Unsatisfactory COVID-19-Related Knowledge, Attitudes and Practices among Undergraduate University Students in Uganda: An Online Cross-Sectional Survey

Background: This study examined the COVID-19-related knowledge, attitude and practices among undergraduate students in Uganda. Methods: An online cross-sectional survey was conducted from 12 th -19 th June 2020 among undergraduate students using a standard questionnaire designed using Google Forms sent via WhatsApp Messenger. The data were analyzed using SPSS. Pearson’s Chi-square test was used to test the differences and odds ratios with their 95% confidence intervals were used for quantifying the association between independent and dependent variables. p < 0.05 was considered significant. A cut-off score of ≥80% was used to denote sufficient knowledge, positive attitude, and good practices. Results: Of the 161 respondents, 102 (63.4%) were males with a mean age of 24.2 (5.0) years. The majority 121 (75.2%) were pursuing health-related programs and overall, 110 (68.3%) had sufficient knowledge while 76 (47.2%) had a positive attitude and good practice each. Knowledge and attitude were significantly associated with health-related programs (AOR 4.78, 95% CI 2.06 - 11.07; p < 0.001) overall. Therefore, concerted efforts to provide tailored health education and behaviour change communication are needed.


Introduction
The Coronavirus Diseases of 2019 (COVID- 19)  COVID-19 is a highly infectious disease believed to spread through respiratory aerosol generated through coughing and sneezing by an infected person [2] [3]. Evidence also suggests that a proportion of asymptomatic and mildly ill COVID-19 cases do also spread the disease [4]. However, the mode of transmission, just as its pathophysiology is still cloudy. The main clinical symptoms of COVID-19 include fever, dry cough, fatigue/myalgia, and difficulty in breathing.
Other symptoms are sore throat, diarrhoea, loss of taste or smell, and arthralgia [5] [6] [7] [8]. About 20% of patients can develop severe disease characterized by acute respiratory distress syndrome (ARDS) requiring oxygen therapy and about 5% will require intensive care unit treatment including mechanical ventilation [9].
To date, there is no proven antiviral treatment or vaccine against SARS-CoV-2, the virus responsible for the current COVID-19 pandemic [8]. Containment of the pandemic, therefore, centres primarily on strong and strict infection control measures arising from the knowledge of the mechanisms of its transmission, to minimize the spread of the infection within the population [10] [11] [12]. The primary preventive measures are largely behavioural and include regular hand washing, social distancing, respiratory hygiene [11] [12], and use of face masks in public, compliance with which is largely dependent on the change in behaviour which is shaped by one's knowledge about the disease and their attitude [6] [8]. While there has been a breakthrough in vaccine development, with several vaccine candidates currently under emergency use in different parts of the world, the global impact is yet to be realized. Uganda launched the COVID-19 vaccination using the Astra-Zeneca vaccine on the 10 th of March 2021, targeting mainly frontline healthcare workers, the elderly and persons with preexisting health conditions associated with increased risk of COVID-19 morbidity and mortality. Vaccine uptake has generally remained low with only approximately 180,969 having been vaccinated as of 10 th April 2021-a month since its launch, and the country is yet to realize any meaningful impact of the vaccine.
University students are among the most exposed to a wide range of information, including social media where misinformation abounds, and yet are not among the current targets of the COVID-19 vaccination drive. This has the potential of clouding their understanding of the current COVID-19 pandemic [13], with dire consequences. To date, there is a paucity of data on knowledge, attitude, and practices related to the current COVID-19 pandemic among undergraduate students undertaking various study programs in Sub Saharan Africa, Uganda inclusive. This study, therefore, aimed to assess the COVID-19-related knowledge, attitude, and practices among undergraduate university students in Uganda to recognize socio-demographic factors which can be useful in sanctioning behaviour change.

Study Design
This was an online cross-sectional survey conducted from 12 th -19 th June 2020, to determine the knowledge, attitude, and practices related to COVID-19 among the respondents.

Study Setting
The study was conducted at three Universities in the northern region of Uganda, namely; Lira, Gulu and Muni Universities. The northern region comprises three sub-regions with distinct indigenous populations and is located approximately 360 -600 km from Kampala, Uganda's capital city.

Study Population
The study targeted students undertaking various undergraduate programs at different years of studies at the three universities.

Sample Size Determination
The sample size was estimated at 285 using the single population proportion formula [14] based on an average previous estimate of good knowledge related to COVID-19 of 79.9% [15], with a marginal error of 5%, a standard normal value corresponding to 95% certainty, and a non-response rate of 15%.

Study Instrument
A precoded structured questionnaire was used to collect quantitative data, developed with questions adapted from previous peer-reviewed KAP surveys [6] [16] [17] and the national COVID-19 Diagnosis and Treatment Protocol [18]. The questionnaire consisted of four parts: 1) Socio-demographic characteristics, 2) Knowledge questions, 3) Attitude-based questions, and 4) Practice questions.

Sampling Procedure and Data Collection
Because the country was in a lockdown with all learning institutions, including the universities, closed, we opted for an online platform using WhatsApp Messenger (Facebook, Inc., California, USA) to enroll potential participants. We employed a convenient sampling method and identified all the existing What-sApp groups of the undergraduate students in the three universities. The Google Form link to the questionnaire was sent via WhatsApp Messenger (Facebook, Inc., California, USA) to identified students' WhatsApp groups.

Independent Variables
The independent variables included respondents' age; gender; year of study; institution, academic program, previous qualifications, residence, and source of information.

Dependent Variables
Respondents' COVID-19-related knowledge, attitude and practices formed the dependent variables.
The knowledge domain was assessed using a 15-item question, aimed at assessing the general knowledge about the cause, transmission, clinical presentation, risk factors and prevention of COVID-19 answered largely on a true or false basis with an additional option of "I don't know" for those who are uncertain. Each correct response weighed 1 point and 0 points for incorrect or uncertain responses, giving a maximum score of 15 and a minimum score of 0. The higher the points scored, the more knowledgeable the student.
The attitude domain was assessed using a 10-item question, each based on a 5 Likert scale with responses of strongly disagree, disagree, not sure, agree, and strongly agree. A score of 1 was assigned for a response reflecting a positive attitude and 0 for a response reflecting a negative attitude or neutral response, giving a maximum score of 10 and a minimum score of 0.
The practice of COVID-19 prevention was assessed using a 7-item question to assess the actual uptake of, and adherence to, the primary preventive measures for COVID-19. Each item was based on a response of "always", "sometimes" and "never" or as specified. A weight of 1 was given for a response reflecting good practice and 0 for a response reflecting poor practice, giving a maximum score of 7 and a minimum score of 0.
Students' KAP levels will be defined as "good"/"sufficient" or "poor"/"insufficient" based on cut off points adapted from that used in previous KAP studies [17] [19]. For knowledge assessment, scores of 80% and above will be regarded as sufficient knowledge, while those below 80% will be taken to reflect insufficient knowledge. For attitude, scores of 80% and above will be regarded as reflecting a good attitude, while scores below 80% will be regarded as a poor attitude. For practice, students with scores of ≥80% will be categorized as having good pre-Open Journal of Preventive Medicine ventive practice, while those with scores <80% will be considered as having poor preventive practice [20].

Data Management and Statistical Analysis
The completed questionnaires were extracted from Google Forms into Microsoft Excel 2010 and the data was exported to Statistical Package for Social Sciences (SPSS) software package (SPSS for Windows, Version 16.0. Chicago, SPSS Inc.) for analysis. Descriptive statistics were calculated for demographic characteristics and responses to given KAP questions, and summarized as frequencies (percentages) for categorical variables and mean (standard deviation) or median (interquartile range) for continuous variables. A cut-off of 80% was used to categorize sufficient knowledge, positive attitude and good practice. Pearson's Chi-square test was used to test the differences between socio-demographic characteristics with regards to KAP, and multivariate logistic regression analysis was used to determine the association between the socio-demographic variables and KAP. Variables with p-value < 0.3 on bivariate analysis were entered into the multivariate model. The adjusted odds ratios (AOR) and their 95% confidence intervals (CIs) were used to quantify the associations between socio-demographic variables and KAP. p < 0.05 was considered for statistical significance.

Ethics Approval Considerations
The study was approved by the Lacor Hospital Institutional Research and Ethics Committee [approval number LHIREC 0165/05/2020]. Online informed consent was made accessible to the respondents and informed consent was obtained before participating in the study. The questionnaire was designed to be anonymous and respondents were informed that all information provided would be anonymous and confidential. The trust in these sources varied widely, with the ministry of health sources being rated as the most trustworthy by 136 (86.3%) of the respondents, while only 21 (13.0%) of the respondents rated social media as a trustworthy source of COVID-19 information (Table 1).
The mean knowledge score was 12.0 (SD 1.5) with an overall knowledge score of 68.3%, indicating a low knowledge of COVID-19 (Table 3) (Table 3).
On multivariate analysis, the study program was the only factor that significantly influenced the level of COVID-19-related knowledge among the respondents. Students pursuing health-related programs were more likely to have sufficient knowledge about COVID-19 than those pursuing non-health-related programs (AOR 4.78, 95% CI 2.06 -11.07; p < 0.001). Respondents' age, year of study, previous qualifications, and source of information did not significantly influence knowledge levels (Table 3).
The overall positive towards COVID-19 was 47.2% (76/161)-indicating that the attitude among the undergraduate students was generally negative (Table 5).
There was a statistically significant difference in attitude between students undertaking health-related programs and those undertaking non-health-related programs (χ 2 4.62; p = 0.032). There was also no statistically significant difference in attitude with regards to the rest of the socio-demographic characteristics on bivariate analysis (Table 5).
On multivariate analysis, students undertaking health-related programs were more likely to have a positive attitude towards COVID-19 than those pursuing Open Journal of Preventive Medicine

COVID-19 Prevention Practices among the Respondents
In responding to questions assessing observance of the COVID-19 preventive practices, only 77 (44.7%) had avoided a crowded place in the period preceding the study and just over half, 88 (54.7%) had worn a mask when leaving home.
When asked about what they would do if they had a fever and dry cough, the Open Journal of Preventive Medicine majority, 159 (98.8%) would self-quarantine or go to the hospital for treatment, while 2 (1.2%) would hide or go to a clinic or buy some medicine and self-medicate. Participants' responses to the other COVID-19 preventive practices are as summarized in Table 6.
Overall, only 76 (47.2%) of the respondents demonstrated good COVID-19 prevention practices (Table 7). There was a statistically significant difference in COVID-19 prevention practices by gender on bivariate analysis (χ 2 5.49; p = Open Journal of Preventive Medicine  (Table 7). On multivariate analysis, gender remained the only socio-demographic characteristic that independently influenced practice, where male students were significantly more likely to exhibit good preventive practices than their female counterparts (AOR 2.37, 95% CI 1.19 -4.73; p = 0.014) ( Table 7).

Discussion
The current COVID-19 pandemic has had devastating global effects on all spheres of human life since first reported in Wuhan, China, in December 2019 and subsequently declared a pandemic by WHO in March 2020. In the absence of proven vaccine or treatment, strong and strict infection control measures to prevent its spread are crucial, compliance to which is influenced by individuals' knowledge, attitude, and practices. To date, there has been limited published data on KAP among university students in Sub Saharan Africa (SSA) and to the best of our knowledge; this is the first study in investigating COVID-19-related KAP among university students undertaking programs beyond the medical field.
The study assessed COVID-19-related KAP among students in three public universities in Uganda to recognize socio-demographic factors which can be useful in sanctioning behaviour change.

COVID-19-Related Knowledge
Overall, 68.3% of the students had sufficient COVID-19-related knowledge, a level much lower than that reported among medical students in Uganda (90%) [21], Iran (87%) [22], China (82.34%) [16] and Pakistan (79.9%) [15]. However, the knowledge level among undergraduate students in our study is higher than 10.5% and 37.5% reported among students in Bangladesh [23] and Saudi Arabia [24] respectively. This could be attributed to the difference in the study populations      In this study, COVID-19-related knowledge was significantly associated with the study program pursued, being significantly higher among students undertaking health-related programs than their counterparts. This finding is consistent with that reported in other studies [15] [16] [24] and has been attributed to the training of these students in clinical medicine and public health, as well as their sense of duty and responsibility as a candidate medical professional [25]. Contrary to findings from other pieces of literature [15] [26], age, education, institution, year of study, and source of information did not significantly influence the level of knowledge about COVID-19 among undergraduate university students in the current study.
The study found a high level of knowledge regarding COVID-19 aetiology, its main route of transmission, the incubation period and the groups at high risk.
This finding is similar to that reported among undergraduate students in Pakistan [15] and Saudi Arabia [24]. This is important because knowledge of the aetiology, transmissibility of, and susceptibility to, the disease is considered as the first step of individuals education and behaviour change communication in adopting the preventive measures that limit its spread [27]. In the same breath, where knowledge of myalgia as one of the main symptoms of COVID-19 was limited. Given that COVID-19 is largely a respiratory disease, this finding may thus not be surprising, since the triad of fever, cough and shortness of breath is a common clinical presentation of respiratory tract infections usually experienced by many in Uganda.
The most cited sources of information on COVID-19 were the national ministry of health; the news media like TV, radios and newspapers, and social media. Reliance on messages through official government sources through TV, radio and official websites which are usually censored has previously been reported in Uganda [21] and remains a widely used source of information in other settings [28]. The reliance on social media as found in the current study is not unexpected since it has become a widely popular source of information among many population groups [8] [15] [21] [27] owing to its ease of accessibility and cost-effectiveness. However, the spread of misinformation through social media remains of great concern [8] [29], a situation which, as asserted by Shu et al.
(2017), can have devastating effects on society [30]. Importantly, while many students in the current study used social media for COVID-19 information, only 13% trusted its worthiness. This is reassuring as it means that those engaged with these channels may scrutinize the information presented to them. Nonetheless, social media continues to play a vital role in educating the public about COVID-19 [24].

Attitude Related to COVID-19
This study reveals a generally negative attitude related to COVID-19 among the undergraduate students, only 47.2% had a positive attitude overall. The attitude was significantly influenced by the study program and the year of study. Students undertaking health-related programs were three times as likely to have a positive attitude towards COVID-19 as those pursuing non-health-related programs. Similarly, students in their second year of study onwards, regardless of the program being undertaken were significantly more likely to have a positive attitude towards COVID-19 than the first-year students. This finding bodes well with that conducted among purely medical students [21]. Unlike the report by other authors [5] [16], there was no statistically significant difference in attitude score by gender among the undergraduate students in the current study.
When asked about the ability of the country to contain the current COVID-19 pandemic, the majority of the respondents held a pessimistic attitude: only 34.8% had confidence that Uganda can win the battle against COVID-19, while only 41.0% were confident that the school environment was enabling in preventing the spread of COVID-19. This finding mirrors that from previous studies in Pakistan [15] and Uganda [17] where a large number of respondents did not believe in the ability of the government to combat the COVID-19 epidemic but contrasts with findings from China, Saudi Arabia and Nepal [5] [6] [26].
This is rather surprising given the government's prompt response in taking stringent control and precautionary measures against COVID-19 but is in tandem with the assertion that people tend to express negative emotions, such as anxiety and panic, during a pandemic that could affect their attitude [31]. The pessimistic attitude could have also been informed by the ongoing rise in the number of cases in the country, the general lack of trust in the national health system, and the information gap between the population and the government.
There is thus a need for authorities in government and the health ministry to instil confidence in the common man, in general, using all the available channels.

Practices Related to COVID-19 Prevention
Overall, only 42.7% of the respondents in the current study demonstrated good practice towards COVID-19 prevention. Male students were twice as likely to exhibit good preventive practices as females. This finding contrasts with that on Middle East Respiratory Syndrome Coronavirus (MERS-CoV) among the Saudis population where males were significantly less likely than females to take safety precautions and prevention [27]. On the other hand, a report by Ping and colleagues (2020) showed no statistically significant difference in preventive prac-Open Journal of Preventive Medicine tices between groups by gender [16]. The finding of positive practice among the male gender has, however, previously been reported [32] and is of significance in that generally, men and late adolescents are more likely to engage in risk-taking behaviour [33] [34] as was evidenced in China where male gender was significantly associated with the practice of defying the principle of social distancing [6].
SARS-CoV-2 is a highly infectious agent just like most viral infections and has been documented to be highly contagious among people in close proximity [35].
However, while close to three-quarters (72%) of the respondents had refrained from shaking hands in the period preceding the study, more than half either frequently or occasionally associated with a crowded environment-defying the recommended social distancing measures. Likewise, up to 45.3% of the respondents either inconsistently or never wore a face mask when leaving home for public places, while 43.5% inconsistently washed their hands before and after touching anything. This is concerning and comes on the backdrop when the national ministry of health has produced a guide on COVID-19 to provide residents with facts and precautionary messages. It, however, calls for more concerted effort in intensifying education and behaviour change communication to the population, especially in as far as dissemination of the available information, education and communication (IEC) materials are concerned. This approach is supported by evidence from a Saudi's study where the provision of education and outreach materials to increase public understating of the disease and influence behavioural change was associated with high rates of adoption of good and safe practices towards COVID-19 [26].
The majority of the respondents in the current study would self-quarantine or go to the hospital for treatment if they had a fever and dry cough, while 1.2% would hide or rather go to a clinic or buy some medicine and self-medicate.
Similarly, the majority of the undergraduate students would proactively report to the authority and stay in quarantine as required if they had close contact with a confirmed case of COVID-19, though 3.1% would rather keep the information to themselves. Concerning what they would do if a colleague or one of their classmates who got cured of COVID-19 wanted to meet them or were to attend class with them, the majority of the respondents would meet or attend class with their classmates and show them more kindness, while 3.1% would rather find an excuse to keep away from such a colleague or classmate. While these findings are largely reassuring, the wrong actions that would be demonstrated by the minority cannot be ignored, especially when it causes stigmatization, since this would hinder other students from coming forward.

Limitations of the Study
The convenience sampling technique used in this study could have introduced sampling bias which, together with the relatively small sample size, could limit the representativeness of the study findings.

Conclusion
Overall, the current findings suggest an unsatisfactory COVID-19-related KAP among undergraduate students. Although the government has taken major steps to create public awareness through various channels to limit the spread of the disease, there is a need for more educational program and behaviour change communication, tailored to target undergraduate students, especially in the non-health-related disciplines.