On 17 March 2020, the French government put in place lockdown measures across the country in response to the rise in COVID-19 cases and deaths. It was in this context that Santé Publique France launched a survey to monitor population mental health and behavior in order to quickly collect useful data to coordinate a preventive response. The survey looked at the degree of adoption of hygiene measures and physical distancing (both key determinants of the evolution of the epidemiological situation), and the mental health of the population.
This article presents the results of the first eight waves of the survey, collected between 23 March – 20 May 2020, and reflects on how they can input into a preventive mental health strategy.
It has been demonstrated that health epidemics and living situations during lockdowns (e.g., overcrowding and lack of privacy, social isolation, loss of income, etc.), pose high risks in terms of mental health, in particular concerning anxiety (fear of oneself or of one’s relatives falling ill, frustration, anger, etc.).
It has therefore been a public health priority to monitor population mental health, and especially to prevent the development of mental ill-health. In the short term, it was important to ensure that the health system, in particular hospitals and emergency services, was not overloaded, as it struggled to cope with the impact of the pandemic.
Each survey wave collected data from 2000 people aged 18+ living in mainland France, through online interviews. In order to make sure that the profile of respondents was representative of the general population in terms of sex, age, occupation, size of town, and region, these independent samples were recruited in the scope of an access panel, by the polling institute BVA, using quota methodology.
French population’s mental health at the beginning of the lockdown
At the end of the first week of lockdown (wave 1), population mental health in France had considerably worsened, compared to pre-lockdown available data (Baromètre Santé publique France 2017). Respondents reported twice the usual prevalence of anxiety disorders (27% compared to 13.5%), and their life satisfaction (score> 5 out of 10) diminished by 20 points (66% compared to 85%).
Evolution during and after lockdown (figure 1)
On average, respondents’ mental health improved between wave 1 (23-25 March) and wave 8 (18-20 May) of the survey, with the end of the lockdown having a positive impact:
life satisfaction, after significantly improving (+10 points) between the 2nd and 3rd week of lockdown, again improved after the end of the lockdown (+5 extra points). This is the case for 81% of respondents in the 8th survey wave (compared to 85% pre-pandemic).
anxiety levels significantly decreased between the 1st and 3rd wave (-9 points). They then stabilised, with a downward trend (not significant), which appears to be continuing. During the 8th survey wave, 17% of respondents reported anxiety disorders (compared to 13.5% pre-pandemic).
levels of depression, which increased between waves 3 and 4, then began to decline, with a significant decline observed since the end of the lockdown (-6 points).
sleep issues (recorded during the 8 days preceding the survey) significantly increased, up until the 4th wave. By the 8th wave, respondents were reporting similar levels of sleep issues as in the 2nd wave, due to the end of the lockdown. Their frequency continues to be higher (63%) than before the pandemic (49%).
Determinants of anxiety
Since the beginning of the lockdown, the main factors linked with higher levels of anxiety (after controlling for socio-demographic factors, factors linked with the epidemiological situation and cognitive factors, e.g. perceptions and knowledge), are the following: (1) in terms of socio-demographic factors: having previous experience of mental ill-health, being in a difficult financial situation, or being a woman; (2) in terms of factors linked with the epidemiological situation: actively researching information on COVID-19 and having had trouble breathing; (3) in terms of cognitive factors: perceiving COVID-19 as a serious illness, feeling vulnerable to COVID-19, not being well-aware of how the virus is transmitted, and perceiving the protection measures as ineffective and constraining.
Lessons learnt and looking ahead
French population mental health was severely impacted at the beginning of the lockdown, with anxiety disorders twice as prevalent, and feelings of life satisfaction dropping by 20 points. The results of the 1st wave of the survey allowed Santé Publique France to rapidly alert health authorities, and to put in place measures to support population mental health. Together with the Ministry of Health, health professionals and the volunteer sector, we set up a mental health helpline linked to the (free) national telephone helpline providing information on COVID-19. Tools, internet ads, posters, and web pages were developed to publicise the helpline, and to direct those in situations of psychological distress towards other resources which could provide responses to specific situations (e.g., isolation, violence, grief, addiction, parenting…).
Population mental health then progressively improved, in particular after the end of the lockdown (significant improvement in life satisfaction; significant decline in depression and sleep issues). People with experience of mental ill-health and financial difficulties continued to report high levels of anxiety.
These results highlight the important impact of economic factors (in particular on mental health), which have led all countries under lockdown to draw up conditions for a progressive re-launch of economic activities. A recent study by the University of Cambridge, to which Santé Publique France contributed for the section on France (not yet published), highlights that the French reported fewer financial difficulties and mental health issues than many other countries. French social and economic protection systems are likely to have played an important role in cushioning the effects of the epidemiological crisis on population mental health. The results also highlight the importance of resuming routine health care – including mental health care – after the end of the lockdown for vulnerable groups.
As we enter a period of “deconfinement”, it will be important in both the short and medium term to ensure that measures intended to protect population health are followed. In order to promote the adoption of these measures, whilst striving to ensure that anxiety levels do not increase, our data points to the importance of clear and simple communication regarding how the virus can be transmitted, and the most effective measures which can be adopted to prevent this (e.g., increasing knowledge and perceived effectiveness). These recommendations are in line with data regarding factors that determine whether people adopt protection measures, which highlight that variations in behaviour are more often due to the perception of social norms (support and adoption of preventive measures by one’s entourage) and confidence in one’s ability to adopt measures than to perceived risk levels. The list of variables included in the regression model as well as the method of analysis and selection of these variables is available in the following article. Chan-Chee C, Léon C, Lasbeur L, Lecrique JM, Raude J, Arwidson P, du Roscoät E. The mental health of the French facing the COVID-19 crisis: prevalence, evolution and determinants of anxiety disorders during the first two weeks of lockdown (Coviprev study, March 23-25 and March 30 – April 1st, 2020). https://www.santepubliquefrance.fr/docs/la-sante-mentale-des-francais-face-au-covid-19-prevalences-evolutions-et-determinants-de-l-anxiete-au-cours-des-deux-premieres-semaines-de-confi  The Winton Centre for Risk & Evidence Communication at the University of Cambridge. Perception of the risk of COVID-19 and governments’ responses to it.