Projected deaths due to Covid-19 will far exceed suicides due to an economic recession in the US: a summary of the scientific evidence

“People get tremendous anxiety and depression and you have suicide over things like this, when you have a terrible economy, you have death…definitely in far greater numbers than we’re talking about with regard to the virus.”

White House press conference, 23rd March 2020, US President Donald Trump

In recent days, US President Donald Trump has repeatedly argued that the economic consequences of public health measures to limit the spread of Covid-19 could result in more deaths, due to suicide, than the virus itself. Existing scientific evidence does not, however, support this claim.

It is important to consider the impact that both the Covid-19 outbreak and related public health interventions will have on mental health, mental healthcare, and suicide rates. However, if no public health measures are implemented, the predicted number of Covid-19 deaths in the US over four months is nearly 500 times greater than the number of suicide deaths occurring between 2007-2010 attributable to the 2008 economic recession. The 2008 economic recession would have to last over 1,300 years, for the number of resultant suicide deaths to surpass the number of lives saved by stringent public health measures. Furthermore, evidence indicates the number of suicide deaths can be reduced with employment support measures and historically, in the US, the total number of deaths in the population falls during periods of recession.

Estimated number of suicide deaths in relation to economic recession

In the US between 2007-2010, there were on average 35,850 suicide deaths annually. An estimated 4,750 of these suicide deaths were attributed to the economic recession. Unemployment is a key mechanism through which economic recessions can contribute to increased suicide rates; the increase in unemployment from 5.8% to 9.6% post-recession (2007-2010) was associated with approximately 1,330 of the suicide deaths.

Estimated number of Covid-19 deaths without public health interventions

A key study, which modelled the impact of non-pharmaceutical measures, and led to a change in UK strategy, indicated that without any public health interventions or changes in individual behaviour there would be approximately 2.2 million deaths in the US due to Covid-19. The figures are based on age-stratified infection fatality ratio, ranging from 0.002% among 0- to 9-year olds and 9.3% for those aged ≥80. Most of the deaths are predicted to occur over approximately a four-month period, and around half would be due to healthcare services exceeding their capacity to treat everyone with the infection. In addition, indirect deaths may also occur as a result of over-stretched services. For example, previously treatable heart attacks may now result in death if cardiac care facilities are unavailable. As such, the number of Covid-19-related deaths would be far higher.

Estimated number of Covid-19 deaths with public health interventions

In the UK, case isolation, home quarantine, and social distancing are predicted to reduce the number of deaths that would have occurred with no measures in place (510,000) to approximately 90,000 (an 82% reduction), while adding school and university closure could reduce the number of deaths to around 24,000 (a 95% reduction). These calculations are based on a reproduction number of 2.4 and the triggering of social distancing and closure of educational settings when more than 200 new Covid-19 cases are diagnosed in intensive care units weekly. A similar percentage reduction in the US would equate to around 396,000 or 110,000 deaths respectively.

The wider context

Despite the implementation of a lockdown in Italy, over a six-week period the number of deaths from Covid-19 has far exceeded the 290 excess suicide attempts and deaths that occurred post-recession (2007-2010) as well as the 4,886 total suicide deaths recorded in 2016, the most recent year for which data are available.

Predicted number of suicide deaths in US if a 2008-level economic recession lasted 20 years, compared with the predicted number of Covid-19 deaths if the most stringent public health interventions are implemented, and if no interventions are implemented

A US economic recession may occur irrespective of whether strict public health measures are implemented, although the extent of the recession will likely vary dependent on approach taken. Economic recessions may be associated with a higher prevalence of mental health conditions and cuts to mental health services. In addition to a recession, there may be other mechanisms such as social isolation, by which strict public health measures contribute to increased mortality. Yet, to surpass the number of lives saved from reduced spread of Covid-19, these additional mechanisms would need to contribute to an extra 1.8 million suicide deaths on top of the number of suicides attributed to the 2008 economic recession. This is 40 times more deaths than the annual number of suicide deaths in the US.

Importantly, in contrast to rises in suicide rates, all-cause mortality decreased in the US during both the 2008 economic recession and the Great Depression (1930-33), for example due to shifts in trends in cardiovascular deaths and road traffic accidents. There is also evidence that the effect of unemployment on suicides can be reduced with more generous employment support measures (1,2,3).

There is no doubt that the impact on mental health and risk of suicide for populations, healthcare workers, individuals and their families as a consequence of the Covid-19 outbreak will be profound. It is a situation that could be worsened by a lack of appropriate, albeit extreme, mitigation measures. However, scientific evidence demonstrates that public health interventions to reduce the spread of Covid-19 should not be withheld on the basis that a resultant economic recession would cause more deaths by suicide.

Prianka Padmanathan is a MRC MARC PhD clinical fellow at the department of Population Health Sciences @_prianka_

Dee Knipe is a Vice Chancellor’s Postdoctoral Research Fellow in the Department of Population Health Science, University of Bristol @dee_knipe 

David Gunnell is Professor of Epidemiology in the Department of Population Health Science, University of Bristol. His research expertise include the effect of the 2008 economic recession on suicide rates (1, 2, 3).

Jacks Bennett is a Bristol doctoral researcher at looking at student mental health and wellbeing @JacksBennett

Group Twitter: @SASHBristol 

If you or anyone you know is affected by the issues covered in this blog, you can contact the Samaritans for free from any telephone on 116 123 in the UK. Alternatively you can email for details of your nearest branch, where you can talk to one of their trained volunteers face to face. The International Association for Suicide Prevention (IASP) has details of support organisations in other countries.

Exceptions to the ‘Werther effect’: Why Jeffrey Epstein’s death is unlikely to trigger further deaths by suicide.

By Helen Fay 

Jeffrey Epstein – a publicly disgraced financier – died by suicide while awaiting trial for charges of sex trafficking. Unsurprisingly, due to the high-profile nature of his case, reports of his suicide have since proliferated across the globe.

Widespread coverage of suicide is usually met with concerns from mental health professionals. These concerns are not unfounded – there is substantial evidence demonstrating that media reports of suicide can initiate an increase in suicide rates, a phenomenon known as the Werther effect. This effect is more pronounced following coverage of the suicide of a public figure. Accordingly, the World Health Organisation advise that:

“Vulnerable individuals may be influenced to engage in imitative behaviours by reports of suicide, particularly if the coverage is extensive, prominent, sensationalist and/or explicitly describes the method of suicide.”

Yet, it would be incorrect to conclude that all media coverage of suicide is harmful. Despite compelling evidence, research in the field is characterised by inconsistent findings, with a proportion of research showing no effect. Publication bias may overestimate the occurrence of the Werther effect. Due to legitimate challenges in designing ethically acceptable studies, most research investigating media influences on suicide rates employ an ecological aggregate design, which assesses change in suicide rates following a specific media exposure. These studies are typically undertaken post-hoc in response to known ‘epidemics’ or high-profile harmful depictions (e.g., celebrity suicide, high-lethality methods).

The variability of media content may also contribute to these inconsistent findings. There are three well-known mechanisms which encourage the Werther effect: i) the introduction of a novel method to a “vulnerable” population; ii) personal identification with the story of the deceased; and iii) the generation of acceptable impressions of suicide by the media (particularly following celebrity suicides). Focusing on the latter mechanism, the media typically frames suicide in a positive manner (e.g., by glorifying the deceased). However, the journalistic ‘norm’ of framing suicide positively could theoretically explain the consistency by which the media is associated with heightened suicide rates, rather than a direct link.

The media’s framing of Epstein’s suicide was atypical. Due to Epstein’s disgraced position, his suicide was characterised as a “cowardly” and an immoral act. Prior research indicates that negative characterisations of suicide are less likely to have deleterious effects. To illustrate, following the suicide of Robin Williams, the press celebrated him as a successful and inspirational entertainer who could not overcome depression – suicide rates following his death increased by 10% (an additional 1841 suicide deaths in the US). In contrast, suicide rates were unaffected following the death of Kurt Cobain whose suicide received widespread condemnation by the press. Quantifying this notion, Stack’s review found that stories with negative descriptions of suicide were 99% less likely to prompt suicide ‘contagion’.

As early as Durkheim’s (1897) Le Suicide, it has been recognised that cultural influences shape attitudes towards suicide. Cultural differences in attitudes correspond with variations in suicide rates, with more permissive attitudes linked to increased suicide rates. The morality attached to suicide is intrinsically linked to the media and its ability to influence public opinion en masse. Some coverage can normalise suicide as a legitimate response to distress. However, Epstein’s alleged crimes – paedophilia and sex trafficking – are almost universally met with societal contempt. Consequently, Epstein, and the act of suicide were typified as immoral, and are unlikely to prompt suicide suggestion. Indeed, media coverage of the suicides of those in disrepute, such as criminals, are rarely accompanied by an increase in suicide rates.

Rather than viewing all coverage of suicide as detrimental, it is important to consider the specific characteristics of news coverage which could encourage suicidal behaviour. In reference to the coverage of Epstein’s death, the negative framing of his suicide, together with an immoral character appraisal could negate the potential harms associated with widespread news attention. Despite this, it would be extremely contentious to suggest that newspapers should frame suicide as an immoral act. The coverage of Epstein’s death was devoid of compassion towards the bereaved, who are known to be particularly vulnerable following a suicide. Instead, journalists should strike a balance between displaying compassion towards the bereaved and avoiding romanticised depictions of suicide, including those which sanctify the deceased.

A Note on Conspiracy Theories:
Numerous conspiracy theories contend that Epstein’s death cannot be explained by suicide. Despite this, news reports on the topic still present a wide-reaching narrative which frames suicide as unacceptable. Conspiracy stories do not necessarily counteract the influence of reporting on suicide. Suicide rates increased following Marilyn Monroe’s (12% in the US, 9% in the UK) and Robin Williams’ (9.8% in the US) suicide deaths, despite a body of conspiracy theories circulating which disputing suicide as their cause of death.

Key things to remember when reporting on suicide:
1. Think about the impact of the coverage on your audience, particularly vulnerable individuals.
2. Exercise caution when referring to the methods and context of a suicide.

– Avoid giving details of the suicide method, particularly novel methods.
– Remember that there is a risk of imitational behaviour due to ‘over-identification’.
– Never say a method is quick, easy, painless or certain to result in death.
3. Avoid over-simplification of the causes of suicide.
4. Steer away from melodramatic depictions of suicide or its aftermath

-Don’t over emphasise community expressions of grief. Doing so may suggest that people are honouring the suicidal behaviour rather than mourning a death.
5. Aim for non-sensationalising, sensitive coverage
– Don’t refer to ‘hotspots’ or ‘epidemics’
– Don’t promote the idea that suicide resolves problems or achieves results.
– Don’t report on contents of a suicide note.
5. Aim for sensitive, non-sensationalising coverage

Abridged from Samaritan’s (2013) Media Reporting guidelines

Helen Fay is a PhD Researcher in Population Health Science, Bristol Medical School. Her PhD investigates media influences on decision-making pertaining to suicide methods.

You can follow @HelenAmyFay and @SASHBristol on Twitter.

If you or anyone you know is affected by the issues covered in this blog, you can contact the Samaritans for free from any telephone on 116 123 in the UK. Alternatively you can email for details of your nearest branch, where you can talk to one of their trained volunteers face to face. The International Association for Suicide Prevention (IASP) has details of support organisations in other countries.

The language of suicide – time to pause and think?

In a recent open letter, campaigners have urged the media to lead the way in using sensitive language when it comes to talking about suicidal behaviour. A similar request has been put forward to researchers and health professionals. There have been several phrases that have been suggested to be unhelpful (primarily from anecdotal accounts from people with lived experience), but the phrase that has gained the most disapproval is “commit suicide”.

You might be asking yourself, “what is the big deal?”. I certainly did when I first started writing about suicide and was consistently corrected. After all commit is a verb indicating that a person was determined to carry out a certain course of action. Yet I soon realised the most common use of the verb commit is in relation to someone committing a crime or an immoral act. People don’t “commit suicide” (i.e. a crime). Suicide has been decriminalised in the majority of countries worldwide. But this phrase has unhelpful undertones, which might further stigmatise this behaviour. With this understanding I’ve been in support of any drive to encourage more appropriate language use, and with the help of my colleagues have been actively encouraging academic journals and conference organisers to dissuade professionals using this language.

As professionals we don’t want to further stigmatise suicide, and as language matters we should be trying to help reduce the associated stigma. Stigma, we know, is a barrier for accessing health care, so in my opinion any effort to reduce stigma should be pursued. Some argue, however, that our push to move away from the term “commit suicide” is over simplistic, and that changing language does not change reality. The evidence being quoted here is weak, and in my view the question remains, does changing our language help? Does it reduce stigma? We simply don’t know. But a small change may help. Getting people to question their language brings into their consciousness the importance of the words they use, and the impact their words might have on others.

An argument has however been made for not moving away from the term “commit suicide”. Perhaps the alternative phrases (i.e. “die by suicide”) remove a person’s agency. I can see this argument, and I know that some use the term “commit suicide” without any understanding of implications of criminality. Take me for example, I certainly didn’t know about the link. But once I did, there was no going back. A friend of mine died by suicide in 2016, and for me reading about him “committing suicide” in the media felt wrong.

For professionals there are two things to consider. The language used to talk to an individual regarding an experience (in this case related to suicide) should be guided by the language that the person feels comfortable using. This should be assessed on a case by case basis. However, when talking to the public about suicide, professionals should use the language that is most acceptable to the greatest number of people. But here is where we stumble. At present we don’t know what phrase the majority of people with a lived experience of suicide find the most sensitive and appropriate. There simply is no empirical evidence to guide us. So perhaps, this is the time to push the pause button on the current drive for us to drop the term “commit suicide”.

As a researcher, evidence-based decision-making is important to me. With the help of fellow colleagues at the University of Bristol, University of Nottingham and the Samaritans, we launched an anonymous online survey for those who have experience of suicide. We hope that this evidence will help inform the debate and plan to release the results of this survey as soon as possible. A copy of the report will be hosted on this website, so stay tuned!

Dee Knipe is a Vice Chancellor’s Postdoctoral Research Fellow in the Department of Population Health Science, University of Bristol. She is an an epidemiologist with a special interest in suicide and self-harm in low and middle income countries.

You can follow @dee_knipe (group account @SASHBristol ) on Twitter.

If you or anyone you know is affected by the issues covered in this blog, you can contact the Samaritans for free from any telephone on 116 123 in the UK. Alternatively you can email for details of your nearest branch, where you can talk to one of their trained volunteers face to face. The International Association for Suicide Prevention (IASP) has details of support organisations in other countries.

Student mental health, suicide and media reporting

By Professor David Gunnell 

In the last 18 months, Bristol has been in the eye of a storm of concern about student mental health following the tragic deaths by suicide of a number of our students.

First and foremost our thoughts are with the families and friends of all those who have lost a loved one. They will quite rightly want to know what could have been done to prevent each death. It will be important that lessons are learnt and widely shared across the university sector.

What do we know about the mental health of the UK’s 2.3 million students and what lies behind the recent tragic deaths? It is widely recognised that levels of depression and self-harm in young people are increasing in the UK and internationally. UUK’s recent analysis highlighted a six-fold rise in the number of UK students disclosing a mental health condition to their university since 2007. Data recently presented to the National Suicide Prevention Advisory Group show rises in suicides in 15-19 year olds in England.  The rises in levels of distress seen by UK universities likely mirror those seen in the wider population of this age. There are several possible additional contributors to student mental health problems. These include:

  • Concerns about the high levels of debt students incur because of course fees in excess of £9,000, in addition to their costs of living. This increases the pressure to perform well at university
  • Difficulties establishing new friendship groups and support networks as they transition to a new geographic location after leaving home for the first time
  • Loss of continuity of medical and mental health care for those previously under specialist care for mental health conditions in their home town
  • Loss of the sense of community provided by friends, families and schools, with students being increasingly able to access course materials on-line and so avoid attending lectures. Furthermore, pressures on university finances have resulted in larger year-group sizes and an associated sense of anonymity. When I studied medicine at Bristol in the 1980s my year group size was 120; now it is closer to 300.
Photo by Davide Cantelli

Concern about student suicide and mental health dates back at least to the 1950s, following the identification of a high incidence of suicide amongst Oxford undergraduates. In 2015-16 York University was the focus of similar concerns and informal conversations with colleagues elsewhere in the UK points to similar worries about student suicides.

The research literature highlights “clusters” of student suicide deaths in the USA and elsewhere around the world. Such clusters have also been reported in secondary schools and communities.

We don’t know what causes these clusters. Psychological theories suggest risk may “spread” by social contagion, social learning, suggestion or imitation.  But research such as Madelyn Gould’s elegant analysis of clusters in the USA tells us clusters are more common in young people and may be precipitated by news reporting. Reporting practices were recently criticised in relation to another UK cluster. For this reason Public Health England’s guidance for media reporting in relation to suicide clusters makes 10 recommendations (see page 24); these include:

  • avoid re-running details of each death in every report, re-reporting previous stories and making links to other suicides
  • do not give undue prominence to a story, such as front cover splash and dramatic headlines and use of photographs and memorials of people who have died – specifically repeated use of image galleries should be avoided

Whilst most news organisations, including our student news groups, have followed this guidance, several, most recently the Sun (Saturday May 19th) have not. Some media organisations have repeatedly published picture galleries of the students who have died, taken in happier times, each time a further death has occurred. These may re-trigger grief in family and friends recovering from their loss. They may make other students feel vulnerable – “if this happy looking person took their own life, then am I too at risk?” It is possible that news coverage may have contributed to some deaths.

Photo by Faustin Tuyambaze

Research suggests that the proportion of students experiencing suicidal thoughts in a 12 month period is around 17%, and 9% have made suicide plans. If these thoughts and plans occur in an environment where models of people taking their lives in similar situations exist, suicide may become a more tenable option.

Furthermore, news reports linking the suicides with long waiting times for support and inadequate service provision, whilst rightly shining a light on important challenges across the university sector, may also deter people from seeking help, the last thing we wish to happen at times like this.

The impact of the deaths on students is exemplified by the 60% increase in referrals for counselling following deaths in Bristol in 2017. These will reflect both heightened concerns amongst students themselves as well as the staff they interact with.

Support, administrative, teaching and other academic staff too are profoundly affected by the death of a student in their care.

These are tough times for young people and universities. We need to better understand the most effective approaches to improving student mental health and support those experiencing difficulties, as well as learn lessons from the recent deaths. Furthermore, we need to understand the drivers behind recent rises in young people’s distress to inform prevention. We also need to review reporting guidance. Perhaps it is time the Independent Press Standards Organisation’s (IPSO) code of practice for reporting on suicide deaths is extended to include avoiding repeatedly publishing pictures of those who have died.

David Gunnell is Professor of Epidemiology in the Department of Population Health Science, University of Bristol. He has a long-standing research and policy interest in the epidemiology and prevention of suicide in the UK and internationally.

You can follow  @SASHBristol on Twitter.

If you or anyone you know is affected by the issues covered in this blog, you can contact the Samaritans for free from any telephone on 116 123 in the UK. Alternatively you can email for details of your nearest branch, where you can talk to one of their trained volunteers face to face. The International Association for Suicide Prevention (IASP) has details of support organisations in other countries.

Can suicide media awareness campaigns sometimes backfire?

By Professor David Gunnell

There have been some truly moving suicide prevention media campaigns recently. For example, Samaritans “Small Talk Saves Lives” campaign tells the story of a woman whose life was saved (from suicide) because someone took the time to talk to her when she was planning to kill herself.

Project 84: CALM’s recent high-profile campaign, based around the sobering statistic that 84 men die by suicide in the UK every week, however, has raised concerns amongst a number of people in the suicide prevention community.

The campaign, which tells the real life stories of 84 men who died by suicide, was launched with a haunting display of 84 life-sized statues of men on the edge of the roof top of ITV studios in London. Friends and families of the bereaved, working alongside a sculptor took part in making the statues. The display reminds one of the moment before someone jumps from a high building.

Over the years CALM have done a huge amount of work raising awareness about the major public health problem of male suicide. However this particular campaign raises concerns. There is a large body of research evidence concerning the impact of reporting / media portrayal of suicide methods on subsequent imitative suicides and the uptake of suicide by populations.

Photo by Mitchell Hollander on Unsplash

It is thought that raising awareness of suicide methods can increase their “cognitive availability” and might influence method choice. A number of research studies – including work by @SASHBristol’s Dr Lucy Biddle – have drawn attention to how the media (print / on-line) may influence people’s preference for particular suicide methods and provide them with knowledge about how to use these. And if this influences them towards more lethal methods, this may result in a rise in suicide rates. Most people who survive a suicide attempt do not repeat the act. Importantly, people don’t have to be suicidal to be affected by the images and reporting. Information may be stored away at the back of someone’s mind and perhaps retrieved later at the time of a suicidal crisis.

Pictures of the statues in the Project 84 campaign have been widely reported by news websites in the UK. The display, which is on the top of a tall building in central London, is highly visible. Samaritans and WHO guidelines on responsible reporting of suicide are clear. Samaritans guidelines state “Avoid giving too much detail. Care should be taken when giving any detail of a suicide method” .

A series of recent studies have shown rises in method-specific (and sometimes overall) suicide incidence may follow high profile media attention given to a particular method of suicide, most often following a celebrity’s suicide. A recent research paper indicated that in the 6 months after Robin Williams’ death by suicide there was a 9.8% rise in suicide deaths (1841 additional fatalities) in the USA over and above expected levels based on recent trends. The excess in deaths was largely due to a rise in deaths using the very method used by Williams.

Jumping from a tall building is a highly dangerous suicide method (and in some parts of the world is one of the most frequently used methods). Here in the UK, suicide deaths by jumping account for approx. 250 deaths per year (around 4% of all suicides). The risk of the Project 84 campaign lies in increasing awareness of this method in the wider population. More deaths might follow as a result of increasing the “cognitive availability” of the method. Indeed these concerns led Australia’s Mindframe Media Advisory Group, putting out a call to Australia’s media advising against sharing imagery from Campaign84.

In many parts of the world suicide prevention teams work closely with local media to reduce risk associated with their reports of deaths from local landmark sites. For the same reason, here in Bristol, we work with our local media on their reporting of suicide deaths from a local site which has in the past often attracted considerable publicity. Of course, it is possible that some reports / approaches to reporting deaths by particular methods may actually deter some people from using those methods. One of our PhD students, Helen Fay, is working on this very issue.

Whilst it’s good to see the important issue of male suicide highlighted, it would be tragic if it were to inadvertently backfire and lead to a rise in deaths by jumping. Of course, as with any public health intervention such as a media campaign, we must careful weigh up the benefits and potential risks. Research on public awareness campaigns is in its infancy. Professor Jane Pirkis and team’s recent systematic review   identified relatively few studies that looked at their impact on the incidence of suicide. Whilst there is some evidence of their effectiveness the authors caution about the importance of taking care “to ensure that campaign developers get the messaging of campaigns right, and further work is needed to determine which messages work and which ones do not, and how effective messages should be disseminated.”

David Gunnell is Professor of Epidemiology in the Department of Population Health Science, University of Bristol. He has a long-standing research and policy interest in the epidemiology and prevention of suicide in the UK and internationally.

You can follow  @SASHBristol on Twitter.

Thanks to @dee_knipe @LucyABiddle @HelenAmyFay for their input

If you or anyone you know is affected by the issues covered in this blog, you can contact the Samaritans for free from any telephone on 116 123 in the UK. Alternatively you can email for details of your nearest branch, where you can talk to one of their trained volunteers face to face. The International Association for Suicide Prevention (IASP) has details of support organisations in other countries.