Suicide reports in the news – impact of Chief Coroner's call to open up suicide reporting

ABSTRACT

Aim: In 2010, the then Chief Coroner called for an opening up of suicide reporting in the press. We aimed to discover if this controversial action influenced the nature and extent of reporting of individual suicide cases in the newsprint media.

Method: A Factiva search of news reports from high circulation newspapers was conducted over four years. We counted the number of news reports of individual suicides and examined any mention of demographics, methods, location and contributing factors.

Results: We retrieved 291 individual suicide reports, 161 before and 130 after the Chief Coroners’ call. Reporting was more detailed after the call, with increased reporting of demographics, contributing factors and methods but less reporting of location. Most of these differences did not reach statistical significance.

Conclusion: The press did not appear to take the action of the Chief Coroner as a licence to report more frequently on individual suicide cases.

KEYWORDS:

• Suicide
• reporting
• news
• media
• coroner

Introduction

The way suicide is reported or portrayed in the media can influence suicidal behaviours among vulnerable people, through contagion, imitation and normalisation (Pirkis and Blood 2001, 2010; Stack 2003; Pirkis et al. 2006; Collings et al. 2011; Thom et al. 2012). In particular, young people and those with mental health problems may be particularly vulnerable to the way suicide is reported or portrayed (Phillips and Carstensen 1986). It is because of this that New Zealand and many other countries have media reporting guidelines the purpose of which is to provide best standard practice for reporting on suicide by journalists and related media professionals (Media Roundtable 2011; Law Commission 2014). Notably, these best practice media reporting guidelines explicitly recommend against reporting methods of suicide (Media Roundtable 2011).

In addition to media guidelines, New Zealand, unlike many other countries, has legislation governing suicide reporting in the news media (Pirkis 2010). Under the 2006 Coroners Act (2006), no one is allowed to ‘make public’ (broadcast in any media, newspaper, book, journal, magazine, sound or visual recording or internet site), without the coroner’s permission, ‘any particular’ of a suicide death other than the name, address and occupation and the fact that the death was self-inflicted (New Zealand Law Commission 2014). Other particulars of the case can be made public if the coroner believes that it is not likely to be detrimental to public safety. However, the precise nature of particulars has recently been a matter of legal debate (New Zealand Law Commission 2014).

In August of 2010, New Zealand’s Chief Coroner suggested publicly that there may be room for a ‘gentle opening up’ of the restrictions on media reporting of suicide; with the caveat that there was a need to consider all viewpoints on the matter (Todd 2010). This comment was accompanied by the release of statistical data on provisional suicides to the mainstream media, including the methods used by those who had recently died by suicide. Although described as a ‘rare move’ in some of the media coverage (3news 2011), this was not the first time data on methods had been released to the New Zealand public (Ministry of Health 2009, 2010) and commented on by the media (New Zealand Government 2009; Piper 2010), the Chief Coroner’s action became a high profile and controversial move, claimed by some to have divided the experts (ONE News 2010; Johnston 2010a; Fuller 2012). The New Zealand Associate Minister of Health at the time was quoted as saying he was ‘sympathetically disposed’ to the Chief Coroners’ comments about the need for more openness about the reporting of suicide (Johnston 2010b). In contrast, the New Zealand Prime Minister’s Chief Scientific Advisor appeared to disagree, quoted as saying ‘there is a lot of evidence … that if we go into details about youth suicide in the practical sense, there’s a high risk of copycat suicide. There is good evidence of clustering youth suicide’ (Fuller 2012). In any case, shortly after the Chief Coroner’s call for more openness, a statement was released by the New Zealand Prime Minister on the on the 25 August 2010, indicating that the government was to conduct a Law Commission Inquiry into the Coroners Act and suicide reporting guidelines.

Over the following months, there was considerable media discourse around the issue of suicide reporting, positioned in the news media as a debate about whether or not to open up the ‘taboo’ issue of suicide and whether existing media reporting guidelines were appropriate.

Given the senior judicial role of the Chief Coroner, his actions could have had considerable authoritative power, potentially influencing the nature and extent of suicide reporting in the news media from this time. For instance, the Chief Coroner’s action may have been interpreted as a license to report more frequently or fully on individual cases of suicide.

This study aimed to investigate whether the Chief Coroner’s actions had an impact on the nature and extent of a specific kind of suicide reporting in the New Zealand news print media – the reporting of individual cases of suicide. We hypothesised that after the Chief Coroner’s actions; (a) there would be an increase in the number of individual suicide cases reported (due to a more liberal approach to reporting); and (b) that there would be more detailed reporting, especially of methods (or locations that imply the method) or other important characteristics of suicide reporting.

Methods

This research was based on news items covering individual cases of suicide from the New Zealand newsprint media, published 24 months prior to and 24 months following the Chief Coroner’s press release on 10 August 2010.

Data sources

We conducted a Factiva database search of all news items (features, editorials and letters) from national (n = 1), metropolitan (n = 4) and provincial newspapers (n = 20). These newspapers have the highest coverage of the topic of suicide (McKenna et al. 2010).

Search terms

Due to the legislation governing suicide reporting in New Zealand at the time, the term ‘suicide’ is not commonly used in the media reporting of individual suicides (Te Pou 2010; New Zealand Law Commission 2014). Instead, consistent with research in this area (Pirkis and Blood 2001; McKenna et al. 2010), our search strategy relied on code phrases that implied suicide (McKenna et al. 2010). These included phrases such as the ‘police are not looking for anyone in connection to the death.’ Retrieved news items were then read in full for clarification of implied suicide. We address the possible impact our reliance on implied suicides may have had on our count of suicide reports in the discussion.

Extent of suicide reporting

Exclusions

As our focus was on news items reporting individual cases of suicide, we excluded news items that: (a) only discussed suicide in general terms as a social or public health issue; (b) contained information about several different suicides – as these focused on the discourse around suicide reporting in the media; and (c) suicide bombings. The remaining eligible news items included reports of single suicides occurring overseas or locally and suicide pacts (single incidents with more than one victim).

Duplicate reports

The same news items were often published in the same form in different newspapers. These duplicate reports were included in the count of suicide reports as it was the frequency of reporting that we were interested in rather than the number of unique suicides actually reported.

Nature of suicide reporting

The nature of suicide reporting was assessed by the mention of specific characteristics aligned with key items from existing media guidelines on suicide reporting, and consistency with those assessed in previous research on the nature of suicide reporting in the media (Thom et al. 2012a).

Demographic characteristics

Due to evidence that the risk of imitative suicidal behaviour may be exacerbated by a reader’s perception of their own similarities with the suicide (Beautrais and Fergusson 2012; Law Commission 2014), we recorded where news items mentioned the age, gender, ethnicity, labour market status and occupation of the deceased person.

Reasons or explanations

Due to media guideline recommendations not to reduce explanations for suicide to a single cause or to normalise suicide as a response to adversity, we documented any reported reasons or ‘explanations’ for the suicide. These included some of the more commonly known risk factors such alcohol and illicit drug use or problems, mental illness, prescription medication, criminal or legal difficulties, financial problems, relationship difficulties (including relationship termination or breakdown, and ‘other’ risk factors (‘other’ included other physical health problems, alleged failure of health services to keep in contact with a patient and any other reason not listed above).

Methods and location when the method is implied

As naming the method or location of suicides at landmarks such as bridges and parks can increase the risk of repeat suicides (Law Commission 2014), we recorded any mention of methods and location of the fatal act or where the body was found where this implied the method. For example, a news item might say the person was last seen at Auckland Harbour Bridge, or their body was found at the bottom of a named cliff or particular railway. We excluded private locations, such as ‘home’ or ‘garage’ and vague location descriptions such as ‘along State Highway 1.’

Statistical analysis

Analyses were conducted in R 3.4 (R Institute, Vienna, Austria) https://www.r-project.org/about.html. Proportions of reports mentioning a specific characteristic are reported by study period (before or after the Chief Coroner’s action), with 95% confidence intervals (CI). Risk ratios were calculated for each characteristic to examine whether the proportion of reports mentioning a factor had changed between the two periods (reported as risk ratio [RR] and 95% CI). The absolute count of suicide reports in each period was contrasted using a Poisson distribution on the absolute number of reports in each period and is reported as the ratio of reports in the second period relative to the first period (with 95% CI).

Results

We retrieved 291 news items meeting our inclusion criteria. Table 1 shows the nature and extent of suicide reporting in the two-year periods before and after the Chief Coroner’s release of data on suicide methods.

Table 1. Nature and extent of suicide reporting in NZ news media before and after the Chief Coroner’s action.

Factors reported in news items	Period 1 (before)		Period 2 (after)		Risk ratio^a (95%CI)
Demographics reported	n	% (95% CI)	n	% (95% CI)
Age	117	72.7 (65.1, 79.4)	105	80.8 (72.9, 87.2)	1.11 (0.98, 1.26)
Gender	148	91.9 (86.6, 95.6)	125	96.2 (91.3, 98.7)	1.05 (0.99, 1.11)
Ethnicity	2	1.2 (0.2, 4.4)	2	1.5 (0.2, 5.4)	1.24 (0.18, 8.67)
Unemployment	4	2.5 (0.7, 6.2)	6	4.6 (1.7, 9.8)	1.86 (0.54, 6.44)
Occupation	40	24.8 (18.4, 32.3)	40	30.8 (23.0, 39.5)	1.24 (0.85, 1.80)
Risk factors reported
Alcohol/drug	6	3.7 (1.4, 7.9)	11	8.5 (4.3, 14.6)	2.27 (0.86, 5.97)
Mental illness	26	16.1 (10.8, 22.8)	28	21.5 (14.8, 29.6)	1.33 (0.82, 2.16)
Prescription medication	4	2.5 (0.7, 6.2)	9	6.9 (3.2, 12.7)	2.79 (0.88, 8.84)
Criminal/legal difficulties	14	8.7 (4.8, 14.2)	21	16.2 (10.3, 23.6)	1.86 (0.98, 3.51)
Financial	6	3.7 (1.4, 7.9)	3	2.3 (0.5, 6.6)	0.62 (0.16, 2.43)
Relationship difficulties	8	5.0 (2.2, 9.6)	22	16.9 (10.9, 24.5)	3.41 (1.57, 7.40)
Other^b risk factors	26	16.1 (10.8, 22.8)	35	26.9 (19.5, 35.4)	1.67 (1.06, 2.62)
Other information reported
Method	43	26.7 (20.1, 34.2)	45	34.6 (26.5, 43.5)	1.30 (0.91, 1.84)
Location of body (where method implied)	44	27.3 (20.6, 34.9)	19	14.6 (9.0, 21.9)	0.53 (0.33, 0.87)
Total news items	161	100	130	100

^aRisk ratio is for probability of including this detail in the suicide reports in period 2 relative to the same detail in suicide reports in period 1.

^bIncluded other physical health problems, alleged failure of health services to keep in contact with a patient and any other reason not listed above).

In terms of the extent of suicide reporting, there was a non-significant 19.3% decrease (ratio of counts = 0.807, 95% CI 0.636, 1.02, p = 0.078) in suicide reporting, with 161 news items retrieved in the two years leading up to the Chief Coroner’s actions and 130 in the two years after.

There was an increase in the reporting of demographic characteristics of the deceased (age, gender, ethnicity, unemployment and occupation) in the second time period, although no changes for these characteristics reached statistical significance.

There was a suggestion of increased reporting of some factors contributing to the suicides. Increases were observed for mentions of relationship difficulties and the residual ‘other’ risk factor categories (RRs and CIs above 1).

Given the evidence of the potential adverse effects of reporting suicide methods, we were particularly interested in whether there was any change in the reporting of methods between the two time periods. There was a non-significant increase in the proportion of articles reporting method (RR 1.30, 95% CI 0.91–1.84) in the second time period, and the reporting of location of the body was significantly less commonly reported in period 2 (RR = 0.53, 95% CI 0.33–0.87).

Discussion

The decline in the number individual suicides reported in the newsprint media after the Chief Coroner’s call to open up the issue of suicide reporting is the opposite of what we hypothesised. However, as expected, we found more detailed reporting of important qualitative features of individual cases of suicide (demographic characteristics, contributing factors and methods). Our findings were not statistically significant. However, they are suggestive that giving prominence to the information and the clear endorsement of such practise by an authority figure such as the Chief Coroner did not change this form of media practice in a way that was likely to be harmful.

One possible explanation for the lowered reporting of individual cases of suicide may have been its displacement with media coverage of the overall debate about suicide reporting that occurred at the same time in the context of the ongoing Law Commission Inquiry into the Coroners Act and suicide reporting guidelines. We found 30 articles covering the debate over suicide reporting in the media in the six months following the Chief Coroner’s actions. If this is correct, once the coverage of the debate ceased, one would expect the reporting of unique cases of suicide would again increase. Further observation of newsprint reporting on suicide over a longer time period would be required to determine if this is the case. It is also possible that suicide reporting may have been declining regardless of the actions of the Chief Coroner. Further analysis of more recent data would be required to assess if this is the case.

Strengths and limitations

There has only been one published study examining suicide reporting in the New Zealand news media (Thom et al. 2012). That study was based on one year of data and designed to assess adherence to media guidelines on suicide reporting. Our study was based on four years of data and, although we looked at some of the factors relevant to media guidelines on suicide reporting, our main purpose was to examine changes in suicide reporting over a period of significant change in public discourse. Unlike the earlier New Zealand study which examined multiple sources of media, our study was limited by its focus on the newsprint media. This was for practical reasons as the newsprint media is freely available via the Factiva database whereas other media sources are not. Future research to examine multiple media channels, including the internet and television, is indicated to address this limitation.

Arguably, a limitation of this research is the reliance on news items that imply suicide by the use of code phrases might have led to a more liberal count of suicide. However, this method is consistent with that used previously (Pirkis and Blood 2001; McKenna et al. 2010; Thom et al. 2012a) and is unlikely to have affected the difference in the number of suicide reports before and after the Chief Coroners’ call. Further, we checked all background text in the news items to confirm implied suicides.

Another issue is the limited ability of this research method to capture more in-depth the nature of suicide reporting, such as glorification of suicides, use of photographs and other features often referred to in media guidelines. Further research is indicated here.

Implications

Media reporting guidelines have an important place in suicide prevention claimed to be one of the most widely recognised public health approaches to reducing deaths by suicide (World Health Organization 2008, 2012, 2014). New Zealand is an internationally unique case in that at the time of this research, due to the Coroners Act, the decision over what can be reported in the media about specific cases of suicide is made by coroners. On the world stage, this makes the New Zealand situation regarding suicide reporting in the media of significant international interest.

August 2010 was a significant time for the New Zealand news media in the fraught history of the debate over suicide reporting because of the controversial nature of the Chief Coroner’s stance on the issue. Although we found no evidence that the Chief Coroner’s action in 2010 led to an increase in the reporting of cases of suicide in New Zealand, policy activity around this time was significant, with the Law Commission review of the Coroners Act and a new set of media guidelines developed in consultation with the media industry. These events have since culminated in the Coroners’ Amendment Act (2016). Restrictions remain in the amended Act around the reporting of method or detail that suggests the method, although the Act has been reworded with the word ‘particular’ removed. Additionally, there is now a process to apply for exemption from the media restrictions. Also of significance is the change under Section 71 of the Coroners Act, that although no person is allowed to describe the death as a ‘suicide’ unless granted an exemption under Section 71 (or when the coroner has completed a certificate of findings under Section 94 stating that the death was a suicide), the phrase ‘suspected suicide’ can be used. This is likely to alter the nature of suicide reporting and research documenting it, from that point forward.

Continued media coverage suggests that suicide remains a newsworthy issue in New Zealand (Johnston 2016) and in 2017 suicide became a significant election issue. We expect that cases of suicide will continue to be reported in the media, and due to the recent changes in legislation there appears to be scope for monitoring of suicide reporting to determine if there have been any further changes in the way it is reported.

Conclusion

While there were some observable differences in the nature of news reports about suicide deaths, the newsprint media did not appear to take the action of the Chief Coroner as a licence to report more frequently on suicide cases, and surprisingly the frequency of individual reports of suicide declined.

Disclosure statement

No potential conflict of interest was reported by the authors.

Additional information

Funding

This research was funded by a University of Otago Research Grant.