Said Shahtahmasebi, Ph.D.

James Hollings article on suicide reporting rules (The Press http://www.stuff.co.nz/the-press/opinion/perspective/8909431/Suicide-reporting-rules-under-review, 12/7/2013) refers to the Vienna subway built in the late seventies which subsequently became a preferred method of committing suicide. It reports that after a group of researchers and the media got together and stopped detailed reporting of each suicide the use of the subway for committing suicide dropped by 75%. The literature (e.g. see 1) on this topic use the terms suicide, suicide attempts, suicide using subway and total suicide interchangeably.The literature suggests that beginning early 1984 suicide cases using the subway method increased until June 1987 when the media stopped detailed reporting of suicide, and directly linked the drop in suicides with this method of media reporting.  These conclusions are misleading and a mis-representation of the data, see Figure 1 below. It should be noted that by drop in suicide numbers the authors of the research and the Press article are referring “the use of the subway method to commit suicide”, but is there any evidence to link this to methods of reporting and a decline in suicide rates?

We are none the wiser. You see, suicide rates have a cyclic trend (e.g. see 2), for all we know this drop in the Vienna subway suicides may have occurred anyway, or it could have been a coincidence, especially when read in conjunction with research reporting subway suicide as a continuing problem. Furthermore, the reported massive percentage drop does not appear to have translated into a commensurate reduction in total suicides in Vienna. It is possible that suicides were completed by switching to another method or choosing an alternative method. Indeed, data on total suicides in Vienna from 1970-2012 (see Figure 1) suggests the suicide trend had turned upward well before the subway suicide became fashionable and continued to go up until 1985 when the downturn occurred, at least two years before the change in method of media reporting in June 1987. Yet, the Press article called it the most conclusive study to date.

Figure 1. Number of deaths by suicide and self-harm in Vienna (source: http://www.statistik.at/web_en/statistics/health/causes_of_death/index.html)

 
One of the biggest problems is uncritical research, uncritical use of the literature, and uncritical reporting. The Press article also refers to Prof Fergusson and Beautrais and the Canterbury suicide project which apparently was one of the main sources of informing policy. This project is flawed and poorly conceived and not only does it not address bias but it introduces additional bias without discussing or controlling for them.

The problem with suicide research is that death is not a recurrent event and the main informant cannot provide information, this has led to a great deal of mis-information without any evidential support (e.g. see 2). For example, there is no statistical evidence to support a causal link between mental illness and suicide, yet, decades of mis-information has helped establish it in the public mind-set as the main cause of suicide. And when suicide research attempts to profile the suicide case based on information obtained from third parties it will be a case of mis-information supporting mis-information. This phenomenon can also be observed at coroners’ inquests, e.g. “I am desperately sad we had no insight into his mental health problem and so were not able to prevent this tragedy.” commented the family doctor of the young person who was widely reported as being popular, not having any problems, being successful at sport and academically, who had committed suicide.

What is often missing from the suicide debate is data (see 2). Only looking at suicide rates over the last few years is not informative and could well be misleading. Historical suicide data suggests a cyclic effect. One of the complications of the data is that suicide trends (age groups, male and female) do not change direction all at the same time, i.e. there is a lagging effect, so when it appears that suicide rates are going down for men it is going up for women. However, at the end of a cycle when overall suicide rates change to a downward direction the authorities and researchers jump up and down and congratulate themselves that their policies are working and that they should be given more resources to apply more of the same to other groups whose rates are going up. And, at the beginning of the cycle when suicide rates change direction to an upward trend, the authorities and researchers proclaim that suicide is a complex public health issue involving lots of social, economic and mental health factors and that they should be given more funds for further research!

The reasons why no prevention plan has worked or can bring total suicide numbers down is related to a lack of understanding of suicide and too much mis-information (see 2).

Suicide prevention is highly politicised and frequently the public are led to believe that the only solution is a medical model intervention (see 2). With the secrecy around suicide basically silencing suicide survivors the public very rarely gets exposure to suicide data. For example, anecdotal information suggests that not many people are aware that suicide is one of the biggest causes of death in New Zealand. And that out of all completed suicides only between one-quarter and one-third have had contact with psychiatric services and yet they managed to complete suicide. Suicide survivors’ stories often paint a top-down approach to an intervention which does not appear to work. So basically, there is no suicide prevention strategy, but an intervention strategy that often gives the public more of the same each year but at much higher costs both in lives lost and in monetary terms.

References
1-    Elmar Etzersdorfer, Gernot Sonneck, Newspaper Reports and Suicide, N Engl J Med 1992; 327:502-503 http://www.nejm.org/doi/full/10.1056/NEJM199208133270720
2-    Shahtahmasebi, S. 2013. “ De-politicizing youth suicide prevention.” Front. Pediatr 1(8), URL (open access, free download): http://www.frontiersin.org/child_health_and_human_development/10.3389/fped.2013.00008/abstract