Covid-19 pandemic and suicide in France: An opportunity to improve information systems.

In these exceptional pandemic times, an important question is to know to what extent this situation will have an impact on the rates of suicide, suicide attempts and suicidal thoughts. We can of course anticipate an increase in these rates due to the joint increase in many known risk factors of suicidal behaviors and thoughts: social isolation, decompensation of sleep patterns, unemployment, discontinuation of education, increase in alcohol consumption and domestic violence, difficulty in medical follow-up, etc. [1].

However, there has occasionally been observed a transient drop in suicide deaths in the aftermath of major disasters as noted in the New York area after the 09/11 attacks [2], with a sense of shared suffering. It is therefore not excluded that the “acute” period of the Covid-19 pandemic — marked by significant increases in the number of deaths and hospitalizations, containment measures and global anxiety — is not accompanied by an increase in suicidal behavior and even is characterized by a decrease in certain indicators as suggested by some preliminary field data. It is also possible that after this initial phase suicide rates will increase in a context of global economic recession as observed in 2008, notably in countries with limited buffering by social welfare systems [3].

So, rather than getting lost in speculations, the easiest way would be to turn to numbers. The problem is that the measurement of suicidal behaviors remains problematic in France, and this since well before the arrival of Covid-19. Note that this is not unique to France either, but let's start by keeping our own house in order.

The measurement of the number of suicides — estimated around 10,000 per year in France in 2016 — is dependent on the identification of the causes of death, which is first based on medical death certificates. The vast majority of these certificates are handwritten and therefore require costly time and personal input. The computerization of these certificates, which would provide fast information on mortality in general and suicide in particular, is surprisingly slow to being implemented. In addition, in the event of a legal obstacle to burial (which is the case if there is a suspicion of suicide), a police investigation and a forensic examination are conducted, the conclusions of which do not always reach the national register of causes of death managed by Inserm (these cases are then classified as undetermined causes of death). Without going into further details, let us remember that multiple administrative and bureaucratic difficulties identified for a long time are currently contributing on the one hand to significant delays before the publication of national suicide figures, currently in the order of 3 years (vs. 12 months in Great Britain for example); and on the other hand, a significant underestimation of suicide cases, in the order of 10% or more. Under these conditions, we are not close to knowing the real impact of the pandemic on suicide rates.

Is the situation better with regard to the measurement of suicide attempts, those gestures carried out with some intent to die but not having led to death? The answer is no. In France, the estimation of these acts is based first of all on the number of people hospitalized for this reason. However, doctors poorly code suicide attempts (“self-harm” in ICD-10 classification) upon discharge from hospital [4], often preferring to highlight the main psychiatric condition (for example depression). A need for training is evident here. Above all, the vast majority of people who carry out a suicidal act are not hospitalized. These people can present to the emergency room and then go home with a prescribed follow-up. However, motives of emergency room visits are still poorly recorded and centralized. In addition, many people — up to 40%, which is considerable — report having seen no physician or mental health professional after a suicide attempt (data from Baromètre Santé 2017) [5]. It is therefore not excluded that many suicidal acts were carried out at home during the pandemic and that the persons concerned, while in physical and mental danger, did not present to the emergency room or call a doctor. Suicide attempts — the number of which possibly far exceeds 200,000 per year in France in 2018 — are therefore very poorly identified in official statistics and it is not certain that we can correctly estimate variations in their number induced by the pandemic.

Finally, suicidal thoughts — which affect around 5% of the general population in France each year — could increase under these stressful conditions. Various research projects are currently aiming to measure the emotional impact of the pandemic including suicidal thoughts. Here, we could therefore have figures relatively quickly. Let's keep in mind, however, that if all suicidal acts are preceded by suicidal thoughts, most suicidal thoughts will not lead to action. We should therefore not seek to generalize these numbers to suicidal acts.

Knowing the impact of the Covid-19 pandemic on these tragic and complex behaviors will take time and we must refrain from any hasty conclusions until we can have valid, robust and multiple statistics. It is also an important and unmissable opportunity for improvement in medical information systems towards more reliability and speed in order to allow steering prevention as close as possible to the reality of the health situation. It is feasible in the short term and at a cost that is certainly negligible compared to the expected benefits. This must be part of the list of urgent changes in our health system to be supported in the immediate aftermath of this crisis, which may not be unique.

Disclosure of interest

The author declares that he has no competing interest.