2.1. Setting and study participants

This online survey was conducted via Survey monkey between 27 April and 17 May 2020, corresponding to the mandatory lockdown period in most SSA countries. At the time, it was not feasible to undertake a conventional Africa-wide community-based sampling survey due to the lockdown and restricted movement. A one-page project information statement that doubled as a recruitment poster was posted/reposted to WhatsApp chat groups and individual WhatsApp accounts. The page also had the contact email of the lead researcher if participants needed further information regarding any part of the study. A link to the online survey was provided. Recipients were further encouraged to send on or ‘snowball’ the e-link of the survey to other WhatsApp groups that they knew as well as to friends and other social media outlets. We also sent out the link by email to selected groups and individuals in all of the target countries relying on the authors’ networks with collaborating academics and local people living in SSA countries. Survey responses were saved and stored on survey monkey regional data center, and anonymized data was retrieved at the end of the study period for analysis.

2.2. Survey design

The survey instrument was adapted and developed from WHO-recommended questions used in a previous survey (Cronbach’s alpha coefficients of 0.74) [21] and had a brief overview of the context, purpose, procedures, nature of participation, privacy and confidentiality statements and notes to be filled out. For this study, we used a twenty-four item self-administered online questionnaire (Appendix A) which was divided into five sections (Sociodemographic and household factors, public attitudes toward compliance to COVID-19, mitigation measures, and risk perception for contracting COVID-19 infection). All questions relating to sociodemographic and household factors were mandatory.

Prior to the launching of the survey, a pilot study was conducted to ensure clarity and understanding as well as to determine the duration for completing the questionnaire. Participants (n = 10) who took part in the pilot were not part of the research team and did not participate in the final survey as well. In order to minimize bias, responses to the risk perception and attitude items of the online survey used a five-point Likert scale with provisions for neutral responses, so that the answers were not influenced in one way or another. The participants did not receive any incentives; their responses were voluntary and anonymous. The Kudar-Richardson 20 (KR-20) Cronbach’s alpha coefficient measuring internal consistency reliability for measures with dichotomous responses for the five mental health symptoms ranged from 0.70 to 0.74, indicating satisfactory consistency.

2.3. Inclusion and exclusion criteria

To be eligible for participation, the participants had to be SSA nationals either living abroad or in their countries of origin, which included Ghana, Cameroon (mostly distributed to the English-speaking individuals), Nigeria, South Africa, Tanzania, Kenya and Uganda, aged 18 years or more, and able to provide online consent. Non-SSA participants were excluded from this study.

2.4. Consent and ethical consideration

The Human Research Ethics Committee of the Cross-River State Ministry of Health, Nigeria approved this study (number of ethical approval: CRSMOH/RP/REC/2020/116). The study was carried out in accordance with the Helsinki Declaration for Human Research. The section also advised participants not to complete the survey a second time if they had already done so, and that only those aged 18 years were eligible to participate. The confidentiality of participants was assured in that no identifying information was obtained from participants.

This was followed by a consent section where participants were required to voluntarily respond with either a ”yes” or ”no” to the question inquiring whether they voluntarily agreed to participate in the survey. Participants who answered “yes” were directed to complete the survey. All participants gave written informed consent before participation in this study.

2.5. Data analysis

2.5.1. Outcome variables

The outcome variables in this study were derived from the item asking participants: ”how do you feel about the COVID-19 lockdown measures?“ (item 16 of Appendix A). The measures were, ”frustrated”, ”angry’, ”bored”, ”anxious” and ”worried” about COVID-19. Each of these five outcome variables were coded as binary, ”1” for yes and ”0” for no.

2.5.2. Confounding variables

The confounding variables used in this study included the demographic characteristics (age group, gender, marital status, and place of current residence, education, employment, occupation, religion; household factors (whether they lived alone and the number of household members which was an open-ended question, see item 12 of Appendix A). The items required a true/false, yes/no response with an additional “I don’t know/unsure” option provided. The public attitude towards COVID-19 mitigation practice variables were obtained from questions on whether the respondents practiced self-isolation, or home quarantine, and adhered to the precautionary public health measures such as, avoiding crowded places or religious events, use of face mask when leaving their homes, and practicing hand hygiene (washing hands with soap for at least 20 s each time or using hand sanitizers). These items were added to identify the effect of compliance to the mitigation practices put in place during the lockdown period to prevent the spread of the virus. For these variables, each question used a Likert scale with five levels with scores for each item ranging from 0 (lowest) to 4 (highest). As with epidemiological studies, the Likert scales were dichotomized to aid epidemiological interpretations and to describe the type of outcome under study (prevalence study and odds ratios). The risk perception variables were derived from questions on whether or not the respondents thought they were “at risk of becoming infected”, “at risk of dying from the infection”, “at risk of becoming severely infected”, “how worried they were because of COVID-19” if they thought “the infection would continue in their country” and how concerned they were of the possibility of being infected. These were included because individuals who perceived the risk as severe are more likely to reduce the spread of the virus.

2.6. Statistical analysis

Initial analyses involved frequency tabulations of all confounding factors in the study population presented in Table 1. This was followed by cross-tabulation to determine the prevalence and their corresponding 95% confidence intervals (CIs) of the mental health symptoms such as feeling anxious and emotional features that included being bored, frustrated, worried and angry. Univariate logistic regression was performed to examine the independent association between the five mental and emotional health symptoms (feeling bored, anxious, frustrated, worried and angry) and confounding factors (see Table 1 for details).

3. Results

3.3. Univariate analysis of factors associated with mental health symptoms

The univariate analysis of factors associated with the symptoms of mental health and emotional effects in the study population is presented in Table 2. Living in Central Africa, with six or more people in the household was associated with increased odds of feeling bored, frustrated, angry and anxious among the respondents. Those who lived with more than six persons in the household showed significantly higher odds for all the dependent variables except for “feeling worried” and East African respondents had remarkably higher odds of feeling frustrated due to COVID-19. Compared to men, women were more likely to feel bored (OR 1.28, 95% CI 1.02, 1.59) and anxious (OR 1.24, 95% CI 1.02, 1.53) and those who were not married were more likely to feel frustrated (OR 1.25, 95% CI 1.02, 1.52) and angry (OR 1.30, 95CI 1.02, 1.66) compared to the married respondents. Higher odds of feeling ‘angry’ was found among those who were unemployed (OR 1.29, 95% CI 1.04, 1.59), and among respondents who thought that COVID-19 would not continue in their respective countries after the lockdown (OR 1.40, 95% CI 1.08, 1.81).

Individuals who were concerned that they or their family members could be infected with COVID-19 were less likely to feel worried and anxious about contracting the infection. Similarly, participants who felt at lower risk of being infected (OR 0.34, 95% CI 0.27, 0.41), or being severely infected (OR 0.26, 95% CI 0.20, 0.33) and those who thought their risk of dying from the infection was low (OR 0.18, 95% CI 0.14, 0.25), were less likely to worry about COVID-19, in this study.

3.4. Multivariate analysis of factors associated with mental health/emotional symptoms

Table 3 shows the factors associated with the symptoms of mental and emotional health, after adjusting for all potential covariates. Age became a significant factor influencing the respondents’ experience of mental health symptoms. Participants aged 29–38 years had higher odds for feeling bored (aOR 1.81, 95% CI 1.05, 3.10), and frustrated (aOR 1.95, 95% CI 1.20, 3.19), while those aged 39-48yrs (aOR 2.09, 95% CI 1.22, 3.56) were more likely to feel frustrated due to COVID-19 compared to younger participants (18–28years). Central African respondents reported higher odds of feeling frustrated (aOR 1.49, 95% CI 1.01, 2.19), angry (aOR 2.12, 95% CI 1.37, 3.29), and anxious (aOR 1.60, 95% CI 1.05, 2.43), whereas respondents from Southern African countries reported higher odds of feeling frustrated (aOR 1.46, 95% CI 1.06, 2.00), compared to those from the West African countries. Other factors associated with higher odds of mental and emotional health symptoms in this study included being unmarried, being unemployed, living with six or more people in the household during the pandemic, perception of low risk of contracting the infection and the thought that COVID-19 will not continue after the lockdown.

Overall, respondents who perceived a low risk of being infected by COVID-19 were less likely to be worried about the disease and those from the Southern African countries were less likely to feel bored during the pandemic (aOR 0.59, 95% CI 0.42, 0.82), compared to the respondents from West Africa.

Table 3. Multivariate analysis of factors associated with mental health/emotional impact of COVID-19 among Sub-Sahara Africans during the lockdown.

Overall, respondents who perceived a low risk of being infected by COVID-19 were less likely to be worried about the disease and those from the Southern African countries were less likely to feel bored during the pandemic (aOR 0.59, 95% CI 0.42, 0.82), compared to the respondents from West Africa.

4. Discussion

The current study explored both mental and emotional health symptoms among SSA respondents, during the COVID-19 lockdown. This is the first study using a web-based cross-sectional survey to examine the prevalence and factors associated with mental and emotional health symptoms of COVID-19 in SSA. This study found that the COVID-19 pandemic had a major impact on the mental and emotional health of respondents in SSAs (including health care workers). More than half of the respondents reported feeling anxious, worried, frustrated and bored, whereas approximately one in four respondents reported feeling angry during the COVID-19 lockdown period. The study also revealed that those older than 28 years who lived in the Central and Southern African countries, respondents who were not married, the unemployed, as well as those who lived with more than six persons in a household, had higher odds of mental and emotional health symptoms during the COVID-19 lockdown period. In addition, respondents who felt that at lower risk of being infected by the virus and those who did not think that COVID-19 will continue after the lockdown, were more likely to feel angry about the pandemic. The study also found that people working in health care sectors were less likely to report mental and emotional health symptoms during the COVID-19 pandemic.

Compared with previous rates of any mental health/emotional symptoms in Low and Middle-Income Countries (LMIC) and SSA countries (10–20% at any one time [20,22]), the impact of the pandemic on the state of mental health (which includes emotional and psychological well-being) of SSAs in this study was profound and increased by three folds. Reports from China [23,24], the USA (53% reported feeling anxious and stress relating to the coronavirus) [25], India (30.5% reported depression) [26] and Italy (41.6% reported moderate stress) [27], found much lower prevalence of mental illness/emotional symptoms than the present study. Although COVID-19 infections and deaths are lower in SSA region than other regions, the higher prevalence of mental health symptoms found in this study suggests that the SSA population could be particularly vulnerable to emotional distress in the current pandemic. Given the already existing situations of poverty, unemployment and weak health systems [20] and the unavailability of effective treatment, it is expected that the rate of mental and emotional health symptoms will increase. This suggestion is substantiated in Table 2, whereby it is observed that respondents were more worried about the likelihood of COVID-19 continuing (63.9%) than their risk of becoming severely infected (25.8%) or dying (19.4%) from the disease. This finding is consistent with a recent study from the UK, which found that the citizens were more concerned about how societal changes would influence their psychological and financial wellbeing, than their risk of becoming unwell with the virus [28]. Prevention efforts such as screening for mental health and emotional problems and psychoeducation [29] focusing on the identified groups at risk for adverse psychosocial outcomes are needed.

The higher odds of mental illness/emotional symptoms among health care workers (HCWs) than non-healthcare workers (NHCWs) found in previous studies among Chinese residents [9,11] were not found in this study. Rather, we found that HCWs were less likely to report any COVID-19 related mental health symptom than NHCWs, particularly with respect to feeling ‘angry’. Despite the lower odds of mental health/emotional symptoms among HCWs, they are particularly vulnerable to emotional distress in the current pandemic. This is due to their level of exposure to the virus, concern about being infected and caring for their loved ones, shortages of personal protective equipment (PPE), longer work hours, and involvement in emotionally and ethically fraught resource-allocation decisions [30]. HCWs should be monitored for a change in routine and behavior. Similar to the previous reports from Italy [27] and China [9,31], the present study found a significantly higher odds of mental health symptoms among women than men during the pandemic (Table 3). This finding may be explained in part by the report that women including those who are pregnant or have young children were more likely to develop the fear of becoming infected or transmitting the virus [32].

As part of the measures to deal with mental health and psychological issues during the COVID-19 pandemic, the WHO had encouraged individuals to stay with friends and families [10]. This guideline was supported by our findings of higher odds for feeling bored, frustrated, and angry among respondents who were single. On the other hand, living with more than six persons in the household was associated with higher odds of mental and emotional health symptoms of feeling anxious, angry, and frustrated. This may also suggest that living with many people in the household could make people feel more anxious. Evidence from a systematic review and meta-analysis on household transmission studies of COVID-19 found that the risk of infection is 10 times higher among household contacts than other contacts [33]. These trends have challenged the traditional SSA social structure of communalism, a crucial socio-cultural factor, whose maintenance could be important in dealing with mental health issues in the SSA context [6]. Many communities in SSA rely on social resources for dealing with mental health issues as utilization of orthodox mental health care services is generally low [12,20]. Some of the resources people access for relief from mental problems within the SSA context include keeping in touch with others, attending faith and religious events, and engaging in prayers [6].

In the wake of the COVID-19 lockdowns in SSA, access to social resources was limited, and alternative ways of delivering mental health resources were needed [12]. Prior to COVID-19, there was already a huge gap of unmet mental health services for older adults in SSA [12,34] which is fueled by factors such as stigma, poor awareness that older adults suffer from mental illness, deficient primary health care services, inadequate community healthcare workers and psychogeriatricians [35]. In addition to these factors, the low levels of digital literacy in SSA hindered the deployment of online or virtual mental health service delivery [6] as alternatives to overcome disruptions to in-person services. While more than 80% of high-income countries have started the utilization of alternative mental health interventions to bridge gaps in mental health services, the patronage by low-income countries has been less than 50% [7]. In line with WHO’s guideline, SSA countries must allocate resources to mental health as an integral component of their response and recovery strategies. Utilization of the mass media to share survivor experiences to mental health patients and the public could be a good alternative for delivering counselling measures [6]. Additionally, educational campaigns are needed to increase the awareness and enlightenment of the public regarding the fact that older adults can suffer from mental illnesses and recognize the benefits of orthodox management of mental illnesses [12].

The present findings indicate that individuals who perceived themselves or their family members to be at a lower risk of being infected by COVID-19 or dying from the disease reported lower odds for worrying about COVID-19. Strong coping mechanisms are necessary to deal with mental health and emotional issues during a pandemic, and one of these coping strategies is to be less concerned about the consequences and impact of the disease and remain cautious. According to the WHO, people should reduce the amount of information they receive about COVID-19 to reduce feeling anxious [10]. Consistent with our finding, a study in China found that people who spent too much time thinking about disease outbreak were more likely to develop symptoms of anxiety [11]. This may explain the higher prevalence of mental health/emotional symptoms, which we found among the unemployed respondents. The mental health and psychological impact of the pandemic because of rising economic recession and massive job losses within the context of struggling economies cannot be underestimated. Therefore, mental health support services should be an integral part of the disease response strategy in SSA.

5. Limitations and strengths

This study has some limitations. First, the data was collected using an online survey and may not be a true reflection of the opinion of other SSAs living in rural areas where internet penetration and connectivity remain relatively low [36] or the older people who are less likely to use the internet. However, the increase in the use of internet recorded among the general population during the pandemic meant that many people may have participated [37]. Also, this was the only reliable, cost effective means to disseminate information at the time of this study and obtain real-time data on the current situation. Second, the survey was available only in English, making it difficult for people living in French-speaking countries including some part of Cameroon, to participate. Third, there was limited participation of East African respondents in this study, which may be attributed to the government instructions for residents to refrain from giving out information regarding the pandemic. Fourth, the research methodology did not allow us to reach people with medically examined mental health symptoms; therefore, the provision of the results may not fully reflect the severity of the mental health symptoms among SSA population. Fifth, this study did not use the tools designed specifically for the COVID-19 pandemic, such as the coronavirus anxiety scale (CAS). Future studies using the tools developed especially for the COVID-19 pandemic will provide a concrete finding and facilitate the demand for a focused public health initiative. Another limitation peculiar to web-based surveys was the inability to verify the eligibility of the participants and the validity of their responses. Despite these limitations, this is the first study to provide comprehensive evidence of the mental health impact of the pandemic across SSA region. With a web-based questionnaire, the study was able to assess the prevalence of mental health symptoms among SSA respondents, while maintaining the WHO recommended “social distance” during the COVID-19 pandemic, which otherwise would be impossible. Furthermore, using a robust analysis, we were able to minimize bias by controlling for all potential confounders in the analysis. That no incentives were not given to participant’s ensured that their participation and response rates were not influenced [38].

6. Conclusions

In conclusion, amidst relatively lower disease and death ratios, this study highlights a high prevalence of mental health and emotional symptoms during COVID-19 in Sub Sahara African region. Despite the lack of a baseline mental health study of the study population prior to COVID-19, the findings strongly suggest markedly elevated mental health symptoms whose rates are consistent with those of other study populations worldwide [39]. Such a high impact of the disease may be related to the weak health systems and low access to alternative mental health service delivery within the sub-region. While three out of every four persons surveyed reported feeling bored, about one in every two persons, felt frustrated and worried about the lockdown. Southern and Central Africans had a greater risk of mental health and emotional symptoms during COVID-19 pandemic. Implementing community-based strategies to support resilience among these psychologically vulnerable individuals such as the older adults, those who neither are married nor employed, as well as people living with many household members during the COVID-19 crisis is fundamental for the SSA communities. The psychological impact of fear and feeling anxious induced by the rapid spread of pandemic needs to be clearly recognized as a public health priority for both authorities and policymakers who should rapidly adopt clear behavioral strategies to reduce the burden of disease, plan for long-term fallout of the disease and the dramatic mental health consequences of this outbreak. Most importantly, mental health service resources must be as an integral component of SSA governments’ response and recovery strategies of the COVID-19 pandemic.