COVID-19 Knowledge, Risk Perception, and Precautionary Behavior among Medical Students in Egypt

Background: During clinical training, medical students may have close contact with COVID-19infected patients. Lack of proper knowledge about COVID-19 dynamics and prevention makes them more liable to infection. Objective: To evaluate the level of awareness, precautionary behavior, and risk perception regarding COVID-19 and identify the factors motivating and hindering medical students to take preventive measures. Methods: A cross-sectional research study was conducted using an online questionnaire of medical students at Ain Shams University, Cairo, Egypt between June 1 and June 30, 2020. The questionnaire design was based on a literature review mainly on the domains of the standard risk perception questionnaire. Results: A total of 351 medical students (mean age of 21 ± 2 years) completed the questionnaire. Over 94% had a high level of knowledge, 96.9% scored at high level of practicing precautionary behavior, and 86.1% had a high level of risk perception. Precautionary behavior practices had significant negative correlation with the knowledge scores associated with COVID-19 and a significant positive correlation with risk perception score. There was a significant association between the precautionary behavior performance and risk perception levels (odds ratio = 4.14). Conclusion: Most medical students showed a high degree of COVID-19-related awareness, precautionary behavior, and perception of risk. Recommendations: There were students who did not practice or that showed a low risk perception as well as reported hindering factors for not practicing precautionary behaviors. Therefore, intensification of health education sessions to support adherence to precautionary measures is still required.


INTRODUCTION
he main goal of public health as well as infection prevention and control is the prevention of infectious disease transmission in population and healthcare settings. (1) Significant global attention has recently been drawn to emerging infectious diseases outbreaks caused by respiratory viruses, specifically, severe acute respiratory syndrome (SARS), Middle East respiratory syndrome (MERS), pandemic influenza A and H1N1, and, finally, to the latest novel coronavirus (COVID- 19) pandemic. (2) Health professionals have always been at risk for infectious diseases. Knowing that even asymptomatic individuals are likely to spread COVID-19, the risk level increases when the infected person is a healthcare worker. (3) Medical students will have direct interaction with affected individuals. In this population, a lack of appropriate relevant knowledge will lead them to overestimate the situation, increase their level of stress and anxiety, and interrupt their medical decisions. (4) The detection of causative agents, the production of infectivity, vaccinations and treatment, as well as contact tracing, isolation, and screening are all essential for controlling emerging infectious diseases. Many of these measures are partially focused on individual actions and how people at risk comply with behavioral guidelines. (5) Compliance with precautionary behavior among at-risk T Original Article populations is often the only way to avoid further spread of the disease. (6) Risk perception is critical for precautionary steps. Precautionary practices should be based on the correct knowledge and skills to facilitate and enable successful risk management of emerging infectious disease epidemics when treatment is not yet feasible. (7) Few studies on the knowledge and risky behavioral responses to infectious disease epidemics have been conducted. (4,8) Medical students are considered health messengers within the community. Their attitude makes a great difference in infection prevention. Therefore, the current study aims to evaluate the level of awareness, precautionary behavior, and risk perception relevant to COVID-19 and identify the factors motivating and hindering medical students to take preventive measures.

Study design and population
A cross-sectional survey was conducted of Egyptian medical students at Ain Shams University, Cairo, Egypt. The sample size was determined using the sample size estimation software PASS11, with 95% confidence intervals (CIs) and a 5% margin of error. After reviewing the results of a previous relevant study (8) that showed an average of 86.96% correct answers of COVID-19 knowledge and an average of 94.2% preventive behavior. Based on these data and using the least prevalence, which was the percentage of correct answers of knowledge related to COVID-19 (86.96% ± 5%), the required sample size was determined to be at least 175 medical students. (9) By the end of the data collection time schedule, the collected sample consisted of 351 medical students. Data collection tools: -A validated, organized, self-administered, anonymous online questionnaire (Google form) was used to collect data via a social media network (WhatsApp). Participants were recruited from June 1 to June 30, 2020. The questionnaire design primarily focused on the main domains of the standard questionnaire for risk perception. (10) In the event of an infectious disease outbreak, this tool offers a standard questionnaire that can be used to test awareness and risk perception. The awareness of disease etiology and dynamics, perception of disease severity, perception of disease vulnerability, perceived self-efficacy of preventive approaches, and motivating/hindering factors for the practice of preventive approaches are the main domains of the questionnaire. The questionnaire consisted of five sections as follows: -Section 1: Demographic information (e.g., age, sex and academic year). -Section 2: Level of COVID-19-related knowledge: Knowledge was evaluated using 15 items based on available information about the disease etiology and dynamics at World Health Organization (WHO) (11) and Centers for Disease Control (CDC) (12) studies and fact sheets in addition to results from other previous studies. (4) A correct response was awarded 1 point and an incorrect answer or "I don't know" was scored 0 points. The overall score was converted into a percentage. Scores of > 75%, 50% to 75%, and <50% were rated as high, moderate, and low levels of knowledge, respectively. Reliability was checked using Cronbach's α; the result was 0.68. -Section 3: Practicing preventive behavior: This section consisted of 8 items based on recommended preventive behavior by the WHO and CDC, (11,12) for example, preventive behavior during coughing, handwashing, and surface disinfection. Choices were "yes, always," "yes, sometimes," or "no" and 2 points were awarded to the participant for "yes, always," 1 point for "yes, sometimes," and 0 points for "no." The overall score ranged from 0 to 16 and was converted to a percentage. A score of 75% was designated as high and < 75% as low results. Item validation has been developed by experts. Reliability (Cronbach's α = 0.72) was checked. -Section 4: Perception of seriousness, susceptibility, and self-efficacy of COVID-19: Five items were used to assess the seriousness perception, susceptibility, and self-efficacy of COVID-19 among participants based on previous studies. (4,10) The answers were given a score using a 4-point Likert-type scale (1 = strongly disagree, 4 = strongly agree). The cumulative total score varied from 5 to 20. The total score was converted into a percentage. Scores of > 75%, 50% to 75%, and < 50% were categorized as high, moderate, and low levels of seriousness perception, respectively. Item validation was developed by experts. Reliability (Cronbach's α = 0.6) was checked. -Section 5: Motivating/hindering factors to perform preventive measures: According to a previous study, (10) 8 items were used to evaluate the motivating factors for using preventive measures, and 10 items were used to evaluate the hindering factors for not using preventive measures among the participants. The choices were "yes" or "no."

Ethical considerations:
This study was approved by the research ethics committee at the faculty of medicine, Ain Shams University (Approval number: FMASU R 31/2020).
The study conformed to the international ethics guidelines and that of Declaration of Helsinki (2013). The students' completion of the questionnaire was considered authorization to participate in the analysis. Data confidentiality has been preserved through questionnaire anonymity.

Statistical analysis
Revision, coding, and computer data entry were done, and then analysis was conducted using SPSS software package version 25 (SPSS, Inc., Chicago, IL, USA). Mean, standard deviation, and range values were used to describe the quantitative data. Numbers and percentages were used to present the qualitative data. Significance was considered at p < 0.05. Total scores for COVID-19 knowledge, precautionary behavior, and perception of seriousness, susceptibility, and self-efficacy were computed. The Kolmogorov-Smirnov test was used to evaluate the normality of the distribution of the quantitative data, which included age and the three scores. Age was normally distributed, but none of the three scores followed a normal distribution, so the median and interquartile range was used to describe the total scores. Score percentages were used among the studied participants to identify their levels of awareness, precautionary actions, and perception of severity, vulnerability, and self-efficacy relevant to COVID-19. The Spearman rank test was used to examine the correlations between scores. The independent student t test and chisquared test were used to study the association between the participants' precautionary behavior and other factors, including age, sex, academic year, presence of previous awareness about COVID-19, level of knowledge relevant to COVID-19, and level of risk perception. In addition, odds ratios and 95% confidence intervals (CIs) were calculated. Binary logistic regression was performed with the following predictor variables: age, gender, students' academic year, previous knowledge about COVID-19, COVID-19 knowledge score, and risk perception score to measure its impact on exhibiting precautionary behavior (i.e., low and high performance) among our sample.

RESULTS
Three hundred and fifty-one undergraduate medical students from all academic years participated in our study (  . The mass media was the most reported source of information (90.6%), followed by healthcare workers (33.3%), friends and relatives (32.2%), and last, academic studies (26.2%). Slightly more than 3% reported the WHO, the CDC, and others as knowledge source (Table 1). Nearly all participants (94.3%) scored a high level of knowledge regarding COVID-19. The lowest reported knowledge level was approximately two items: "The origin of COVID-19 is unclear, but it appears that wild desert life has transmitted it to humans" and "It can be diagnosed by the CBC when lymphocytosis is found" (29.9% and 50.7%, respectively) ( Table 2).
The self-reported performance of precautionary behavior by the studied group is presented in Table 3. Most of our study group (96.9%) had a high performance level of precautionary actions, while only 3.1% had a low level. The lowest precautionary activity recorded (52.4%) was "I increased the frequency of cleaning and disinfecting objects that can be easily touched with hands (i.e., door handles and surfaces)," followed by "I speak to my family and friends about COVID-19 prevention" (60.4 %), and "I cancelled or postponed friends' meetings, dining out, and sporting activities" (61.8%).
The COVID-19 risk perception level was rated as high, moderate, and low in 83.2%, 15.9%, and 0.9% of the participants, respectively. Approximately half of our participants reported strong agreement about the seriousness of COVID-19, and more than half strongly believed that social distancing and wearing facial masks are effective measures in COVID-19 prevention (Table 4).  Students' motivating/hindering factors that affected the performance of preventive measures against COVID-19 are listed in Table 5. The highest motivating factor was "I feel responsible for my health," followed by "COVID-19 can be serious" and "I don't want to transfer COVID-19 to people around me" (98.9%, 98.6%, and 98.6%, respectively). Conversely, the highest hindering factor (47.0%) was "I feel that very little knowledge about preventive measures is given," followed by "I believe that preventive measures will not be carried out by people in my environment" (43.3%).  A weak significant positive correlation between students' risk perception and their precautionary behavior regarding COVID-19 infection and a significant weak negative correlation regarding their COVID-19-related knowledge are shown in Table 6.
Regarding the factors affecting students' performance of precautionary behavior toward COVID-19, the results indicate that age, sex, academic year, and degree of COVID-19-related knowledge were not significantly associated with participants' performance of precautionary behavior, whereas participants who had previous information about COVID-19 and those with a high-level risk perception had a significant association with the performance of precautionary behavior (p < 0.01; Table 7). The logistic regression model showed that the risk perception score was the only predictor variable that had a statistically significant association with the performance of precautionary behavior by our study group (Table 8).  ( †) The χ 2 test was used for all, except for the (#) independent t test,* p < 0.05.

DISCUSSION
The COVID-19 pandemic is threatening the general population as well as healthcare workers all over the world. COVID-19 knowledge remains limited to its status as a novel virus. Vaccination and effective antiviral therapies are currently under trial. Until now, we could only prevent human-to-human transmission by implementing infection control measures. (3) Medical students showed a high level of knowledge. The majority of the participants (96.9%) had a good performance level of precautionary behavior. Over 85% had a high level of risk perception. The two most important hindering factors were not feeling that enough information was provided on the pandemic and not trusting that people would perform preventive measures. The participants' performance was affected significantly by their degree of risk perception. Regarding the sources of COVID-19 infection knowledge, only 26.2% and 3.4% of the respondents reported academic and other educational sources, respectively. The mass media, including television, the radio, and the Internet, were the highest reported sources of knowledge. This agrees with the study by Olaimat et al. in Jordan among university students. Like our results, only 24.2% of the subjects cited scientific sources of knowledge, whereas the highest reported sources were the Internet and the mass media. (13) The present study shows that the majority of students (94.3%) scored a high knowledge level, which disagrees with the study by Olaimat et al., who found that 56.5% scored a high level when their knowledge was tested. This difference may be due to different study populations; their study included students from all university faculties whereas ours included only medical students. Medical students would know academic basics that apply to all viruses, including COVID-19. Our results are aligned with those of Baseer et al. (14) in Saudi Arabia, who found a high level of MERS-related knowledge among dentists during the 2016 coronavirus outbreak.
Regarding knowledge questions, two items on which participants scored the least are worth mentioning: "The origin of COVID-19 is not clear, but it seems that it has been transmitted to humans by wild desert life" (29.9%) and "It can be diagnosed by the CBC when lymphocytosis is found" (50.7%). This ambiguity necessitates the inclusion of COVID-19 study in medical school curricula.
The highest scores were for questions related to the mode of transmission and virus fatality: "COVID infection spreads by respiratory droplets, such as cough and sneeze," "Infection with COVID-19 is transmitted via close contact with a confirmed case (particularly in families, crowded places and health centers)," and "For certain individuals, COVID-19 can be fatal, as older age group and people with current health problems are more at risk of severe complications." This agrees with the findings of Taghrir et al. (4) as both study populations were of medical students.
High performance of preventive behavior was found in the majority of the study participants (96.9%). This result is similar to those of Bashirian et al. (15) in Iran, who found desirable behavior in 73.1% of the hospital staff evaluated. Three items in our results were alarmingly low: "I cancelled or postponed friends' meetings, dining out, and sporting activities" and I increased the frequency of cleaning and disinfecting objects that can be easily touched with hands (i.e., door handles and surfaces)," in 61.8% and 51.4% of respondents, respectively. These results disagreed with those of Olapegba et al. (16) in Nigeria, who explored similar items but found more strict preventive behavior among their study population such as "I changed my way of living because of coronavirus" in 76.8% and "I'd like to wash my hands or sanitize them after handshaking" in 81%. This discrepancy may be because they performed a country-wide survey with a different sociodemographic distribution (e.g., a wider age range). The third item in our results, "I speak to my family and friends about COVID-19 prevention" was found in only 60.4% of the participants, which is considered unsatisfactory as medical students should consider health education regarding a current pandemic in their surrounding community.
High risk perception was found in 86.1% of the participants, which is similar to the results found by Dryhurst et al., (17) who analyzed data about risk perception regarding COVID-19 throughout 10 European countries. It was found that risk perception varied among the 10 studied countries (4.78 and 5.45) on a 7-point scale, which was fairly high according to their evaluation. This result also agrees with that of Bashirian et al., (15) who found a desirable level of perceived threat of infection among healthcare workers.
A significant positive correlation between students' risk perception and their precautionary behavior was detected. This significant relationship was confirmed using binary logistic regression. This result agrees with that of Iorfa et al., (18) who found that risk perception significantly predicted precautionary behavior. Conversely, a significant but weak negative correlation was found between knowledge and precautionary behavior. Although this result may seem paradoxical; it was previously found in more than one study. (8,19,20) This can be explained by the fact that the more technical information people know, the more secure they feel, which reduces their precautionary behavior. (19) This also implies that despite their good knowledge level, risk perception is the most important variable that could be the basis for reliable precautionary behavior, and it could possibly be a means of preventing newly coming waves of the COVID-19 pandemic.
Regarding motivating and hindering factors, the majority of the study participants agreed that there is a lack of trust in other people efficiently performing preventive measures. These worries of medical students in our research may be explained by the high percentage of health illiteracy among the Egyptian population as shown by Almaleh et al. (21) Therefore, the study participants exhibited a lack of trust in other people.

CONCLUSION AND RECOMMENDATIONS
The study participants exhibited a high knowledge level. The majority of the participants (96.9%) had a high performance level of precautionary behavior. More than 85% had a high level of risk perception. The two most important hindering factors were not feeling that enough information was provided about the pandemic and not trusting that people would perform preventive measures. The participants' performance was significantly correlated with their degree of risk perception. Including COVID-19 education in medical curricula of microbiology, public health, and internal medicine is therefore highly recommended. We also recommend massive public information campaigns to prevent or at least mitigate other potential waves of the pandemic. Adopting preventive measures on a university campus encourages the students to adopt preventive behavior.

Limitations of the study
The study was conducted with only volunteer participants; namely, medical students attending Ain Shams University who filled out the online questionnaire. Thus, there is no generalizability of the results for all medical students at Ain Shams University or all medical students at other universities. Therefore, further studies should be conducted involving medical students from different universities.