Deterioration of mental health and insufficient Covid-19 information among disadvantaged immigrants in the greater Paris area.

OBJECTIVES: The aim of this study is to provide information on changes in mental health among disadvantaged immigrants from Sub-Saharan Africa in the Greater Paris area and their level of information about Covid-19.
METHODS: Prior to the Covid-19 epidemic, the Makasi community-based cohort followed 850 immigrants from sub-Saharan Africa in the Greater Paris area. Between the 1st of April and the 7th of June 2020, all participants scheduled for a follow-up survey were systematically included into an additional COVID-19-related wave of data collection (N = 100). We compared participants' type of housing, level of food insecurity, work and mental health (PHQ9) before and during the first COVID-19-related lockdown, using paired-Mc Nemar chi-2 tests. We next described their level of information on Covid-19 and policy measures, broken down by sex.
RESULTS: Among the 100 participants, 68% had no legal residence permit. Food insecurity was more often reported during lockdown than before (62% vs 52%). 9% of participants had a score indicative of severe depression (PHQ9) before lockdown and 17% afterwards (p = 0.17). Only 51% knew about the possibility of asymptomatic transmission of the COVID-19 virus.
CONCLUSIONS: This study brings original information on a hard-to-reach population group. Our results suggest that the lockdown had a detrimental impact on various economic and mental health aspects among disadvantaged migrants residing in the Greater Paris area.

Background

Immigrants appear to be disproportionally affected by Covid-19 [1]. Far from biological mechanisms, many scholars pinpoint the role of socioeconomic factors to explain this disparity and call for collecting evidence [2]. In July 2020, the French National Institute of Statistics published a report revealing that the mortality rate was twice as high among immigrants as among natives during the first wave of the COVID-19 epidemic in April and March 2020 [3]. Among immigrants from sub-Saharan Africa, excess mortality was estimated at +114%. Although these are all-cause mortality data, there is no doubt that immigrants were more affected by Covid-19. Indeed, a large nationally-representative seroprevalence study showed that immigrants from countries outside EU experienced a higher seroprevalence of Sars-Cov-2 (9.4% versus 4.1% in the non-immigrant population) [4].

However, very little data are available on immigrants' experiences during the epidemic, and even less among immigrants who are undocumented or disadvantaged. Yet undocumented migrants are probably at even greater risk for Covid-19 than other immigrants because of a limited access to healthcare and lack of information (Bhopal 2020). A recent study conducted in Switzerland found poor mental health among undocumented migrants during the COVID-19-related lockdown [5] and a recent meta-analysis underlined that having a lower socio-economic status and experiencing social isolation are risk factors for depression and anxiety during the pandemic [6]. Previous qualitative work showed how the fear of the new coronavirus and the fear of being arrested exacerbated by police presence in the public space, as well as global uncertainty regarding livelihood, were major determinants of mental health during the first lockdown [7].

The aim of this study is to provide detailed information on changes in mental health among disadvantaged immigrants from sub-Saharan Africa in the Greater Paris area and their knowledge and level of information about the Covid-19 during lockdown (17th March - 12th May 2020). The Greater Paris area was one of the regions most affected by the Covid-19. Our main hypotheses were 1) deterioration of mental health and 2) lack of information about Covid-19 due to distance from the healthcare system and public services. Our results can inform public response in case of another epidemic and potential lockdown.

Methods

Prior to the Covid-19 epidemic, the Makasi community-based cohort followed 850 immigrants from sub-Saharan Africa in the Greater Paris area to assess the effectiveness of an intervention aiming to enhance participants' empowerment to reduce sexual risk.

Participants were recruited in the Greater Paris area via outreach activities conducted by the community-based charity Afrique Avenir. They were above 18 years old and met at least one of the following criteria of disadvantage: having unstable housing, being unemployed, having food deprivation, being undocumented, experiencing violence, having no medical insurance, and ignoring where to go to see a doctor. The level of acceptation was 63%, which is rather high for this kind of research among hard-to-reach populations [8,9]. If they accepted, a written consent was collected by the outreach team.The research protocol included a baseline quantitative questionnaire assessing participants' living conditions and mental health and two follow-ups at 3 and 6 months (The study is registered in Clinical Trials: NCT04468724) [10]. Between the 1st of April and the 7th of June 2020 (first lockdown in France), all participants who had a follow-up were systematically included in an additional COVID-19-related wave of data collection (N = 100). The follow-up and additional data collection were conducted by telephone and included questions on health literacy and Covid-19-related knowledge. It should be noted that prior to the lockdown, some follow-ups were already made by phone if a person could not come to the interview. Therefore, we were able to compare participants' social and family situation, isolation and mental health during the lockdown and three months before.

We described the study population according to sociodemographic characteristics. We compared the type of housing, food insecurity, employment and mental health (PHQ9) before and during the lockdown, using paired-Mc Nemar chi-2 tests. We then described the level of information on the Covid-19 epidemic.

Ethical approvals

The MAKASI study was approved both by the Comité de Protection des Personnes Sud-Ouest et Outre-Mer (ID RCB 2018-A02129–46) and by the CNIL (Commission Nationale Informatique et Libertés), n°2,215,270 in France.

Results

The population recruited in the Makasi study mostly includes undocumented migrants hosted by their social network. Among the 100 study participants recruited before March 2020 and interviewed by telephone during the lockdown, 79% were men, 68% had no legal residence permit and 40% had no health insurance before lockdown (Table 1 ). The majority came from a Western African country, and 43% arrived in France because of risks to their safety in their country of origin. Although one quarter of participants had been in France for less than a year, the sample also comprises persons living in France for 7 years and more (7%).

Table 1

Sociodemographic characteristics of disadvantaged immigrants, MAKASI project, April–June 2020.

	Total N = 100
Sex
Men	79%
Women	21%
Age
19–29	34%
30–39	39%
40+	27%
Region of birth
West Africa	54%
Central Africa	41%
Eastern and Southern AfricaMissing	4%1%
Reason for coming to France
Find work, try one's luck	37%
Join a family member	10%
Threatened in his/her country	43%
Studies	5%
Medical reasons	3%
Others	2%
Level of education
None/Primary	27%
Secondary	57%
Superior	16%
Length of stay in France
First year	24%
Second year	19%
Third year	15%
3–6 years	34%
7 years and more	7%
Missing	1%
Has a stable partner before lockdown	47%
Has children before lockdown	62%
Legal status before lockdown
Undocumented	68%
Permits <1 year (incl. Asylum seekers)	22%
Permits ≥1 year	3%
10-y residency permit or French nationality	7%
Health insurance before lockdown
None	40%
State Medical Aid (for undocumented persons)	22%
Basic health insurance	23%
Basic health insurance and complementary insuranceIn the process of renewing health insurance	9%6%

Food insecurity was more often reported during lockdown (62% vs 52% before lockdown, p < 0.24) (Table 2 ) and 61% of participants reported that they had more difficulty getting what to eat. Besides, fewer participants reported living on the street during lockdown (5% vs 15% before lockdown) and a higher proportion were hosted by friends or acquaintances (66% vs 50%, p < 0.09). However, 61% of participants reported that they had no personal space to sleep during the lockdown. Participants' mental health worsened during lockdown. Indeed, the proportion of the sample suffering from depression was 72% (vs. 65% before), and 17% of participants had a score indicative of severe depression (PHQ9) (vs. 9% before lockdown, p < 0.17).

Table 2

Comparison of social situation before and during lockdown among disadvantaged immigrants, MAKASI project, April–June 2020.

	Before lockdown	During lockdown	pa
Stable housing
Yes	23%	18%	0.42
No	77%	82%
Type of housing
Personal dwelling	23%	16%	0.09
Hosted by family/friends, …	50%	66%
Hosted by a charity	11%	13%
Nowhere to go/street	15%	5%
Has spent at least a day without eating in the preceding month for lack of money
Yes	52%	62%	0.24
No	48%	38%
Mental Health (PHQ9):
Absence of significant depression symptoms of depression (score PHQ9: 0–9)	35%	28%	0.32
Presence of significant symptoms of depression (score PHQ9 10 and more)	65%	72%
Severe depression (score PHQ9:20–27)	9%	17%	0.17

Paired McNemar chi2 tests comparing distributions before and during lockdown.

Regarding their professional activity (in the formal or informal sector), only 38% of participants had a job before lockdown. Among them, 82% lost their job. Those who kept their job were mainly working in frontline jobs such as delivery persons, home helpers, supermarket logistics and construction (data not shown).

Participants' level of knowledge of the Covid-19 appeared insufficient: 17% of the respondents reported that they did not know what to do in case of Covid-19 symptoms, and only 51% knew about the possibility of an asymptomatic transmission of the virus (Supplemental Material 1). Additionally, few participants had a precise knowledge of the specific policy measures implemented by the government: only 45% and 24% of them knew that the validity of all residence permits and the State Medical Aid (health insurance for undocumented persons) had been extended. Overall, 30% stated that they did not receive sufficient information from the authorities concerning the epidemic. Finally, regarding strategies to preserve one's health, the majority of participants mentioned barrier measures (masks, hand washing, physical distancing), which is in line with qualitative evidence showing a strict respect of barrier measures and lockdown instructions in this population [7].

Discussion

This study brings original information on the impact of the COVID-19 epidemic on a hard-to-reach population. Our results suggest that the lockdown had a detrimental impact on various economic and mental health aspects among disadvantaged immigrants residing in the Greater Paris area. Moreover, their level of information on Covid-19 was insufficient. First, the sanitary crisis worsened their professional and financial situation which was already critical prior to the epidemic. Second, though mental health outcomes have deteriorated during lockdown in the general population of France, the level of depression found in Makasi (72%) was still more than three times higher than the highest proportion of depression measured in France during the same period (20.4%) [11].

Additionally, it is likely that overcrowded housing and working conditions in frontline jobs have resulted in greater risk of Covid-19 exposure, whereas economic hardship and deteriorated mental health may have decreased individuals' access to healthcare, even in case of Covid-19 symptoms. Thus disadvantaged immigrants are at higher risk of Covid-19 infection and delayed access to healthcare [12]. While strongly exposed to the disease, they did not have sufficient information about the epidemic and policy measures, even if some information campaigns were launched in France to address the specific needs of this population [13]. More research is therefore needed to better understand what kind of interventions could be effective in conveying prevention messages to disadvantaged immigrants.

The main limitation of our study is its moderate sample size, which restricts the statistical analyses and inference, and calls for caution in the interpretation of the results.

However, our study focusing on a sub-sample of disadvantaged immigrants, still provides a useful insight into their specific difficulties during lockdown. The follow-up during lockdown was a unique opportunity to document its impact on such a population which is often not represented in official statistics. In France, a large population survey revealed that the COVID-19-related lockdown worsened the financial situation of 39% of immigrants from non-EU countries. However, this study did not include undocumented immigrants [14]. NGOs have also reported the situation of undocumented immigrants placed in emergency shelters during lockdown [15] [16]. Yet our study shows how the lockdown has impacted immigrants who are invisible in most studies, i.e. those who are hosted by relatives or friends and who have no legal residence status.

Conclusion

Although information campaigns have been specifically designed in France to inform immigrants, our results clearly indicate that more needs to be done to reach-out the most disadvantaged ones. In particular, community-based approaches [17] could be key to develop multilingual, community-based information campaigns and interventions regarding prevention, access to healthcare, vaccination, and social support during the pandemic.

Temporary citizenship could greatly improve the situation of undocumented immigrants in France, facilitating their access to public social services and healthcare [18]; unfortunately it seems that we are witnessing a restriction of immigrants' access to healthcare in France, through new limitations of the State Medical Aid and digitalization of all administrative procedures [19,20].

Ethical statement

This project was approved both by the CNIL (Commission Nationale Informatique et Libertés), n°2,215,270 and by the Comité de Protection des Personnes Sud-Ouest et Outre-Mer (ID RCB 2018-A02129-46).

Declaration of Competing Interest

The authors have no conflict of interest to declare.