Perceived risk, anxiety and protective health behaviours during the early phase of coronavirus (COVID-19) pandemic in Nigeria

Background: Pandemics such as the current COVID-19 pandemic are often associated with heightened fears and signicant adjustments in health behaviours. Aim: This study was aimed to assess perceived risk, anxiety and protective health behaviours of the general public during the early phase of the coronavirus (COVID-19) pandemic in Nigeria Methods: A cross-sectional study among 1197 respondents aged 18 years and above between April 27th to May 16th, 2020. Result: More than half (61.9%) of the respondents have high risk perception towards COVID-19. Using the seven-item Generalized Anxiety Disorder (GAD-7) scale, high anxiety level was found in 37.2% of the study respondents. Male gender (OR=1.38,CI=1.09-1.74), being employed (OR=1.53,CI = 1.21- 1.95) and high risk perception (OR=1.33, CI = 1.05-1.69) were positively associated with observance of more than one protective measure against COVID-19. Having not more than 12 years of education (OR=0.51, CI=0.32-0.81) was negatively associated with observance of more than one protective behaviour. High risk perception (OR=1.82; CI=1.42-2.34), having not more than 12 years of education (OR=1.73; CI= 1.13-2.64), and being employed (OR =1.29; CI=1.01-1.64) were positively associated with anxiety. Being aged 25-34 years (OR=0.34; CI =0.12-0.94) and being aged 35-54 years (OR=0.30; CI=0.10-0.85) were found to be protective against developing COVID-19- related anxiety. Conclusions: This study showed that risk perception has inuence on both anxiety and observance of protective behaviours. Being a novel experience, this research has implications to support current and future response to a pandemic experience.


Introduction
Undoubtedly, one event that would de ne the year 2020 for a long time to come is the Coronavirus  pandemic. The respiratory disease which was rst discovered in the Chinese city of Wuhan in December of 2019 shares similarities with Severe Acute Respiratory Syndrome (SARS) and Middle East Respiratory Syndrome (MERS)-two other previously known coronavirus infections in humans (WHO 2020a). The core symptoms of the disease, according to the World Health Organisation (WHO), are dry cough, fever and tiredness (WHO, 2020b).
On the 30th of January, 2020, WHO declared SARS-CoV-2 which is now renamed COVID-19 a Public Health Emergency of International Concern (WHO2020c). Becoming a global crisis on the 11th of March 2020, WHO characterized COVID-19 as a pandemic, thus becoming the rst ever Coronavirus to be declared a pandemic (WHO, 2020c). As at 29th April 2020, the virus had spread exponentially with over 3million con rmed cases, and 200,000 deaths in more than 200 countries. On February 27, 2020, Nigeria Page 3/16 con rmed its COVID-19 index case-an Italian citizen-and as at April 29, 2020, the Nigeria Centre for Disease Control (NCDC) had reported over 1,500 laboratory-con rmed cases of COVID-19 in Nigeria with 307 discharges and 51 deaths (NCDC, 2020).
According to WHO, the infection has no proven treatment or vaccine yet, but can be properly managed even to the point of recovery (WHO, 2020b). Therefore, parallel to efforts aimed at identifying, isolating and treating cases, WHO and other scienti c communities have been emphasising the need for individuals to adopt practices aimed at preventing and reducing the spread of the disease. These recommendations which are similar to previous pandemic containment strategies include using tissues when sneezing, washing hands regularly with soap and water, and social distancing (Rubin et al 2009). Due to the severity and spread of the virus, severe measures have been taken which involve forcing people to stay at home, closing of businesses and o ces, exempting only essential workers. Majority of the Nigerian population, however, belongs to the informal economy, and their livelihoods are consequently being threatened by the lockdown since much of their activities and businesses involve face-to-face contact.
According to the Health Belief Model, a set of core beliefs are at the root of the adoption or otherwise of preventive health behaviours (Rosoenstock et al., 1966). One of these core beliefs is 'perceived susceptibility'-an individual's perception of their likelihood of developing a certain health condition. Therefore, the more likely people believe they are to contract a disease, the more likely they are to adopt behaviours meant to prevent the disease. Currently in Nigeria, there are those who strongly believe that the novel coronavirus is a hoax and therefore all the measures being put in place by the authorities are unwarranted. Another group of Nigerians (about 26%), as reported by the Ngozi Okonjo-Iweala (NOI) poll believe that they are shielded from the virus owing to their religious dispositions, viable genes and the hot weather of the country (NOIPolls, 2020). Yet still, there are those who hold the position that COVID-19 is a disease of a uence of some sort and so only the rich (especially those who could afford to travel overseas) could be infected with it. Even though the aforementioned assertions are obviously unfounded, the fact that some people hold such beliefs about the disease shows clearly that there are many who believe that they cannot contract the virus. Therefore, given the huge public health implications of these erroneous dispositions (such individuals are more likely to be careless and propagate the spread of the disease), it became expedient to investigate the perceptions of Nigerians about their vulnerability to COVID-19. Such knowledge is needed in the development of health messages that are relevant to the peculiarities of this society.
It is only expected that a health crisis of this magnitude, which has signi cantly disrupted the normal lives of people in Nigeria, would create anxieties and fears in different categories of people. Drastic measures such as the lockdown have put signi cant strain on businesses and livelihoods of many in the country, leading to palpable apprehension about both the present and the future. Also, fake news and conspiracy theories have been having a eld day, thriving especially on social media. Therefore, given the fact that Nigeria has over 25 million smartphone users with access to various social media platforms, this puts an appreciable number of people at an increased risk of being exposed to unveri ed information that can trigger panic (Statista, 2020). Also, constant exposure to news and updates about the pandemic, even when they are accurate, can instill fear and panic in some individuals, especially children and people who may have pre-existing mental health conditions (Omigbodun & Abdulmalik, 2020). In addition, individuals who have been infected by the virus as well as their families may also be on edge owing to the uncertainty about the outcome of the disease (Omigbodun & Abdulmalik, 2020).
Given the signi cant interlink between human psychological and behavioral factors, therefore, it is important to assess the role risk perception plays in the development of COVID-19-related anxiety and adoption of protective behaviours in Nigeria. Understanding the role of speci c perceptions in people may help health communicators to improve their messages about outbreaks of new infectious diseases generally and COVID-19 speci cally. For majority across the world, this is the rst pandemic experience and so there is limited evidence on reactions to prior pandemic experience-in Nigeria especially.
Additionally, we are unaware of any data of this kind on the current COVID-19 pandemic in Nigeria. This study was carried out to assess perceived risk, anxiety and protective health behaviours of the general public during the early phase of the coronavirus (COVID-19) pandemic in Nigeria.

Sampling
The study was a cross-sectional study. Due to the lockdown and restriction imposed by the government to curb the spread of the virus, an online Google form questionnaire was created and the link was shared using snowballing sampling technique to respondents across the 6 geo-political zones of the country. The semi-structured anonymous online questionnaire took about 15 minutes to complete. In a bid to ensure representativeness of the data, ve (5) online data o cers were recruited from across the country to oversee the data collection process. The strategies adopted in sharing the questionnaire included direct and group broadcast on social media platforms such as WhatsApp, Facebook and Telegram. The survey was carried out during the rst phase of COVID-19 in Nigeria between April 27th to May 16th, 2020. The questionnaire was completed in English Language. A 1216 responses were received, 1197 of which were valid.

Inclusion criteria
Any individual who is 18 years and above with access to internet and who gave their consent.

Exclusion criteria
Children were excluded from the study due to their inability to give direct consent.

Instrument
The questionnaire comprised four sections. The rst section obtained socio-demographic information including sex, age at last birthday, marital status, highest level of education, employment, religion, current state of residence and country of origin.
Level of Anxiety was assessed using the 7-item General Anxiety Disorder (GAD-7) scale. Our study assessed respondents' anxiety symptoms over the period of two weeks using a 4-point likert scale. The total score (0 to 21) was calculated from their response "not at all (0), several days (1), more than half a day (2), and nearly every day (Spitzer et al., 2006). Using the median, the presence of anxiety was de ned as GAD-7 total score of 8 points or greater.
Questions on perceived risk of COVID-19 were adopted from a previous pandemic questionnaire by Rubin et al., (2009), which consists of 6 questions measured on a 4-point likert scale (1-Strongly Agree, 2-Agree 3-Stongly Disagree and 4-Disagree). The total score on the risk perception questionnaire ranged from 0 to 24 and a median score of 16 was used in this study to dichotomise respondents into high and low risk perception categories.
Protective health behaviour was assessed with questions on protective behaviour in the last 7 days and family protective measures against COVID 19. Observance of protective behaviour was assessed using the following questions: avoidance of crowded place, wearing of face mask before leaving home, ensuring everyone who visits wash their hands before entering and having a personal sanitizer or easy access to sanitizer use. Participants were subsequently categorised into two: practicing no or one protective behaviour and practicing more than one protective behaviour.

Data Analyses
Data from the online Google forms were downloaded as Microsoft Excel document before exporting it into Statistical Package for Social Sciences (SPSS) version 24.0 and analyzed using the same software. Descriptive statistics (frequency and mean) were used to describe the socio-demographic characteristics, risk perception, anxiety, and practice of protective behaviours of the respondents. Logistic regression analyses were performed to explore the predictors of anxiety and protective behaviours during COVID-19 outbreak. Level of statistical signi cance was set at p<0.05.

Ethical consideration
Ethical approval was received from Oyo State Ethical Review Board, Nigeria (AD 13/479/1791 B ) for the study. Participation was completely voluntary (responses that were not accompanied by consent were rendered invalid). Ethical principles of anonymity and non-male cence were adhered to in the conduct of the study.   Perceived risk, anxiety, and protective health behaviours A total of 445 (37.2%) scored 8 and above using the seven-item General Anxiety Disorder scale, indicating anxiety about COVID-19. Table 3 contains participants' response on measures taken to protect themselves. Wearing a face mask (67.9%) was the commonest response followed by use or access to hand sanitizer (57.6%) while the least practiced behavior was avoidance of crowded place (9.1%). On measures taken to protect family members against the COVID-19 pandemic, regular hand washing or use of hand sanitizer was reported by less than half (44.7%), a few (2.2%) reported taking chloroquine regularly while a little below one-quarter (23.6%) reported practicing physical distancing.

Discussion
Results from this study showed that about 6 out of every 10 of respondents had a high level of perceived risk of contracting COVID-19. This is signi cantly higher compared to ndings among Myanmar population where less than one-fourth of respondents had high level of risk perception towards COVID-19 (Mya et al., 2020). The elevated perceived risk level might be attributed to disparity in country experience. As at the time of this study there was a continuous increase in the number of cases and deaths reported across various states in Nigeria, a reality which could have raised people's perception of their risk of contracting the virus. In addition, obtaining adequate information from public health professionals, government, media platforms has been associated with elevated awareness about risk during a pandemic (Van-der-Weerd, 2011;Khosravi, 2020). Therefore, aggressive media and government engagement across in the country on COVID-19 might be associated with the high-risk awareness found in this study.
About two-fth (37.2%) of the participants in this study reported high level of anxiety, a nding which is consistent with the 37.4% prevalence of anxiety found among high school students in China (Zhou et al., 2020) and 32.1% among a Nepalese population (Sigdel et al., 2020). This nding is, however, higher than the 28% obtained among Indian adults (Verma & Mishra, 2020). This relatively high level of Covid-19related anxiety may not be unconnected to the daily increase in numbers of COVID-19 con rmed cases and deaths being experienced in the country at the time this study was being carried out.
In this study, participants who were categorised as having high risk perception were almost twice more likely to be anxious about COVID-19 compared to those who had low risk perception. This nding is in tandem with the postulation that a novel risk or threat is usually accompanied with heightened fear compared with a more familiar threat (Cori, 2020). Therefore, given the novelty of COVID-19 coupled with the not-yet-successful attempts at getting a vaccine or a cure to for it, it is expected that individuals who perceive themselves as being particularly susceptible to the virus exhibit greater anxiety levels.
This study revealed that individuals within the age range of 25-34 years and 35-54 years were each associated with less likelihood for developing COVID-19-related anxiety compared to those aged 55 years and above. This is in contrast to the ndings of Lee (2020) in a study that assessed the psychometric properties of the Coronavirus Anxiety Scale among a multiethnic sample in the United States, as younger age was associated with higher coronavirus anxiety. The widespread information about the elevated vulnerability of the elderly and those with underlying conditions to COVID-19 may have resulted in this signi cantly higher anxiety in the 55 years and above age group.
Furthermore, it was found that those who were employed were more likely to experience anxiety compared with those who were unemployed; a nding which is consistent with what was reported by Verma & Mishra (2020) among Indian adults, where participants who were employed were almost two times more likely to experience anxiety. This nding seems rather opposite to our a priori position, as we had expected a contrary nding. It is, however, possible that the sudden disruption in the routine professional and economic activities of the employed respondents had resulted in signi cant anxieties in that group, while the unemployed may seem relatively adjusted to this reality already, as they do not necessarily possess the luxury of a routine job or engagement.
This study also found that having more than 12 years of formal education seemed protective of developing anxiety. This result contradicts the nding of Qiu et al (2020) in a national survey of Chinese population, where higher education was associated with greater psychological impact of COVID-19. This is, however, in tandem with the nding reported by El-Zoghby et. al (2020) among Egyptian adults, where it was found that higher levels of education was associated with lower psychological impact of COVID-19 (although the comparison was between bachelor's degree holders and postgraduate degree holders). People with higher levels of education may possess greater awareness and more authentic information about the virus, the various preventive measures and may comply better with those measures, thereby leading to greater feelings of safety. Future studies will, however, do well to explore the mediating role of factors such as risk perception, gender, and culture in this relationship.
More than half of the participants reported wearing face mask when going out to the public. Similar report was found among a South Korean population (Lee & You, 2020). Findings indicate that hand washing and use of hand sanitizer was practiced by 57.6% of respondents while, notably, social distancing was only observed by 9.1%. This nding is incomparable with the 95.8% and 88.1% level of hand washing and social distancing respectively observed among Hong Kong and Korean populations (Kwok et al., 2020). This relatively low rate of reported compliance with these protective behaviours may be a re ection of a lower perceived risk and severity of COVID-19 amongst Nigerians. In congruence with our study, however, a study among a populace in South Korea found social distancing to be the least practiced precautionary behaviour compared to wearing face masks, hand washing and use of sanitizers (Lee & You, 2020).
Males were more likely to report engaging in more than one protective behavior against COVID-19, a nding which contradicts those from studies in Chinese and Turkish adults by Duan et al (2020) and Yildirim et al (2020) respectively, where females were found to be more likely to engage in all the recommended protective behaviours against COVID-19. Certain cultural factors may be responsible for the contrasting nding in our study. However, it is also not impossible that, given the patriarchal nature of our society many male respondents might have given a socially desirable response and so this may not necessarily be an accurate representation of their practice.
Similar to what was obtained by Yildirim et al (2020) among Turkish adults as well as by Bashirian et al (2020) among health care workers in Iran, high perceived risk was associated with reported adoption of preventive behaviours in this study. This a rms the position of the Health Belief Model (Rosenstock, 1966) and similar health behaviour models, that the higher an individual's estimation of their susceptibility to a health problem, the more likely they are to adopt behaviours aimed at preventing such problems.
Participants who were employed were more likely to engage in more than one protective behaviour against COVID-19. This nding may not be unconnected to the direct relationship found between being employed and experiencing greater anxiety. The heightened anxiety associated with the coronavirus might be a major motivation for subsequently engaging in the protective behaviours. This nding may also be as a result of the fact that, some employed people may still have need to go to work sometimes and as a result be more likely to be in closer proximity with people, thus resulting in a greater perceived need to protect themselves against virus.
In addition, similar to what was reported by Yildirim et al (2020), it was found that lower level of education (not more than 12 years of formal education) was associated with lesser odds of practicing more than one preventive behaviour. Lower level of education may be associated with rejection of the very existence of the coronavirus, lower awareness of the risk of contracting the virus and subsequently the lower need for the adoption of protective behaviours (El-Zoghby et. al, 2020).
This study has limitations. Firstly, the study was conducted during the early period of the lockdown, and since the pandemic is still on, there is a tendency for a change in severity and transmission. Also, due to the use of social media platforms for the recruitment of participants, the responses were limited to people with access to the internet which could limit the generalisability of the ndings in this study to the entire Nigerian populace. However, the study ensured representation of responses from the six geopolitical zone in Nigeria including remote communities with internet access, which is deemed a strength of this study.

Conclusion
Findings from this study revealed that an appreciable proportion of the Nigerian population perceive themselves to be at risk of contracting the coronavirus and are anxious about the situation of the coronavirus. It also highlighted the different adaptive behaviour and psychological responses among the Nigerian population during the COVID-19 pandemic. The study found a very low practice of social distancing among the study population which is a key preventive measure to curb communal transmission of the virus. High risk perception of contracting COVID-19 was found to have a role in determining both anxiety and observance of protective behaviours. A balanced information will be required during a pandemic season to increase the willingness of the population in taking protective measures without leading to increased anxiety. Also, awareness activities about COVID-19 as well as interventions to mitigate the psychological burden of the pandemic should be tailored to meet the needs of speci c demographic groups.