Knowledge, practice and associated factors towards the prevention of COVID-19 among high-risk groups: A cross-sectional study in Addis Ababa, Ethiopia

Background Coronavirus disease (COVID-19) is a highly transmittable virus that continues to disrupt livelihoods, particularly those of low-income segments of society, around the world. In Ethiopia, more specifically in the capital city of Addis Ababa, a sudden increase in the number of confirmed positive cases in high-risk groups of the community has been observed over the last few weeks of the first case. Therefore, this study aims to assess knowledge, practice and associated factors that can contribute to the prevention of COVID-19 among high-risk groups in Addis Ababa. Methods A cross-sectional in person survey (n = 6007) was conducted from 14–30 April, 2020 following a prioritization within high-risk groups in Addis Ababa. The study area targeted bus stations, public transport drivers, air transport infrastructure, health facilities, public and private pharmacies, hotels, government-owned and private banks, telecom centers, trade centers, orphanages, elderly centers, prison, prisons and selected slum areas where the people live in a crowded areas. A questionnaire comprised of four sections (demographics, knowledge, practice and reported symptoms) was used for data collection. The outcomes (knowledge on the transmission and prevention of COVID-19 and practice) were measured using four items. A multi variable logistic regression was applied with adjustment for potential confounding. Results About half (48%, 95% CI: 46–49) of the study participants had poor knowledge on the transmission mode of COVID-19 whereas six out of ten (60%, 95% CI: 58–61) had good knowledge on prevention methods for COVID-19. The practice of preventive measures towards COVID-19 was found to be low (49%, 95% CI: 48–50). Factors that influence knowledge on COVID-19 transmission mechanisms were female gender, older age, occupation (health care and grocery worker), lower income and the use of the 8335 free call centre. Older age, occupation (being a health worker), middle income, experience of respiratory illness and religion were significantly associated with being knowledgeable about the prevention methods for COVID-19. The study found that occupation, religion, income, knowledge on the transmission and prevention of COVID-19 were associated with the practice of precautionary measures towards COVID-19. Conclusion The study highlighted that there was moderate knowledge about transmission modes and prevention mechanisms. Similarly, there was moderate practice of measures that contribute towards the prevention of COVID-19 among these priority and high-risk communities of Addis Ababa. There is an urgent need to fill the knowledge gap in terms of transmission mode and prevention methods of COVID-19 to improve prevention practices and control the spread of COVID-19. Use of female public figures and religious leaders could support the effort towards the increase in awareness.


Results
About half (48%, 95% CI: 46-49) of the study participants had poor knowledge on the transmission mode of COVID-19 whereas six out of ten (60%, 95% CI: 58-61) had good knowledge on prevention methods for COVID-19. The practice of preventive measures towards COVID-19 was found to be low (49%, 95% CI: 48-50). Factors that influence knowledge on COVID-19 transmission mechanisms were female gender, older age, occupation (health care and grocery worker), lower income and the use of the 8335 free call centre. Older age, occupation (being a health worker), middle income, experience of respiratory illness and religion were significantly associated with being knowledgeable about the prevention methods for COVID-19. The study found that occupation, religion, income, knowledge on the transmission and prevention of COVID-19 were associated with the practice of precautionary measures towards COVID-19.

Conclusion
The study highlighted that there was moderate knowledge about transmission modes and prevention mechanisms. Similarly, there was moderate practice of measures that contribute towards the prevention of COVID-19 among these priority and high-risk communities of Addis Ababa. There is an urgent need to fill the knowledge gap in terms of transmission mode and prevention methods of COVID-19 to improve prevention practices and control the spread of COVID-19. Use of female public figures and religious leaders could support the effort towards the increase in awareness.

Background
Coronaviruses are a large family group of viruses that cause illnesses that range from the common cold to more severe diseases found in both animals and humans [1]. A sudden outbreak of coronavirus disease 2019  caused by infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) began in December 2019 in Wuhan City, Hubei Province, China [2]. The most recent novel coronavirus, currently known as COVID-19 virus, had not been detected before and could be a new strain of coronavirus that has no antecedent known in humans [3].
Fever, dry cough, fatigue, and multiple systemic illnesses including respiratory (shortness of breath, sore throat, rhinorrhea, hemoptysis, and chest pain), gastrointestinal (diarrhoea, nausea, and vomiting), musculoskeletal (muscle ache and joint pain), and neurologic (headache or confusion) were the symptoms reported in the first cases in China [1,4,5]. The best estimate of the incubation period of the disease ranges from 1 to 14 days [3], it causes a huge burden on the healthcare system and is a global public health emergency for all countries in the world [3,6].
According to the Worldometer (As of August 4, 2020, 7 pm), COVID-19 has spread to more than 210 countries and territories and accounted for 18,543,936 confirmed cases. Of these, 63.43% (11,762,600) cases recovered and 3.77% (699,987) of total reported cases have died [7]. In Africa, according to the Center for Disease Control (CDC) dashboard, since the first case was registered in 14 February 2020 in Egypt, 968,020 cases, 629, 612 recoveries and 20, 612 deaths have been reported. According to the Ministry of Health and the Ethiopian Public Health Institute's COVID-19 pandemic preparedness and response daily situation report (4 August 2020), since the first COVID-19 case was reported on 13 March, 2020, a total of 19, 877 COVID-19 confirmed positive cases have been registered; out of these, 41.45% (8249) have recovered and 1.72%(343) have died [8].
The global community is struggling to slow down and eventually halt the spread of COVID-19 that has claimed thousands of lives and sickened tens of thousands through improving the knowledge and practice of COVID-19 prevention methods, testing and screening [9,10].
WHO reports that the best way to prevent and slow down the transmission of COVID-19 is to accurately and widely inform the public about the disease, the causes, mode of transmission, and simple prevention methods such as hand washing with soap or use of hand sanitizers, maintaining social distance and staying home to remain protected from the infection [11]. On the other hand, poor hand hygiene practices, overcrowding and close physical contacts like handshaking contribute to the fast spread of the virus within a very short period [12].
Implementing personal hygiene and public health interventions especially in priority highrisk groups is necessary to curb the spread of coronavirus [13]. Therefore, enhancing the community's knowledge and practice of COVID-19 symptoms & prevention methods will have a significant contribution to reduce the spread of the outbreak. A study conducted in United States of America on awareness, attitudes, and actions related to COVID-19 revealed that seven out of ten participants (71.7%) correctly identified three COVID-19 symptoms and 69.8% were able to identify three prevention methods [14]. Studies in China and the United Arab Emirates revealed that 73.8% of health workers identified fever, cough, sore throats and shortness of breath as symptoms of COVID-19 [15,16], while, 98% of health workers suggested washing hands with soap and water, social distance and using face masks could help in the prevention of disease transmission [15]. In line with this, 96% of Chinese residents had mentioned fever, fatigue, dry cough, and myalgia as the main clinical symptoms of COVID-19 [17]. Another study in Bangladesh also indicated that 98.7% of students believed that hand washing with soap and water followed by avoiding touching your nose, mouth and eyes with unwashed hands, the use of face mask/tissue when coughing or sneezing and wearing a clean surgical mask during their respiratory illness as the prevention methods for COVID-19 [18].
Ethiopia has showed a high commitment to prevent and slow down the COVID-19 pandemic before it causes significant health damages and an economic crisis in the community. Case identification, contact tracing, isolation and mandatory quarantine, large scale screening in high-risk groups are the actions being taken to control the spread of the disease [12]. Controlling the spread of infection in high-risk groups must be the target in the containment strategy of COVID-19 responses. To our knowledge, no study has assessed the knowledge and practices towards the prevention of COVID-19 among high-risk groups. Therefore, this study aimed to assess the knowledge, practices, and associated factors towards the prevention of COVID-19 among the high-risk groups of Addis Ababa community.

Study area and period
The study was conducted in Addis Ababa, the capital city of Ethiopia. Addis Ababa is often referred to as "the political capital of Africa" for its historical, diplomatic and political significance for the continent. The largest city in the country, with a total population of 4,793,699 (52% Female Vs 48% Male). The religious composition of the population showed the overwhelming majority were Orthodox Christians (81.8%), followed by Muslims (12.7%), Protestants (3.9%) and Catholics (1%). The city is divided into 10 boroughs, called sub-cities. Addis covers 527 square kilometere of area in Ethiopia. The population density was estimated to be near 5,165 individuals per square kilometre [19].
Addis Ababa is located at 9˚1 0 48@N 38˚44 0 24@E. It has an elevation of 2,326 meters above sea level, at the lowest point around Bole International Airport in the southern periphery and it rises to over 3,000 meteres above sea level in the Entoto Mountains to the North. The largescale assessment of the knowledge and practice in the prevention of COVID-19 was conducted from 14 to 30 April, 2020.

Study population
A population assumed to be more likely within high risk groups of acquiring COVID-19 was the study population. More likely high-risk groups were classified as groups who have high exposure due to different risk factors of COVID-19 such as, exposure to public areas, travellers, recently been overseas, close contact of COVID-19 positive confirmed cases, and being in correctional and detention facilities. Health workers (including emergency services and non-clinical staff) regardless of their risk, and street children, prisons, schools, long-term living facilities, bus stations workers etc) were also classified as a high-risk.

Study participants
Most of the respondents were formal sector employees. These formal sector target facilities included; bus stations, public transport drivers, airport, cargo, health facilities, public and private pharmacies, hotels, government and private banks, telecom centers, trade centers, children's villages, custody centers, geriatric centers, prisons, detention centers and selected slum areas where people live in crowded housing (namely; Lideta, Atobis Tera, Kirkos).

Sampling strategy and data collection
Random sampling method was applied to identify the study participants. A total, 6,007 people were included in the study.
Background and demographic information were collected using a questionnaire that was adapted from the surveillance system for COVID-19. Senior statisticians developed the data entry template using Open Data Kit. Internet-based cloud was used to transfer the data and computer tablets were used for data collection. Information on reported symptoms, travel history, any suspicious exposure to COVID-19 and attending gatherings/crowded areas was collected.
Thirty-two teams, each of them with two members, were involved in the fieldwork. Sixtyfour health professionals (Medical Doctors, Public Health Officers, Laboratories, Nurses, Environmental health professionals and Midwives) were trained for three days. The training included a classroom lecture and practical exercise on the use of the data collection tool. During the face-to-face interview, the field teams used all the necessary PPE. They also practiced all the necessary precautions such as practicing hand hygiene, to avert the risk of acquiring infection from the study participants.

Data source
The analysis was done on the data collected for the large-scale COVID-19 screening in Addis Ababa. The collected data includes socio-demographic and economic information (such as age, sex, religion, occupation and income), and, knowledge about the method of transmission and prevention for COVID-19. Precautionary practices for the prevention of COVD-19 and reported symptoms were also captured.

Definition of variables
The knowledge on the transmission of COVID-19 was measured using four items. The items were, 1) during contacts (handshake and physical contact), 2) during coughing and sneezing, 3) during breathing, and 4) Eating and drinking contaminated foods. Good knowledge was classified when the scoring was �3 (75% and above) out of four items and below three indicated poor knowledge on the transmission methods of COVID-19.
Knowledge on the prevention of COVID19 was measured using four items. The items were awareness on, 1) Physical distancing, 2) Hand washing, 3) Use of mask, and 4) Isolation and quarantine. Good knowledge of prevention was classified as when the scoring was �3 (75% and above) out of four items and score below three indicated poor knowledge on prevention methods of COVID-19.
The score of the practice was measured based on four items. These four items were, 1) practice of hand hygiene, 2) use of mask, 3) maintaining social distance and 4) staying home. A score �3 (75% and above) were classified as a good practice towards the precautionary measures of COVID-19 and a score of �2 (50% and below) indicated a poor practice.

Data quality control
To maintain the quality of the data, three days training was provided for data collectors on the aim of the study and methods of data collection. A computer-based data collection was used to control entry errors. Field level supervision was done to control the data collection and data quality.

Data management and analysis
We obtained the data from EPHI. The data was collected for the routine surveillance and screening activity among more priority high risk groups of Addis Ababa people. The collected data were directly transferred from the field staff tablet to EPHI server whenever the field team completed each questionnaire.
The items were first coded as "1" favouring the good outcome and "0" not favouring a good outcome. Then, all the items were added into one to represent the outcome. The new variable had a total score range from zero to four. A score of �3 (75%) would favour and �2 (50%) would disfavour the outcome.
A descriptive statistics was done to present the results in count, proportion and graphs. Multivariate Logistic regression was done to determine the factors associated with knowledge on transmission and prevention of COVID-19 and practice towards the prevention of COVID-19.

Socio-demographic characteristics of the participants
Six thousand seven participants were involved in the survey. Of these, the majority, 61% (3,634) were male, 41% (2,470) were in the age between 25-34 years, and around 4% (236) were above 60 years old. The mean age of the participants was 33(SD ±11.73) years. Over half (50%) of the participants were non-formal employees such as street children, clients living in centres, in children's villages, cardiac centres and cancer centres, prisons etc. While a small proportion were health care workers (14%) and public transport drivers (11%). Around, threequarters (75%) of the participants were Orthodox Christian in religion and half of the study participants (50%) had a monthly income between 3,001 and 10,000 ETB ( Table 1).

Experience of COVID-19 related illness and communication
The experience of COVID-19 related illness and communication were assessed. Among the study participants, 3% had a history of fever and an experience of respiratory illness within 14 days before the study. Of those who had experience of respiratory illness, 51% had cough, 34% had runny nose, 18% had sore throat and the rest, 11%, had shortness of breath. Free call centres and hotlines have been made publicly available to create awareness, receive information and for other COVID-19 related issues; however, only 11% of participants contacted these facilities.

Knowledge of participants on the transmission, prevention, and practice for prevention of COVID-19
The majority, 87%, of the respondents mentioned direct contact with COVID-19 confirmed patients, and 71% mentioned breathing as possible transmission mechanisms for COVID-19. Regarding prevention, 85% and 83% had mentioned hand washing and social distancing as prevention mechanisms for COVID-19, respectively. More than 80% of the respondents practiced hand washing, while 76% practiced social distancing to break COVID-19 transmission ( Table 2).

Level of knowledge on transmission modes and prevention methods, and practice for the prevention of COVID-19
Findings demonstrated that study participants have low level of knowledge on transmission modes, prevention methods and practice of prevention of COVID-19. Over half (52%; 95% CL: 51-54) of the respondents had good knowledge on transmission of COVID-19; about 60%

Factors associated with knowledge on the transmission mechanisms of COVID-19
Multiple logistic regression analysis showed that gender, age, occupation, income and having contacted the 8335 free call center were significantly associated with knowledge on the transmission mechanisms of COVID-19. The odds of being knowledgeable on the transmission mechanisms of COVID-19 among male participants were 0.87 times less likely than females (AOR = 0.87, 95% CI: 0.78-0.97). As age increases, the likelihood of being knowledgeable about the transmission modes of COVID-19 increases. Age group older than 64 years were 14 times (AOR: 14.38, CI = 6.79-30.48) more likely to be knowledgeable on the transmission of

PLOS ONE
Knowledge, practice and associated factors towards the prevention of COVID-19 among high-risk groups COVID-19 than those who were younger than 18 years old. Regarding the occupation of respondents, health workers and grocery store workers were 2.25 and 2.22 times, respectively, more likely to be knowledgeable on the transmission mode of COVID-19 than other occupations. As income increases, the likelihood of being knowledgeable about the methods of transmission of COVID-19 also increases; however, above some level of income this is no longer the case. Participants whose income was greater than 10,000 Ethiopian Birr were 49% (AOR: 0.51, 95% CI = 0.39-0.65) less likely to be knowledgeable about the transmission mechanism of COVID-19 than low-income participants (<1000 ETB). Those participants who had contacted 8335 free call center were 1.3 times more likely to be knowledgeable about transmission mechanisms of COVID-19 than those who did not (AOR = 1.30, 95% CI = 1.10-1.54) ( Table 3).

Factors associated with knowledge on the prevention of COVID-19
Multiple logistic regression analysis showed that age, occupation, income, religion, and experience of respiratory illness 14 days before the study were significantly associated with knowledge on the prevention of COVID-19. At age the age of 64 years or greater, knowledge on the prevention methods of COVID-19 was 11 times higher than at ages below 18 years old.
(AOR = 11.46, 95%CI = 5.57-23.59). Regarding the occupation of respondents, health workers and grocery store workers were four and two times more likely, respectively, to be knowledgeable on the prevention methods, while drivers were 18% less likely to be knowledgeable about the prevention methods of COVID-19 than other occupants. Those participants who experienced respiratory illness in the past 14 days were 1.6 times (AOR = 1.61, 95%CI = 1.17-2.21) more likely to be knowledgeable about the prevention methods of COVID-19 than those who did not experience respiratory illness (Table 4).

Factors associated with practices for the prevention of COVID 19
Using logistic regression analysis, the study found that occupation, religion, income, and knowledge on the transmission and prevention of COVID-19 were found to be associated with the practice of precautionary measures against COVID-19. As depicted in Table 5, health care workers were more than two times (AOR = 2.33 (95% CI: 1.85-2.93) more likely to implement the precautionary measures of COVID-19 than other occupants. Also, drivers (AOR = 1.94, 95% CI: 1.49-2.52) and guards (AOR = 1.83, 95% CI: 1.40-2.39) had higher odds of implementing the precautionary measures of COVID-19, respectively. Implementation of precautionary measures against COVID-19 was found to be lower among grocery store workers, at 0.71 (95% CI: 0.53-0.95) and students at 0.56, (95% CI: 0.32-1.00) compared to other occupants. Regarding religion, Muslims participants were at 0.42 (95% CI: 0.19-0.88) lower odds of implementing the precautionary measures of COVID-19 than others. The implementation of precautionary measures against COVID-19 was found to be lower among higher income groups. Those who earned more than 10,000 ETB per month had 0.42 lower odds of practicing the precautionary measures against COVID-19 than those with lower income. Participants who had good knowledge on mode of transmission and prevention mechanisms of COVID-19 were four times (95% CI: 4.05-5.62) and fifteen times (95% CI: 12.55-18.97) more likely, respectively, to implement the precautionary measures of COVID-19 (Table 5).

Discussion
The study found that the overall knowledge of the respondents about transmission and prevention of COVID-19 was 52% and 60%, respectively. The level of knowledge on COVID-19 transmission mode reported in our study was lower compared to reports from Ethiopia and elsewhere [20,21]. The discrepancy could be attributed to the type of research participants and the time period when the study was conducted. The practice of the prevention of COVID-19 was low (49%). Regarding the factors, gender, age, occupation, income and having used the 8335 free call center were factors associated with knowledge on the transmission of COVID-19. Age, occupation, income, religion, and experience of respiratory illness 14 days before the study were significantly associated with the knowledge on the prevention of COVID-19. The study found that occupation, religion, income, and knowledge on the transmission and prevention of COVID-19 were associated with the practice of precautionary measures against COVID-19. This study was conducted two months after the first COVID-19 positive case was reported in the country; thus we expect the study community to have good awareness by now. However, the level of knowledge that this study identified was unexpectedly low. A study conducted in Iran revealed that knowledge about the mode of transmission of the disease found a similar figure of 56% [9]. In contrast, a study conducted in China revealed that the overall knowledge of COVID-19 was high at 90% [17]. Our study has found better knowledge on the prevention of COVID-19 than a study conducted in Thailand with a result of, 73.4% on the knowledge of prevention and control of COVID-19 [22].
The level of practices for the prevention of COVID-19 in Ethiopia was 49%, which was lower than the study conducted in Iran, 71%, but much higher than the study conducted in Thailand, 17% [9,22]. The socio demographic characteristics of the study participants, study differences in methodology, study time, and area of the study could explain the differences, while the low-levels of use of the 8335 call centers and hotline may have limited people from acquiring knowledge in the subject matter.
In this study, age, occupation, and income have shown significant association with the knowledge on the transmission and prevention mechanisms of COVID-19. Gender, religion, and contacting the 8335 free call center were significantly associated with the knowledge on the transmission mechanisms of COVID-19, while participants who had a respiratory illness 14 days before the study were significantly associated with knowledge on the prevention mechanisms of COVID-19. The findings of women being more knowledgeable than men about the transmission of COVID-19 was in line with the studies conducted in China and Iran [9,17].
Being a health professional was associated with better knowledge about COVID-19. This finding is supported by the study conducted in Iran [9]. A further similarity was found between two studies on the association between age groups and knowledge, with both studies showing better knowledge at older ages. The increase in income was inversely associated with being knowledgeable about the transmission methods of COVID-19. In addition, the study revealed that, participants who had contacted the 8335 free call center had better knowledge on the transmission methods of COVID-19, and who had experience on respiratory illness for 14 days before the study had better knowledge on the prevention of COVID-19.
The study also found that occupation, religion, income, and knowledge on the transmission and prevention of COVID-19 were found to be associated with the practice of precautionary measures against COVID-19. Health care workers, drivers, and guards reported practicing prevention mechanisms such as hand washing, social distancing, use of facemasks, and staying home. On the other hand, participants such as grocery/store workers, students and highincome groups were not practicing the prevention methods of COVID-19 in the proper ways. Contrary to our study, having health-care-related occupation was found to be significantly associated with lower practice towards the prevention of COVID-19 [9]. The hand hygiene practice had strong relationship and it is widely practiced among Muslim religion followers. However, the measurement of precautionary measures were estimated in combined with other three preventive practices that lead to pulling the result towards negative; such as the use of facemask, physical distancing and staying at home. Our study revealed that having good knowledge on the transmission and prevention of COVID-19 was found to be associated with good practices on COVID-19 prevention. This finding is not in line with the study conducted in China, which revealed that study participants who had a good knowledge score on the prevention mechanisms of COVID-19 were more likely to visit crowded places and failed to wear face masks [17].

Strengths and limitations
Our study was large and had a broad scope with several strengths. It was based on a large sample size and focused on participants from areas that were likely to be high risky; therefore, the findings can be generalized across the city's population. Unlike other studies, we did face-toface interviews in this challenging situation. However, it has also limitations. The study did not gather important characteristics of the study participants, such as educational status, marital status and visiting public places, smoking etc that could affect knowledge on COVID-19. Further, the initial assessment was done as part of the routine COVID-19 response rather than specifically for this study.

Conclusion
The study pointed out that population who had high risk of acquiring COVID-19 had moderate knowledge about transmission mode, prevention mechanisms and practice on the prevention of COVID-19. Awareness creation programs on targeted groups, such as men, youth and those with higher incomes are required in order to improve the knowledge and their practices on the prevention of COVID-19. Advocacy of the hand hygiene, physical distancing, use of facemask and others strategies should fulfill the characteristics of the targeted individuals.