COVID-19 and suicide: Just the facts. Key learnings and guidance for action

1. Introduction

In an international healthcare crisis such as the COVID-19 pandemic, real-time dissemination of accurate information becomes critical in order to enable healthcare and policy decision-making in a situation of urgency with substantial uncertainty. Amidst an atmosphere of fear, rife with wild speculation and rumor, information is vital; when gathered and processed more rapidly, however, its quality can be compromised. In several prior editorials, we have shared our Journal’s challenges in balancing the twin imperatives of timeliness (expedited reviews) and scientific rigor (managing contributions with new and less certain information) with regard to our responsibility in this effort (Tandon, 2020a; and 2020b). In this editorial, I briefly examine how our field has addressed the topic of suicide in the context of COVID-19 and what we have learned about rates/patterns of suicide during the 16 months of the pandemic thus far. I present guidance from an international collaborative focused on this problem (quoting key excerpts of a letter that was sent to the Journal editor; Knipe et al., 2020) and suggest some key learnings.

2. The fear of increasing rates of suicide during pandemic- the alarm sounds

At the onset of the pandemic, several mental health experts warned the world to prepare for a concurrent increase in rates of suicide using phrases such as “a tsunami of suicide”, “dual pandemic of suicide and COVID-19″, and “suicide mortality and COVID-19- a perfect storm” (Banerjee et al., 2021; McIntyre and Lee, 2020; Reger et al., 2020; Thakur and Jain, 2020). International organizations stepped in with recommendations as to how to address this impending public health challenge (Gunnell et al., 2020; Wasserman et al., 2020).

3. What actually happened to suicide rates during the COVID-19 pandemic?

In the first few months of the viral pandemic, there were scattered media reports of individual instances of suicide ostensibly related to the effect of COVID-19 and several articles published in scientific journals that principally relied on such media reports (e.g., Mamun and Griffiths, 2020; Rajkumar, 2020; Syed and Griffiths, 2020). Other articles in scientific journals on the topic either based their observations/conclusions on non-databased theorizing or data derived from unrepresentative and inadequately designed cross-sectional surveys (Charlier, 2021; Inoue et al., 2020). The principal implication of these reports was that suicide rates were increasing in association with the COVID-19 pandemic, with particular attention focused on a particular group in some of these articles.

Analyses of actual suicide rates during the COVIDE-19 pandemic based on national and regional/local health statistics from across the world, however, provide a different picture. A majority of such comprehensive data-based studies report a modest reduction in total suicide rates whereas the remainder report no significant net increase in suicide rates during the first year of the COVID-19 pandemic (Behera et al., 2021; Deisenhammer and Kemmler, 2021; Ferrando et al., 2021; John et al., 2020; Kahil et al., 2021; Kim, 2021; Leske et al., 2021; Mitchell and Li, 2021; Nomura et al., 2021; Woolf et al., 2021). Whereas the net decline or no change in overall suicide rates was evident in the data presented in these reports, that observation was not mentioned in the abstract or conclusions of many of these articles. Additionally, occasional attention was drawn to real gender (Kim, 2021; Nomura et al., 2021) differences in suicide rates during the pandemic.

4. What did we learn, what should we do, and why does this matter?

When the COVID-19 pandemic began sixteen months ago, mental health advocates raised the concern about an increased risk of suicide that warranted attention. In view of the already existing significant public health problem of suicide across Asia and the rest of the world (Tandon and Nathani, 2018) and the many likely consequences of viral pandemics that are known to increase suicidality, drawing attention to this issue seemed appropriate. Fortunately, those fears were not realized and overall rates of suicide have declined or remained stable across the world during this pandemic thus far. Does our failure to correct the initial narrative of a COVID suicide pandemic matter? I believe that there are some important lessons to be learned with significant potential implications for public health policy, scientific integrity, editorial responsibility, and mental health advocacy. Allow me to share some perspectives and summarize associated minor changes in our Journal practices as a result.

4.1. What did we learn about rates of suicide during the COVID-19 pandemic (first 16 months)?

In contrast to alarmist predictions of increasing rates of suicide as a consequence of the COVID-19 pandemic, overall rates actually declined or at worst stayed stable. There was a suggestion that COVID differentially impacted rates of suicide among certain groups. For example, in contrast to decline in suicide rates reported among males, rates among females decreased to a lesser extent or increased minimally (Kim, 2021; Nomura et al., 2021). Furthermore, systematic reviews of scientific articles about suicide rates during major international respiratory outbreaks including COVID-19 (Kahil et al., 2021; Leaune et al., 2020) have noted the absence of any good data supporting an association between such pandemics and suicide or suicide-related outcomes. Data do suggest, however, an increased risk of suicide in the immediate aftermath of such pandemics.

A closer scrutiny of the published literature that have titles or abstracts that highlight “findings of increased suicide” indicates that many of these articles actually found no increase in rates of suicide or a decrease in rates of suicide although they reported otherwise. For example, in an article published in this journal, Seposo (2021) entitled their article “COVID-19 threatens decade-long suicide initiatives in Japan” and noted a 3 percent increase in monthly suicide numbers across prefectures in Japan in 2020 as compared to 2019. It should be emphasized, however, that when excess mortality statistics are reported for a given year, they are generally done in comparison to the average of the preceding five years- not just the previous year (Aron and Muellbauer, 2020). That is because of the relative stability of a five-year baseline. If one examines the data in the article by Seposo (2021), one finds that while the monthly suicide rate was 3% higher in 2020 than 2019, it was actually a little lower than the rates in 2018 and about 10 percent lower than in each of the prior three years, giving one an actual 5 percent reduction in monthly suicide rates in 2020 in comparison to the average of the prior five years.

In contrast to the absence of findings of increased rates of suicide, there were consistent findings of excess mortality in the general population that disproportionately affected those with psychiatric disorders (Giattino et al., 2020, Nemani et al., 2021). This higher mortality was related to direct and indirect effects of COVID-19 and unrelated to suicide. Because of this increased risk, it has been suggested that persons with severe psychiatric disorders should be a priority group for receiving vaccinations against COVID-19 (Komrad, 2020; Mazereel et al., 2021), but unfortunately this has not happened anywhere in the world until this time.

4.2. Why it matters. Implications for scientific publication and media coverage of topics related to suicide

There are lessons to learn both for public media and editors of scientific publications. Greater attention to quality of data, awareness of the potential adverse effects of disseminating questionable alarmist news and details about suicide, and the consequent need for closer attention to title and content of such communications is necessary. The International COVID-19 Suicide Prevention Research Collaboration (ICSPRC) is an international group of researchers and leaders from 39 countries that have sought to “share knowledge to minimize the impact of the pandemic on suicide deaths globally”. They have sent personal letters to editors and publishers of scientific journals and others about their concern pertaining to articles unhelpful to that goal, including one to me as the Editor of the Asian Journal of Psychiatry. They declined my invitation to submit a formal article/communication for publication in the journal and so, with their permission, I provide key excerpts of their letter to me (Knipe et al., 2020) below:

15/12/2020

Dear Prof Tandon:

We are writing to you on the behalf of the International COVID-19 suicide prevention research collaboration (ICSPRC) – an internal group of suicide prevention researchers and charity leaders from 39 countries. Our group aim is to share knowledge to minimize the impact of the pandemic on suicide deaths globally.

We have recently noticed an increasing number of publications related to suicide and the COVID-19 pandemic being published in your journal. We are delighted Asian Journal of Psychiatry has been able to prioritize and rapidly publish articles related to the impact of the pandemic on mental health. However, the titles and the content of some articles have given us some cause for concern.

There is a large body of research literature documenting the potentially harmful effects of news reporting of suicide deaths on population suicide rates. Concerning aspects of reporting include description of suicide methods, sensational headlines, and excessive reporting – these can lead to suicidal behavior among vulnerable people. Furthermore, associating the negative effects of the pandemic with suicidal behavior can normalize it as a way of coping at a time of crisis. A frequent source of these news stories is academics or journals promoting (e.g. through social media) and/or press-releasing findings, sometimes with sensational headlines. These may be used by individuals and/or news organizations with particular agendas which can have dangerous consequences.

In the UK, Samaritans closely monitors news reporting of suicide. They have observed that media stories linking suicide to COVID-19 mostly originate from bereaved families and mental health/suicide prevention experts and organizations releasing data and using alarmist language. In keeping with this, over the last few months have seen several sensational headlines in news stories based on reports of academic articles and wanted to alert you to the potential impact this may have on suicide risk more widely.

At this challenging time, it is vitally important that the research community does not contribute to increasing the risk of suicide in vulnerable populations. We recommend that editors consider the following points when reviewing articles or commentaries about suicidal behavior during the COVID-19 pandemic and its aftermath (adapted from Hawton et al. (2020) and Samaritans’ guidelines):

• 1)Remove references to methods of suicide from article titles and avoid detailed (e.g. how a ligature was attached) description of methods in the body of the article. Descriptions of a novel method of suicide should be avoided.
• 2)Avoid speculation about ‘triggers’ or cause of suicide (in this case COVID-19 and its associated public health measures). Suicide is extremely complex, and it is rarely the case that a single event or factor leads someone to take their own life. We recommend that a statement about the complexity prefaces any speculation.
• 3)Avoid sensational language, such as “surge”, “spike”, “crisis”, tsunami” and “epidemic” when describing the potential impact of the pandemic – these terms have been used out of context, generating sensational news headlines. There is currently no strong evidence of increases in suicide deaths during the first few months of the pandemic ( John et al (2020) ) .
• 4)Particular care should be taken when referring to suicidal behavior in young people, as this group is particularly susceptible to suicide contagion.

In the era of open access publishing the accessibility of academic research has improved, which increases subsequent media coverage and engagement by members of the public. This is a particular problem when articles are published as either correspondence or pre-prints as findings are yet to undergo careful scientific assessment via peer review.

We urge editors to be mindful of the potential negative impact of articles related to suicide might have on your readers because of inappropriate reporting at this time. For comprehensive guidance on reporting of suicide during the COVID-19 pandemic, please refer to The International Association for Suicide Prevention guidance.

We are very happy to answer any questions you may have about the content of this letter.

Yours faithfully:

Dr Duleeka Knipe, On behalf of ICSPRC

dee.knipe@bristol.ac.uk

Steering group: Mgs. Pablo Analuisa; Prof Louis Appleby; Prof Ella Arensman; Prof Jason Bantjes; Prof David Gunnell; Prof Keith Hawton; Prof Ann John; Prof Nav Kapur; Prof Nurad Khan; Dr. Olivia Kirtley; Dr Duleeka Knipe; Prof Thomas Niederkrotenthaler; Dr. Emma Nielsen; Prof Rory O’Connor; Prof Joseph Osafo; Prof Michael R. Phillips; Prof Jane Pirkis; Prof Steve Platt; Dr. Mark Sinyor; Prof Ellen Townsend; Dr. Lakshmi Vijayakumar; Prof Gil Zalsman.

4.3. Changes in Journal Practice

We, in the Asian Journal of Psychiatry, have learned from this experience and recognize the need to make improvements. At the outset (Tandon, 2020a and 2020b), we had recognized the importance of expeditious communication of COVID-19-mental health related information to our readership, but recognized the somewhat tenuous nature of early reports, particularly those that were reliant on media as their data source. With this in mind, we only published such observations of a preliminary or “soft” nature but with potential and therefore worthy of being shared with the field as a letter to the editor (Tandon, 2021a), hoping that the format would alert the reader to the more tenuous nature of such contributions. We recognize that this approach was less than optimal (Tandon, 2021b) and consequently will be more diligent in appraising the rigor/robustness of correspondence published in the journal and not publish any articles wholly reliant on dubious media reports. At the same time, while we recognize the potential for misinterpretation or sensationalization by others of scientific articles published in our journal, scientific rigor and provision of all necessary information needed for article appraisal by the reader will be the sole determinant of our decision-making in furtherance of our responsibility to our readership and scientific community at large.

There are some key lessons for mass media in their coverage of issues related to suicide and more broadly to their coverage of mental health topics. Their coverage of suicide in the context of COVID-19 has been less than stellar (Ganesh et al., 2020; Kar et al., 2021; Sripad et al., 2021; Menon et al., 2021b; Panigrahi et al., 2021) and a range of specific corrective remedies have been suggested (Menon et al., 2021a).

4.4. Why it matters. Implications for mental health advocacy

In a time of decreasing trust in all sources of information, it is critical that one do everything one can to preserve the trust of the scientific community, public health and political leadership, and the general public. Scientists and physician leaders are still among the most believed sources of information. The greater scrutiny that their information goes through prior to its dissemination (editorial and peer review) and the self-correcting nature of the scientific process are the key basis for this greater trust. although propagation of confusing and discrepant information by members of our community has weakened trust even among this group. We need to work harder to police ourselves and the rigor and accuracy of information that is disseminated in our name. We recognize the challenge of balancing timely sharing of potentially useful but preliminary information. We also recognize the importance of conveying public health messages in simple and actionable language. But if we are perceived as being more invested in highlighting a message we consider important at the expense of some distortion of the truth, we will only be hurting our cause. This caution is perhaps even more germane when it comes to mental health advocacy. As it is, society at large pays grossly inadequate attention to issues of mental health and stigmatizes and systematically discriminates against those with mental illnesses. Attention to the needs of those with mental illness is scant to begin with and when we distort or sensationalize presentation of relevant information, even this attention is compromised. In the context of COVID-19, there are several issues of high significance that have not received any attention; this is, in part, to the overly dramatized presentation of “increased suicide during COVID” message that is not entirely consistent with actual data.

Instead, the increase in rates of suicide rates documented in the immediate aftermath of a global infectious outbreak warrants urgent attention and preventive measures (Wasserman et al., 2020). Similarly, the higher excess (non-suicide) mortality among those with severe mental illness is a real thing (Wang et al., 2021) and requires urgent measures (such as prioritization of vaccination against COVID-19) to address this challenge. These messages will get lost if we lose public trust and are perceived as crying wolf. We have an obligation to educate the public at large about issues relevant to mental health and illness; we can do better.