Socio-demographic characteristics in ﬂ uencing knowledge, attitude and preventive practices of COVID-19 among Ghanaians: A cross-sectional study

Background: In recent times, the novel coronavirus disease is one of the most challenging public health burdens. Low-and-middle-income countries including Ghana have not been spared by the infection. Several measures are being undertaken by countries to control and prevent the future waves of the disease. Previous studies have established the importance of knowledge, attitude and practices towards an infection as a remedy for controlling the spread of such infection. However, further studies are required to ascertain the socio-demographic characteristics in ﬂ uencing the knowledge, attitude and preventive measures of COVID-19. Methods: This study investigated the socio-demographic characteristics in ﬂ uencing knowledge, attitude and preventive practices towards COVID-19 among the Ghanaian population. A cross-sectional design was adopted to recruit 369 participants for the study utilizing convenience and snowball sampling methods. With the use of an electronic semi-structured questionnaire and social media platforms, an online survey was conducted from March 26 to May 1, 2020. Descriptive statistics, independent t-test, chi-square test, one-way analysis of variance (ANOVA) and linear regression were used to analyze data. Results: The average mean percentage of good knowledge, positive attitudes and good preventive practices regarding the novel coronavirus disease was 83.2%, 88.2% and 69.2%, respectively. In relation to occupation, there was signi ﬁ cant difference (t = -2.984, p-value = 0.0030) in COVID-19 knowledge. Income level was signi ﬁ cantly associated with attitudes towards COVID-19. Male gender (vs female, β = -1.276, p-value = <0.00001), less than ¢500 income earners (vs above ¢1000, β = -0.845, p-value = 0.017) were associated with lower COVID-19 preventive practice score while being the 42-64 age category (vs 18–25-year group, β = 1.128, p-value = 0.028), health worker (vs non-health worker, β = 0.816, p-value = 0.001) and belonging to Ga ethnic group (vs Akan ethnic group, β = 0.89, p-value = 0.027) were associated with higher scores. Conclusion: Although this study found higher knowledge and attitudes toward the novel coronavirus disease, it did not re ﬂ ect in higher preventive practices towards the infection. In our ﬁ ght against this infection in Ghana, COVID-19 preventive measure educational campaigns targeting the male gender and low-income earners should be considered and implemented.


Introduction
Since the emergence of the novel coronavirus disease  in December 2019 from Wuhan city in China, it has spread to many other countries around the world including those on the African continent [1]. The World Health Organization (WHO) on January 30, 2020, declared the outbreak of the disease as a Public Health Emergency of International Concern, serving as a clarion call to the international community to provide political, fi nancial and technical support to a public health emergency [2]. Subsequently, COVID-19 was declared as a pandemic on March 11, 2020, due to its widespread across the globe [3].
In Ghana, the Ministry of Health (MOH) in a release on March 12, 2020, announced the fi rst two confi rmed cases of COVID-19 [4]. After the pronouncement of these initial cases, the Government of Ghana (GoG) put in place numerous measures including the closing of borders, schools, churches, mosques and ban of all social gatherings [5]. Citizens were also advised by GoG to practice precautionary measures such as the washing of hands with soap, sanitizing, social distancing among others. Notwithstanding, as of May 1, 2020, Ghana was among the top African countries that had recorded most cases of COVID-19 with 2074 confi rmed cases, 212 recoveries and 17 deaths [6].
The battle against COVID-19 by GoG and Ghanaians as a whole continued through various phases with strict measures such as partial lockdown in various parts of the country such as Greater Accra, Greater Kumasi and Kasoa Metropolis for three weeks. This was accompanied by the compulsory wearing of nose masks, particularly in the Greater Accra region. In the wake of all these measures put in place by GoG, people's adherence to these measures are signifi cant and is chiefl y affected by Knowledge, Attitude and Practices (KAP) towards COVID-19. This is per the KAP theory [7,8].
The second wave of the COVID-19 in India can be attributed to many factors including neglect of COVID-19 proper behaviour [9]. And it is obvious the Indians let their guards down and did not follow the protocols and guidelines of the previous wave of COVID-19. Lack of knowledge, bad attitudes and poor preventive practices towards COVID-19 among them may have also accounted for the violation of the COVID-19 appropriate behaviour. The knowledge, attitudes and practices of standard preventive measures are signifi cant in the prevention of future waves in India and other parts of the world.
According to some studies [10,11]. Conducted during the SARS outbreak in 2003, knowledge and attitudes toward infectious diseases like COVID-19 have a causal connection with the level of panic emotion among the general population and can subsequently complicate the efforts made to prevent the transmission of the disease. So, some basic knowledge about the COVID-19 and its modes of transmission as well as basic hygiene principles to reduce the panics related to the disease and its spread is essential.
In the specifi c case of the COVID-19 virus, SARS-CoV-2, the routes of transmission remains a paradox [12]. Hence, there is an urgent call globally to research to unravel these contradictory routes of transmission and other factors including demographic characteristics that may contribute to the transmission of COVID-19. These investigations are essential to inform policy, and consequently, policymakers and public health agencies may need to upgrade their precautionary approaches to disrupt all the probable modes of transmission based on new scientifi c evidence [12].
Currently, studies have been done in some countries such as China [13]. Egypt [14]. Iran [15]. Kenya [16]. On KAP towards COVID-19 among the general population. However, the situation in Ghana has not been adequately assessed, especially the investigation of socio-demographic characteristics infl uencing the knowledge, attitude and practices of the COVID-19 pandemic.
Given this and all the precautionary measures put in place by the GoG together with daily education on television, radio, print and social media platforms concerning COVID-19, the study investigated knowledge, Attitude and Practice towards the novel coronavirus disease among Ghanaians. It also ascertains the socio-demographic characteristics infl uencing the knowledge, attitude and preventive practices of the novel coronavirus disease in the country.

Study design
A cross-sectional study was conducted from 26 March to 1 st May 2020. The study adopted an online survey approach among Ghanaians, due to the restricted movement imposed on Ghanaians during the COVID 19 pandemic period. Social network platforms such as WhatsApp, Facebook, Instagram and Twitter were used as the channels for reaching people to respond to the survey.

Study area
The study was conducted among Ghanaians through online social media platforms. Ghana is a West African Country bordered by the countries; Burkina Faso, Ivory Coast and Togo in the north, west and east respectively. However, the south is bordered by the Atlantic Ocean. Ghana is divided into 16 administrative regions with Accra as the national city. The population is estimated to reach 31,072,940 in the year 2020 with Akans and Christians forming 47.5% and 71.2% of the population correspondingly. The health system is manned by the Ghana Health Service offering primary, secondary and tertiary health care in various regions.  The questionnaires were multiple choice and limited format.

Data collection technique
The data collection instrument was reviewed by professionals including medical offi cers, medical laboratory scientists and disease control offi cers to ascertain its reliability and validity.
This was followed by a pretesting. The tool was shared with authors' social contacts on WhatsApp, Facebook, Twitter and Instagram. Respondents were also encouraged to share with their networks after completing the survey.

Demographic characteristics of respondents
The study received participation from three hundred and thirty-nine (369) people. Out of this, the majority, 231 (62.6%) were males. A greater number, 156 (42.3%) of participants were within the 26-33 age bracket while the least, 26 (7.1%) were in the 42-64 age category. More than half, 207 (56.1%) were not health workers. Almost all, 356 (96.5) were Christians and had attained tertiary education. Also, most, 221 (60.0%) and 215 (58.3) of study respondents earned an average income of more than ¢1000 and were Akans, respectively. Again, the majority, 123 (33.3%) respondents were residents of the Greater Accra region, this was followed by the Ashanti and Eastern region (Table 1).

Participants knowledge of COVID-19
The average mean of knowledge on COVID-19 among study participants was 18.3 ± 2.36 out of a maximum score of 22, showing a mean percentage of good knowledge of 83.2% ( Figure   1). The correct answer rate of questions (K1-K12) relating to knowledge on COVID-19 ranged from 4.3-99.2%. However, the

Attitudes of the participant's towards COVID-19
The average mean of participant's attitude towards COVID-19 was 5.29 ± 0.88 out of a maximum score of 6, representing a mean percentage positive attitude of 88.2% ( Figure 1). The majority, 61.5% of above ¢1000 per month income earners had good attitudes towards the novel disease ( Figure 2). Based on the questions to ascertain attitudes towards  (Table 3).

Preventive practices against COVID-19 by study participants
The average mean of COVID-19 preventive practices was  (Table 4).

Differences between knowledge, attitudes and practices towards COVID-19 by socio-demographics characteristics
There was signifi cant difference in knowledge on COVID-19 in relation to the occupation (t = -2.984, p-value = 0.0030) of study participants. Again, signifi cant differences existed in attitudes with respect to income level (F = 4.12, p-value =    *, ** and *** represent good practices and were scored 1, 2 and 3 respectively.  (Table 7).

Discussion
This study investigated the knowledge, attitudes and practices of Ghanaians towards the novel coronavirus disease through an online survey. Averagely, participants of this study showed a high knowledge of the disease. This fi nding was not different from a similar study conducted among the Egyptian population [14]. Additionally, an extremely higher mean score of knowledge was recorded in a study conducted among Chinese residents [17]. All these results show the concern that was attached to the disease. Besides, the high COVID-19 knowledge level may also have resulted from the educational level of participants who took part in the study [18].
Again, our study revealed that health workers had signifi cantly higher knowledge scores compared to non-health workers. This result is expected since health workers are the front-liners in the fi ght against the coronavirus pandemic. However, a study conducted by Yue et al. showed that being a health care worker had a signifi cant impact on COVID-19 knowledge, but there was no signifi cant difference in knowledge of health workers and non-healthcare workers [19]. Signifi cant differences in COVID-19 knowledge rather existed in age groups, marital status, gender and educational level [19].
The fi ndings of our study depicted a high mean percentage attitude score of almost 90%. A similar result was reported in a population-based study done in Iran [20]. Further, both recent and previous studies conducted within Africa and other parts of the world within this pandemic period have reported high positive attitudes towards the novel coronavirus disease [17][18][19]21]. Knowledge is usually associated with attitudes. Therefore, the high COVID-19 knowledge among participants from different countries may have infl uenced the general attitude of people worldwide. A study conducted in Nigeria Earlier modelling studies had predicted the severity of the novel coronavirus pandemic globally; the severity is expected to be intense in low-income settings and among people earning low income [22,23]. In our study, we found that attitudes towards COVID-19 were signifi cantly associated with the income level of participants. And, it was confi rmed in our study that a positive attitude towards the COVID-19 was signifi cantly higher among participants who received higher income compared to those who received lower income.
Also, positive attitudes towards COVID-19 were found among the majority of above ¢1000 income earners. Therefore, it is obvious that negative attitudes among low-income earners may be contributed by the cost of COVID-19 preventive items, such as nose masks and sanitizers.
In a study [24]. Conducted among ethnic minorities in Hong Kong, participants expressed some misconceptions on the prevention of novel coronavirus disease; however, they had positive attitudes towards the approved preventive measures of the infection including wearing masks and social distancing.
Though in our study, ethnicity was not associated with attitudes towards COVID-19, signifi cant differences in attitude were among ethnic groups, implying that an ethnic group had better attitudes towards COVID-19 compared to the others.
These fi ndings are expected because earlier studies [25]. Have reported a relationship between ethnicity and the spread of According to a recent Chinese study, COVID-19 preventive measures were practiced by participants who earned higher income [24]. However, the results of this study revealed that participants who earned between ¢500-1000 as income practiced COVID-19 preventive measures compared to those who received less than 500 and above ¢1000. Furthermore, our study showed a strong association between the income level of above ¢1000 and lower practice scores. These results suggest the unpredictability of human behaviour, which may be affected by factors such as knowledge and complacency.
This study, fi nally, found out that being within the age category of 42-64 years and belonging to a Ga ethnic group were associated with a higher COVID-19 preventive practice score. The numerous data that churned out at the early stage of the COVID-19 pandemic revealed that the older age group were at high risk compared to the youth [27]. So, educational and preventive campaigns were all targeted at this age bracket.
This may be the reason for the higher practice score among them [28]. The Ga ethnic group are inhabitants of Accra, the epicentre of the novel coronavirus in Ghana. In essence, it was expected that this group may adhere more to the preventive measures compared to the other ethnic groups in Ghana.

Limitation of the study
Due to the COVID-19 pandemic, the study was limited to only participants who could read and write, since the survey was made available on solely social media platforms.

Conclusions
I n summary, our research suggests that knowledge and attitudes towards novel coronavirus disease were high among the Ghanaian population; however, preventive practices towards the infection was just above average. Occupation and income level had a signifi cant relation with knowledge and attitudes towards COVID-19 infection, respectively. Also, COVID-19 preventive practices had a signifi cant association with gender, income level, occupation, age and ethnic group. Educational interventions on COVID-19 should be targeted towards the population involved in non-health related occupations. Good attitudes towards the pandemic should be encouraged among the low-income earners of the population. The male gender and low-income earners should be admonished to practice measures towards the curbing of the novel coronavirus disease.

Declarations
Ethics approval and consent to participate: The Ghana Heal th Service Ethical Review Committee (GHS-ERC) guidelines and standard operating procedures exempt studies that present minimal risks such as online surveys, public behaviour observations and interviews from ethical review and clearance. Hence, the study did not require ethical approval. However, study participants' anonymity and confi dentiality were ensured since the survey did not require any identity before partaking. An informed consent section was created as part of the questionnaire to seek consent before respondents proceeded to complete the online Study Questionnaire.

Availability of data and material
The study data and materials will be provided upon request. wrote the fi nal manuscript. All authors read and approved the fi nal manuscript.