OR coast OR jump* OR leap* OR fall OR height OR lie OR lying OR moving object OR carbon monoxide OR car exhaust OR hang* OR firearm OR gun* OR burn* OR drown* OR fenc* OR barrier* OR parapet OR net* OR pit* OR sign* OR poster* OR helpline* OR surveillance* OR CCTV OR patrol* OR media OR reporting OR television OR radio. Reference lists of key review papers and included studies were also searched. We sought only English-language publications.
Study inclusion criteria
We gave careful consideration to the kinds of studies that we would include in the review. Hotspot interventions are generally offered as universal preventive strategies, rather than as preventive strategies targeted at individuals. This means that evaluations usually draw on aggregate or ecological data on completed or attempted suicide, rather than individual-level data. It is practically and ethically difficult to mount cluster randomised controlled trials (cRCTs, regarded as Level I evidence) in this area. Suicide is an emotive issue and suicide hotspots generate considerable community concern, which means that randomly selecting some sites to receive the intervention is generally not feasible. Ecological studies with quasi-experimental designs (non-randomised studies with before-and-after designs and comparison sites, regarded as Level II evidence), are the next best solution, but comparable sites are not always available (for example, where one bridge is a recognised hotspot and nearby bridges do not present the same problem). This means that ecological studies with before-and-after designs and no comparison sites (regarded as Level III evidence) are often the most acceptable and appropriate option in the circumstances.
We included studies in the review if they described an intervention relating to a known suicide hotspot, evaluated it using at least a before-and-after design with no comparison (but preferably a stronger design), and used suicides as the outcome of interest (with or without other outcomes, such as suicide attempts). Studies which only measured suicides after an intervention was put in place and/or only considered outcomes other than suicides (such as calls to a crisis telephone service) were excluded.
The following data were extracted for each study:
Author(s) and date of publication;
Study design and observation period; and
With respect to the findings, data were examined on changes in the number or rate of suicides at the hotspot site and at comparison sites where these were available. Consideration was also given to whether there was any evidence of substitution, either in the form of site substitution (reduction in suicides at the hotspot being accompanied by increases at other nearby sites) or method substitution (reduction in use of one method being accompanied by increases in others).
Nineteen papers describing 14 studies at 13 locations worldwide met our inclusion criteria [15–33], and are summarised in Table 2. There were several instances where the same group of authors used the same core data in more than one paper, augmenting it with data from other sources or with follow-up data [15–20, 26, 32]. In these cases, we took the conservative approach of regarding the different papers as relating to the same study to avoid double-counting of any observed impacts, thereby circumventing the possibility of multiple publication bias. There was one instance in which the same data were examined independently by separate investigators to determine the impact of the same intervention at the same site [24, 27]. These were regarded as separate studies, but their findings are discussed together, again to avoid artificially inflating the collective magnitude of any impact.
The interventions studied fall into four broad categories representing different approaches to suicide prevention. The most commonly-investigated of these is restricting access to means (for example, by installing barriers at a jumping site). The second approach involves encouraging help-seeking (for example, via signs and telephone crisis lines). The third intervention involves increasing the likelihood of intervention by a third party (for example, offering training for staff working at or near suicide hotspots and/or surveillance methods). The final approach is the provision of guidance on responsible reporting of suicide to media professionals, in order to minimise the risk of ‘suicide contagion’ at hotspots. Most of the studies consider a single intervention, but some consider several together.
Restricting access to means
Nine studies have examined the effectiveness of restricting access to lethal means by installing physical barriers at sites that are used for jumping from a height or jumping in front of a train. All of these studies suggest that suicides reduce once means restriction measures are put in place [15–18, 22–24, 27–31], or rise when they are removed [15, 16].
Pelletier , Reisch and Michel  and Sinyor and Levitt  observed no further suicides after barriers were installed on the Memorial Bridge in Augusta, Maine (United States), Muenster Terrace in Bern (Switzerland) and the Bloor Street Viaduct in Toronto (Canada), respectively. Isaac and Bennett  and Skegg and Herbison  reported the same ‘reduction to zero’ finding when access was blocked to Beachy Head in Sussex (United Kingdom) and Lawyers Head Cliff in Dunedin (New Zealand). Bennewith and colleagues [17, 18] also reported substantial decreases in the number of suicides (though not a complete elimination of them) following the erection of fencing on the Clifton Suspension Bridge in Bristol (United Kingdom). The fencing was accompanied by an expansion of the role of bridge staff to include monitoring of incidents, and the installation of CCTV cameras (see below). Lester  and O’Carroll and Silverman  reported similar findings to those of Bennewith and colleagues [17, 18] when they independently examined data on suicides before and after modifications to the Ellington Bridge in Washington, DC (United States). Law et al. also noted a significant decrease in suicides following the introduction of platform screen doors on the Hong Kong underground railway system (Hong Kong) . Conversely, when barriers on the Grafton Bridge in Auckland (New Zealand) were removed for aesthetic reasons, Beautrais and colleagues observed an increase in suicides [15, 16]. Replacement of the original barriers by new ones with an improved design was followed by a decrease in suicides [15, 16].
Eight of the above studies made some attempt to examine whether the reductions in suicide at the sites in question were associated with any commensurate increases in suicide at alternative sites in the given city. Of these, seven demonstrated that the number of suicides from other sites remained the same or decreased for the total population [23, 24, 27–29], or for males (who account for the majority of suicides by highly lethal methods, such as jumping from a height) [17, 18]. However, Sinyor and Levitt  found no reduction in the overall numbers of suicides by jumping in Toronto during their respective study periods, suggesting that some substitution may have occurred.
Pelletier  and O’Carroll and Silverman  explored the substitution phenomenon further, and considered whether the observed reductions in suicides at the means-restricted sites equated to decreases in rates of suicide by any method (that is, not just jumping) in Augusta and Washington DC, respectively. Both of these studies identified small decreases in overall suicide rates following the means-restriction interventions.
Three studies have examined the effectiveness of installing signs and telephones at specific hotspots as a way of encouraging suicidal individuals to seek help [21, 25, 33]. King and Frost  evaluated an intervention that involved the placement of signs providing contact details for the Samaritans placed in car parks in the New Forest in Hampshire (United Kingdom). Lester  evaluated crisis emergency telephones on the Sunshine Skyway Bridge in St Petersburg, Florida (United States) which were installed at the same time as the introduction of a police presence on the bridge (see below). Wong et al.  evaluated an integrated community-based program of initiatives designed to reduce suicides by visitors to Cheung Chau (Hong Kong), an island where a number of people had taken their own lives by charcoal burning in rented holiday flats. This included, among other things, the introduction of a 24-hour telephone hotline service to support people in emotional distress, and the provision of hotline numbers in all holiday flats.
All three studies showed reductions in suicides at the specific sites following the introduction of the intervention [21, 25, 33]. King and Frost  and Wong et al.  examined patterns of suicide at comparison sites (that is, at car parks with no signs in the New Forest, and at other Hong Kong islands, respectively) and found no change over the relevant observation period. King and Frost  also explored the impact of the reduction at the intervention sites on the overall number of suicides in the district and found that this decreased, suggesting that substitution had not occurred.
Increasing the likelihood of intervention by a third party
Three studies have explored the extent to which increasing the likelihood of intervention by a third party at hotspots can reduce suicide at these sites [17, 18, 25, 33]. Two of these studies – by Lester  and Wong et al.  – have already been mentioned. They considered the impact of a range of activities on suicides from the Skyway Bridge and on Cheung Chau Island, respectively. This included a police presence on the former, and gatekeeper training and suicide patrols on the latter. Both studies reported positive findings in terms of reductions in numbers of suicides [25, 33].
The third study, by Bennewith and colleagues [17, 18], was also mentioned earlier. These authors examined an intervention that primarily involved the installation of barriers on the Clifton Suspension Bridge, but also included other components, namely an expansion of the role of bridge staff to include monitoring of incidents, and the installation of CCTV cameras. As noted, the intervention was associated with a reduction in suicides. The number of incidents remained stable, but bridge staff were more likely to be involved in incidents.
Providing guidance on responsible media reporting of suicide
A single study has considered whether providing guidance on responsible reporting of suicide at a hotspot can lead to reductions in suicide at that site, although the authors have strengthened the evidence by adding further data and conducting a follow-up analysis some years after they did the original work [19, 20, 26, 32]. Etzersdorfer and Sonneck [19, 20], Sonneck et al.  and Niederkrotenthaler and Sonneck  found that completed and attempted suicide on the Vienna underground railway system (Austria) rose significantly in the latter part of the pre-intervention period (when sensationalist reporting of suicide was common). They then observed that, following the introduction of guidelines on responsible reporting, suicidal acts dropped dramatically to a level that has been sustained since [19, 20, 26, 32]. There was some evidence that this contributed to an overall reduction in the national suicide rate over time [19, 20, 26, 32].
Interpreting the findings
Our review represents a consolidation of current knowledge about preventing suicides at hotspots, and an identification of gaps in that knowledge. Our starting point was the research question, ‘What interventions are available to reduce suicides at hotspots, and are they effective?’ The answer to the first part of this question is relatively straightforward. Four main approaches have been used at suicide hotspots: (a) restricting access to means (through installation of physical barriers); (b) encouraging help-seeking (by placement of signs and telephones); (c) increasing the likelihood of intervention by a third party (through surveillance and staff training); and (d) encouraging responsible media reporting of suicide (through guidelines for media professionals).
The answer to the second part of the question is more complex. The strongest evidence for effectiveness comes from studies that have looked at restricting access to means through the installation of barriers at jumping sites and on railway networks. This body of evidence consistently suggests that these measures are associated with a reduction in suicides at these sites because they limit access or make it difficult to perform suicidal acts. In the main, the evidence also suggests that restricting access to means at one site does not drive suicidal individuals to seek alternative locations, thereby shifting the problem elsewhere. There are also indications that reducing suicides by a particular method does not lead to substitution of different methods; instead it may have a positive impact on the overall suicide rate. The apparent effectiveness of installing barriers at suicide hotspots is consistent with the broader literature on restricting access to means as a population-level suicide prevention strategy. Reviews by Mann et al.  and Beautrais et al.  suggest that this is one of very few approaches for which there is strong evidence of effectiveness. The theory behind restricting access to means is that it may ‘buy time’ for the individual to reconsider his or her actions, particularly in situations where these actions are associated with impulsivity or ambivalence [36–38].
Beyond this, the evidence is weaker. The evidence for the effectiveness of interventions designed to encourage suicidal people to seek help (for example, crisis telephone lines) and interventions designed to increase the likelihood of intervention by a third party (for example, suicide patrols and CCTV) is limited. Relatively few studies have looked at their impact. Those which have done have tended to examine these strategies in the context of broader suicide prevention programs at given hotspots, and this has made their independent influence difficult to evaluate. It is fair to say, however, that they show sufficient promise regarding reducing suicides at hotspots to be worthy of further testing. Different models (for example, crisis telephone lines which link directly to mental health services ) may be worth exploring.
The provision of guidance on how to report suicides at hotspots has also been subject to limited testing. Only one study has evaluated this strategy specifically. Although this study was well-designed and covered a lengthy observation period, it was restricted to a single setting in a single country. The study suggested that supporting journalists and editors in responsible reporting led to a decrease in suicidal acts on the rail network in question, and that this translated into a genuine overall reduction in suicides nationally. This is consistent with the literature on responsible reporting of suicide more generally (not just in the context of suicide hotspots), which suggests that guidelines can be effective in modifying the behaviour of media professionals and that this, in turn, can minimise imitation suicides [40–42].
Further research in this area is clearly warranted, and good quality evaluations of the latter three types of interventions are particularly necessary. These evaluations should adopt the strongest designs possible in the context of implementing interventions at hotspots. Recognising that these are not likely to be randomised controlled trials, suicide prevention researchers should draw on other areas of public health which have grappled with the issue of what constitutes sufficiently good evidence of effectiveness [43, 44], and on the broader evaluation literature which stresses the importance of underpinning evaluations with sound program theory . Triangulation of data from multiple sources is desirable, and might include both quantitative and qualitative data collection approaches. For example, ecological data (on pre- and post-intervention suicides from the site) could be combined with individual-level data (on the decision-making processes of those who sought help as a result of the intervention) where this is practically and ethically possible. Selecting comparison sites is also advantageous if this is feasible, particularly if the intervention can be rolled out across sites in a staggered fashion . Likewise, examining the dose–response effect of gradually introducing the different components of a multi-faceted intervention may help to tease out the extent to which each individual component is effective. A consideration of cost-effectiveness is also important.
The above approaches to the reduction of suicides at known hotspots have been deemed to constitute current best practice, and have been advocated as part of a suite of potential measures in guidelines developed in England in 2006 (and translated into Japanese in 2007) , Scotland in 2012  and Australia in 2012 . This is appropriate under the circumstances; the need to strengthen the evidence base regarding interventions at hotspots should not prevent us from taking action in the immediate term, and the interventions considered in this review certainly show some promise. However, there is an onus on those who are responsible for funding and implementing these interventions to monitor them carefully to ensure that they are achieving their potential as effective suicide prevention strategies and are not having unintended consequences. Partnering with suicide prevention researchers with expertise in evaluation may be one way to do this.
Limitations of the review
The current review has several limitations which must be acknowledged. We adopted as comprehensive a search strategy as we could, but resource constraints meant that we only had the capacity to search one database, we did not look for grey literature or conference abstracts, and we did not contact any authors for additional information. We did search two other databases (PsycINFO and Scopus) for a subsequent meta-analysis of studies on means restriction at hotspots and we found no additional papers . Nonetheless, it is possible that some papers may have been missed. Publication biases may have operated, such that positive findings about particular interventions were more likely to have appeared in print than negative ones. It is also worth noting, however, that there would be instances where interventions have been implemented, with an effect, and not reported. This is particularly likely to be the case with barriers on bridges, which are not uncommon and may lead to reductions in suicides, but may not be reported in the scientific literature.
It was not always possible to determine the exact nature of the intervention. This was not so much a limitation of the review per se, but rather a limitation of the original studies. This limitation may have been a particular problem for the studies of restricting access to means, which tended to be reported as evaluations of a stand-alone intervention (the installation of barriers). Of these nine studies, only the one by Bennewith and colleagues [17, 18] provided detail of additional activities that were put in place alongside the barriers on the Clifton Suspension Bridge. A commentary on the study by Sinyor and Levitt  by Sakinofsky  indicated that the barriers at the Bloor Street Viaduct were complemented by a telephone crisis service, but this was not evident from the original paper. This may not have been an isolated case. The upshot of this is that impacts may have been attributed to a single intervention that in reality were at least in part due to other activities that ran alongside it.
These limitations were compounded by more basic difficulties of definition. Although we have provided a definition of the term ‘hotspot’, in practice this may have been applied fairly loosely across studies, with the result that different authors would use different thresholds to deem a site a hotspot. Similarly, the term ‘intervention’ is not universally understood. Some of the activities classified as interventions were in fact opportunistic activities (e.g., road closures), other interventions within the same group undoubtedly varied in scope and scale, and still others, as noted above, were delivered as part of a suite. Finally, the term ‘effectiveness’ must be interpreted in the light of the evidence available, with due acknowledgement being given to the study design issues mentioned above.
Notwithstanding the above limitations, we believe that our review demonstrates that there is consistent and relatively strong evidence that reducing access to means through the installation of barriers can be effective in averting suicides at hotspots and does not lead to substitution effects. The evidence is weaker for the other approaches that have been used, namely encouraging help-seeking, increasing the likelihood of intervention by a third party, and providing guidance on responsible media reporting of suicide, although they all show promise. This picture is consistent with the broader literature on suicide prevention, which suggests that there is good evidence that restricting access to means can work and that the majority of other interventions require further testing. Our review adds to this literature by focusing on the specific suite of interventions that have been used at suicide hotspots.
There is often community resistance to restricting access to means, despite this being the approach for which there is the greatest evidence of impact. Policy-makers and planners should be encouraged to recognize the demonstrated benefits of this approach, because they will often be faced with arguments of aesthetics and up-front costs which mean that they are likely to preferentially choose the other approaches. The other approaches may be useful too, but they clearly require further testing. More well-designed intervention studies are needed to strengthen this evidence base.
Beautrais AL: Suicide by jumping: A review of research and prevention strategies. Crisis. 2007, 28 (Suppl. 1): 58-63.
Owens C, Lloyd-Tomlins S, Emmens T, Aitken P: Suicides in public places: Findings from one English county. Eur J Public Health. 2009, 19 (6): 580-582. 10.1093/eurpub/ckp052.
Blaustein M, Fleming A: Suicide from the Golden Gate Bridge. Am J Psychiatry. 2009, 166 (10): 1111-1116. 10.1176/appi.ajp.2009.09020296.
Glasgow G: Do local landmark bridges increase the suicide rate? An alternative test of the likely effect of means restriction at suicide-jumping sites. Soc Sci Med. 2011, 72 (6): 884-889. 10.1016/j.socscimed.2011.01.001.
Cantor CH, Hill MA: Suicide from river bridges. Aust N Z J Psychiat. 1990, 24 (3): 377-380. 10.3109/00048679009077705.
Kerkhof A: Railway suicide: Who is responsible?. Crisis. 2003, 24 (2): 47-48.
De Leo D, Krysinska K: Suicidal behaviour by train collision in Queensland, 1990–2004. Aust NZ J Psychiat. 2010, 42 (9): 772-779.
Caldwell T, Jorm A, Dear K: Suicide and mental health in rural, remote and metropolitan areas in Australia. Med J Australia. 2004, 181 (7 Suppl): s10-s14.
Elnour A, Harrison J: Lethality of suicide methods. Inj Prev. 2007, 14: 39-45.
Tranah T, Farmer R: Psychological reactions of drivers to railway suicide. Soc Sci Med. 1994, 38 (3): 459-469. 10.1016/0277-9536(94)90448-0.
Yum B, Roh J, Ryu J, Won J, Kim C, Lee J, Kim K: Symptoms of PTSD according to individual and work environment characteristics of Korean railroad drivers with experience of person-under-train accidents. J Psychosom Res. 2006, 61 (5): 691-697. 10.1016/j.jpsychores.2006.07.005.
Hamilton S, Metcalfe C, Gunnell D: Media reporting and suicide: A time-series study of suicide from Clifton Suspension Bridge, UK, 1974–2007. J Public Health. 2011, 33 (4): 511-517. 10.1093/pubmed/fdr043.
National Institute for Mental Health in England: Guidance on Action to be taken at Suicide Hotspots. 2006, London: National Institute for Mental Health in England
Moher D, Liberati A, Tetzlaff J, Altman D: Preferred Reporting Items for Systematic Reviews and Meta-Analyses: The PRISMA Statement. PLoS Med. 2009, 6 (6): e1000097-
Beautrais AL: Effectiveness of barriers at suicide jumping sites: A case study. Aust NZ J Psychiat. 2001, 35 (5): 557-562. 10.1080/0004867010060501.
Beautrais AL, Gibb SJ, Fergusson DM, Horwood LJ, Larkin GL: Removing bridge barriers stimulates suicides: An unfortunate natural experiment. Aust NZ J Psychiat. 2009, 43 (6): 495-497. 10.1080/00048670902873714.
Bennewith O, Nowers M, Gunnell D: Effect of barriers on the Clifton suspension bridge, England, on local patterns of suicide: Implications for prevention. Br J Psychiatry. 2007, 190: 266-267. 10.1192/bjp.bp.106.027136.
Bennewith O, Nowers M, Gunnell D: Suicidal behaviour and suicide from the Clifton Suspension Bridge, Bristol and surrounding area in the UK: 1994–2003. Eur J Public Health. 2011, 21 (2): 204-208. 10.1093/eurpub/ckq092.
Etzersdorfer E, Sonneck G: Newspaper reports and suicide. N Engl J Med. 1992, 327: 502-503.
Etzersdorfer E, Sonneck G: Preventing suicide by influencing mass-media reporting: The Viennese experience, 1980–1996. Arch Suicide Res. 1998, 4: 67-74.
King E, Frost N: The New Forest Suicide Prevention Initiative (NFSPI). Crisis. 2005, 26 (1): 25-33.
Isaac M, Bennett J: Prevention of suicide by jumping: The impact of restriction of access at Beachy Head, Sussex during the foot and mouth crisis 2001. Public Health Med. 2005, 6 (1): 19-22.
Law CK, Yip PSF, Chan WSC, Fu K-W, Wong PWC, Law YW: Evaluating the effectiveness of barrier installation for preventing railway suicides in Hong Kong. J Affect Disorders. 2009, 114 (1–3): 254-262.
Lester D: Suicide from bridges in Washington, DC. Percept Motor Skills. 1993, 77 (2): 534-10.2466/pms.19220.127.116.114.
Lester D: Suicide by jumping from bridges. Percept Motor Skills. 2005, 100 (3 (Pt 1)): 628-
Niederkrotenthaler T, Sonneck G: Assessing the impact of media guidelines for reporting on suicides in Austria: Interrupted time series analysis. Aust N Z J Psychiatry. 2007, 41 (5): 419-428. 10.1080/00048670701266680.
O’Carroll P, Silverman M: Community suicide prevention: The effectiveness of bridge barriers. Suicide Life Threat. 1994, 24 (1): 89-91. discussion 91–89
Pelletier AR: Preventing suicide by jumping: The effect of a bridge safety fence. Inj Prev. 2007, 13 (1): 57-59. 10.1136/ip.2006.013748.
Reisch T, Michel K: Securing a suicide hot spot: Effects of a safety net at the Bern Muenster Terrace. Suicide Life Threat. 2005, 35 (4): 460-467. 10.1521/suli.2005.35.4.460.
Sinyor M, Levitt AJ: Effect of a barrier at Bloor Street Viaduct on suicide rates in Toronto: natural experiment. BMJ. 2010, 341: c2884-10.1136/bmj.c2884.
Skegg K, Herbison P: Effect of restricting access to a suicide jumping site. Aust N Z J Psychiatry. 2009, 43 (6): 498-502. 10.1080/00048670902873698.
Sonneck G, Etzersdorfer E, Nagel-Kuess S: Imitative suicide on the Viennese subway. Soc Sci Med. 1994, 38 (3): 453-457. 10.1016/0277-9536(94)90447-2.
Wong PWC, Liu PMY, Chan WSC, Law YW, Law SCK, Fu K-W, Li HSH, Tso MK, Beautrais AL, Yip PSF: An integrative suicide prevention program for visitor charcoal burning suicide and suicide pact. Suicide Life Threat Behav. 2009, 39 (1): 82-90. 10.1521/suli.2009.39.1.82.
Mann JJ, Apter A, Bertolote J, Beautrais A, Currier D, Haas A, Hegerl U, Lonnqvist J, Malone K, Marusic A: Suicide prevention strategies: A systematic review. J Am Med Assoc. 2005, 294 (16): 2064-2074. 10.1001/jama.294.16.2064.
Beautrais A, Fergusson D, Coggan C, Collings C, Doughty C, Ellis P, Hatcher S, Horwood J, Merry S, Mulder R: Effective strategies for suicide prevention in New Zealand: A review of the evidence. NZ Med J. 2007, 120 (1251): U2459-
Gunnell D, Nowers M: Suicide by jumping. Acta Psychiatr Scand. 1997, 96 (1): 1-6. 10.1111/j.1600-0447.1997.tb09897.x.
Hawton K: Restricting access to methods of suicide: Rationale and evaluation of this approach to suicide prevention. Crisis. 2007, 28 (Suppl. 1): 4-9.
Yip P, Caine E, Yousuf S, Chang S-S, Wu K, Chen Y-Y: Means restriction for suicide prevention. Lancet. 2012, 379: 2393-2399. 10.1016/S0140-6736(12)60521-2.
Glatt K, Sherwood D, Amisson T: Telephone help lines at a suicide site. Hosp Community Psych. 1986, 37 (2): 178-180.
Michel K, Frey C, Wyss K, Valach L: An exercise in improving suicide reporting in print media. Crisis. 2000, 21: 71-79.
Michel K, Maillart A, Reisch T: Monitoring of suicide reporting in print media 10 years after the publication of media guidelines: Did you expect anything else?. International Association for Suicide Prevention XXIV World Congress - Preventing Suicide Across the Lifespan. 2007, Killarney, Ireland: Dreams and Realities
Pirkis J, Dare A, Blood RW, Rankin B, Williamson M, Burgess P, Jolley D: Changes in media reporting of suicide in Australia between 2000/01 and 2006/07. Crisis. 2009, 30 (1): 25-33.
Victora C, Habicht J-P, Bryce J: Evidence-based public health: Moving beyond randomized trials. Public Health Matters. 2004, 94 (3): 400-405.
Sanson-Fisher R, Bonevski B, Green L, D’Este C: Limitations of the randomized controlled trial in evaluating population-based health interventions. Am J Preventive Med. 2007, 33 (2): 155-161. 10.1016/j.amepre.2007.04.007.
Brousselle A, Champagne F: Program theory evaluation: Logic analysis. Eval Program Plann. 2011, 34 (1): 69-78. 10.1016/j.evalprogplan.2010.04.001.
Brown C, Lilford R: The stepped wedge trial design: A systematic review. BMC Med Res Methodol. 2006, 8 (6): 54-
NHS Health Scotland: Guidance on Action to be Reduce Suicides at Locations of Concern. 2012, Edinburgh: NHS Scotland
Lockley A, Williamson M, Robinson J, Cox G, Cheung Y-T, Grant L, Pirkis J: Preventing Suicide at Suicide Hotspots. 2012, Canberra: Australian Government Department of Health and Ageing
Pirkis J, Spittal M, Cox G, Robinson J, Cheung Y-T, Studdert D: The effectiveness of structural interventions at suicide hotspots: A meta-analysis. Int J Epidemiol. In press
Sakinofsky I: Barriers to suicide. Strategies at Bloor Viaduct. BMJ. 2010, 341: c4447-10.1136/bmj.c4447.
The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2458/13/214/prepub
This review was funded by the Australian Government Department of Health and Ageing. CO was supported during the writing of this paper by the UK National Institute for Health Research Collaboration for Leadership in Applied Health Research and Care for the Southwest Peninsula.
The authors have no competing interests to declare.
JP, GRC, CO and JR were responsible for the original conceptualisation and design of the review, and AL, MW and YTDC helped to refine its parameters. GRC conducted the literature search and retrieved all potential articles. GRC, AL and JR screened abstracts and full text articles to arrive at the final pool. GRC and AN prepared early drafts of the paper, and all authors commented on and refined these drafts. All authors read and approved the final manuscript.
Authors’ original submitted files for images
Below are the links to the authors’ original submitted files for images.