Knowledge and perception of COVID-19 pandemic. Before-and-after online community survey.

Background COVID-19 pandemic impacts many communities worldwide. In this study the Poles' knowledge about COVID-19 as well as people’s behaviours, attitudes and fears during the pandemic were assessed. Changes in these between the outset of the pandemic and the imposition of the strictest lockdown measures in Poland were investigated. Methods Physicians, nurses, students of medicine-oriented faculties, non-medical professionals, students of non-medicine-oriented faculties and secondary school students were surveyed by an anonymous online questionnaire two times: at the onset of the pandemic and in the second week of the strictest lockdown. Statistical analyses were performed using non-parametric tests – Pearson Chi Square, Kruskal-Wallis tests. Results In total 2618 responses were collected. At the beginning people knew that the respiratory system was attacked (97.9%); correctly identied the major symptoms of COVID-19 (95.0%) and ways to prevent infection: hand washing (99.8%), covering mouth (85.9%) and the need to call sanitary-epidemiological services if one experienced COVID-19-like symptoms (92.1%). The management Respondents reported higher level of fear of the pandemic than other groups included in this study. This study uses before-and-after approach which highlights the changes in people’s knowledge and perception of the COVID-19 pandemic during the pandemic’s progression.


Background
Coronavirus disease  is the disease caused by the novel coronavirus 2019-nCoV (SARS-CoV-2). It is currently one of the biggest acute global threats to human health. The outbreak of the pandemic has affected many countries, causing problems concerning almost every aspect of everyday life. It has led to realignment in the world economy, deepened the medical crisis in many already vulnerable countries and it is currently the predominant topic covered by all media. The pandemic poses an unprecedented challenge for all health care systems worldwide. This is the rst medical crisis in the recent history affecting different populations, countries and continents, requiring everybody to be aware and prepared for it.
The new coronavirus strain, discovered in December 2019 in the city of Wuhan in China [1], spread globally within few months. The rst cases of COVID-19 in Europe were observed in January 2020 [2]. On March 4 th , when the rst case of COVID-19 was reported in Poland, there had already been more than 93,000 con rmed cases of infection and almost 3,200 deaths reported globally [3]. As it was planned to distribute the questionnaires when the rst case of  in Poland will be observed, the rst wave of the survey was launched on that day. The questions and answers of the survey were based on the current knowledge and prevailing atmosphere at that time, when many facts about the disease have not been discovered yet.
Within few days, the Polish government decided to implement precautions to stop the pandemic from spreading in Poland. Schools and universities were closed on March 12 th [4]. Various public facilities and shops were closed on March 13 th for a period of two weeks or longer. People were instructed to stay at home and refrain from travelling outside the country and the "state of epidemic threat" was declared nationally [5]. After a week, this o cial status was changed to the "state of epidemic" [6]. On April 1 st , further restrictions amounting to a lockdown were introduced: leaving home was prohibited unless necessary, and only allowed individually or in groups of two. As a consequence of refusal to comply with the regulations a ne could be imposed [7].
The second wave of the survey began on April 9 th , when there were about 5,600 con rmed cases of infection and 174 deaths in Poland [8]. Globally, there were over 1,430,000 con rmed cases of infection and more than 85,500 deaths [9]. At the time the second phase of the study began, after nearly a month of lockdown, the Polish government had not abolished any of the restrictions yet.
The rst aim of the study is to evaluate Poles' knowledge about COVID-19 as well as people's behaviours, attitudes and fears during the pandemic in Poland, depending on the socio-educational pro le of the respondent. The second aim is to compare responses at the beginning of the epidemic before the lockdown restrictions and during its peak, at the time when the number of COVID-19 cases in Poland was very high and the restrictions imposed by the government had already affected life of the respondents.
To assess the level of knowledge and attitudes regarding SARS-CoV-2 and COVID-19 a short survey was created to evaluate the knowledge about coronavirus in the Polish population as a spontaneous reaction to the situation. The survey also covered social aspects of the pandemic, its in uence on everyday life and peoples' attitude to information disseminated in the media.

Materials
The survey was based on 2,618 responses to the anonymous questionnaires, which were collected in two waves: at the outset of the epidemic and during the most severe COVID-19 lockdown in Poland. Demographic characteristics of the sample are presented in Table 1. Due to the pandemic conditions, it was decided to collect data online. The following cohorts were distinguished among the respondents: medicine-oriented community: physicians, nurses, medicine-oriented students, and non-medicine-oriented community: non-medical professionals, non-medicine-oriented students, secondary school students (over 18 years old). Such division was introduced, as physicians, nurses, and non-medical professionals are mentioned in other studies on COVID-19 perception [10][11][12][13][14]. Moreover, researches did not nd many studies where the secondary or university students were asked about their knowledge and perception of the pandemic. These cohorts are also easily accessible for an online survey. The questionnaires were send via e-mails and posted on Facebook groups for medical professionals, students and non-medicine-oriented communities. Each group consisted of at least several thousand people, which aimed to reduce the selection bias. Respondents were surveyed to evaluate their knowledge regarding coronavirus and awareness of the threats of the epidemic outbreak in Poland. Although every resident of Poland was eligible for this survey, the biggest number of the respondents hailed from the northern regions of the country.

Methods
The questionnaire was designed by the students of the Medical University of Gdańsk based on the WHO recommendations and information. Answers to the knowledge questions were based on the o cial WHO guidelines [15]. The survey consisted of twelve questions divided into four sections: demographics, knowledge, attitudes, and behaviours. The rst section gathered data on the occupation, age, and the current place of residence of the respondent. The second part evaluated general knowledge about COVID-19 including its symptoms, routes of infection, and prevention methods. The third part measured the fear of the pandemic and which lockdown restrictions were the most di cult for people to cope with. The fourth part focused on sources of information on COVID-19. In the 4 th question (infected body systems) a respondent could mark only one answer while the questions 5 to 11 (symptoms, transmission routes, prevention, what to do when experiencing symptoms, daily life di culties, fears, knowledge sources) were multiple choice questions in which a respondent could pick more than one answer. The 12 th question was a rating scale question. The research was conducted at two different phases of the epidemic in Poland.

Data analysis
Demographic data are presented in Table 1 as counts and (for age data) means with standard deviations. The percentage rates of answers were calculated for questions 4 to 11 for each of the six cohorts and for the whole sample ( Table 2, Table 4, Table 5). For question 12 (how well-informed respondents felt), mean scores (on a 1-5 scale) were calculated (Table 3). All calculations were performed separately for both waves of the survey. All analyses of statistical signi cance were performed using non-parametric tests -Pearson Chi Square (questions 4-11) and Kruskal-Wallis tests (question 12). Threshold of statistical signi cance was set at p<0.05. Statistical analysis was performed in Statsoft Statistica 13.3 software.

Results
In total, 2,618 surveys were collected, with about 80% of respondents residing in the Pomorskie (Pomeranian) voivodeship, Poland. Age and profession characteristics of the sample are presented in Table 1.
When comparing the results from the rst and second wave of the survey, more people reported wearing facial masks and staying at home (+37.5%, +66.1%, Table 4, Table 5). Respondents were also more aware that they should not visit a general practitioner or an emergency department (1.6%, -6.3%) if symptomatic; that the SARS-CoV-2 could spread via direct (+18.1%) or indirect contact (+9.2%) and that tiredness was a symptom of COVID-19 (+9.1%). People were more worried about the pandemic generally (+19.6%), economic (+64.1%) and education (+37.3%) crises, and about their families (86.8%, +26.5%). However, they were less afraid of being quarantined (-18.2%) or price increases (-36.7%). Among nurses and physicians not a single respondent selected "not worried about anything" answer, while only these two populations were not concerned about being absent at work (Table 4). More respondents in the second phase of the survey reported visiting governmental and science-oriented websites (+27.4%, +9.6%, respectively), especially among the non-medical section of the sample, and watching TV (+16.4%) to learn about COVID-19. Nonmedical professionals and nurses were the only cohorts that felt slightly worse informed about the COVID-19 in the second wave of the survey. In other groups, information con dence increased (Table 5).

Discussion
The COVID-19 pandemic is a rapidly spreading global threat that in one way or another affects everyone, regardless of their profession and age. The study reported in this paper was a spontaneous and quick response to a unique and new social situation, designed to gain a measure of understanding of Poland's population knowledge of and reactions at two very different phases of the epidemic's development in Poland. Due to the crucial role of medical professionals in tackling the pandemic, the sample was divided into the two main cohorts, medical and non-medical.
Communities' knowledge about COVID-19 Even at the beginning of pandemic in Poland, nearly all of the respondents knew that the SARS-CoV-2 attacked the pulmonary system. Moreover, this level of knowledge was maintained a month later.
The three main symptoms of the infection (cough, fever, shortness of breath) were well known. This is similar to some studies [16][17][18], but different to other studies [19,20]. Only the three main symptoms of the infection were shown on the governmental posters [21] and that is probably the reason why the non-medical respondents were mostly aware just of those. At the beginning of the epidemic, muscle pain and tiredness were mentioned only by specialised sources of scienti c knowledge [1,22] and that is probably the reason why only physicians knew about them. After a month, an overall increase in the awareness of other symptoms was observed.
In March 2020, Poles knew the COVID-19 was an airborne disease, similar to the UK and US populations [16]. After a month, they also knew that the virus could spread via direct and indirect contact. Researches marked answers pointing to these three ways of transmission as correct, as the preventive measures promoted by the government [21] to limit the spread of COVID-19 were: hand washing that limits indirect contact, covering mouth to reduce airborne transmission and social distancing, which reduces direct contact. People were highly aware of these prevention steps.
At the beginning of the epidemic, Poles knew they should call sanitary-epidemiological services or visit an infectious diseases ward if experiencing COVID-19 symptoms. Unfortunately, some people would also visit a general practitioner clinic or an emergency department. This would be a fundamental mistake as these places were not taking su cient anti-COVID-19 preventive measures [23]. The same behaviour was noticed in South Korea during the MERS epidemics [24]. This is most likely due to the typical way the healthcare systems operate, including the Polish one. If one needs to see a physician, one goes to a general practitioner (mild cases) or an emergency department (severe cases). This changed during the pandemic as the healthcare system had to adapt to the new conditions. Fortunately, during the second assessment, a threefold and fourfold drop was observed in the percentage of respondents mentioning a general practitioner or emergency department visit with COVID-19 symptoms, respectively. This is an excellent improvement in the society's knowledge that limits the COVID-19 spread. This happened as the "do not's" were added on posters, too [25].
Why did the Poles have such high levels of knowledge about COVID-19 that further increased later on? At the beginning, the most crucial information (transmission routes, prevention, what to do if one has symptoms) was broadly disseminated via lea ets, posters [21,25], and mass media news reports, and repeated multiple times so people had a chance to encounter and learn it. A similar pattern was observed in other countries affected by the pandemic [26]. What is more, before the virus arrived in Poland in March, the pandemic had already been ongoing for several weeks (months, if counting from the rst Wuhan infections) and had already affected many other countries worldwide (China, South Korea) and in Europe (Italy) [3]. Thus, people in Poland had more time to prepare for the new threat. Searching for information using various media could be used to reduce anxiety levels by helping people to understand the upcoming, unprecedented situation [12,26,27]. On the practical level, increasing one's knowledge helps to avoid infection [16].
The differences between medical and non-medical populations are understandable in the light of the requirement for continuous professional education among medical professionals. Furthermore, medical professionals form the rst line of defence against COVID-19. Thus, this group had naturally more interest and a larger need to prepare early and more effectively to protect themselves.

COVID-19 information sources and their evaluation
In general, people felt quite well informed to start with, and after a month the information con dence had slightly increased further. This is a result similar to other studies [18,19,28]. General news websites and social media were the main sources of knowledge for non-medicine-oriented respondents, which explains why at the beginning of the pandemic many people were not familiar with the up to date events and ndings presented in scienti c knowledge sources. Conversely, medicine-oriented community preferred science-oriented or governmental websites. Again, this group choose these sources to educate themselves as they were the rst line of defence against COVID-19.
In the second wave of the study the respondents indicated using other sources of knowledge than they had used a month earlier. More respondents visited governmental and science-oriented websites. This difference was particularly noticeable among the non-medical cohort. Additionally, due to the lockdown restrictions and isolation [29], people were spending much more time at home. Thus, more people had time to watch television and learn about COVID-19 from TV broadcasts, which explains the greater popularity of this information source in the second round of the study.

Pandemic-induced fears and their changes
At the beginning of the pandemic in Poland the virus had already spread to all inhabited continents and clearly affected societies worldwide. In general, insecurity was the most alarming factor for the Poles, who sometimes felt more threatened with COVID-19 compared to, for example, people in China [26]. The probable reason was that at the beginning of the epidemic in China, the residents were not aware of the gravity of the situation and the consequences that the pandemic would bring. Moreover, media were devoting a lot of coverage to the epidemic in China [30]. All these induced additional fears and psychological impact on the Poles and other nations [31].
A month after the rst survey was conducted, the situation in Poland had changed dramatically. Numerous governmental restrictions were imposed on the citizens in order to stop COVID-19 from spreading [4][5][6]. As a result, more than 95% of the respondents surveyed in the second wave of the study were scared of something (picked at least one item in the 10 th question) and a 20% increase in this measure was observed between the rst and the second wave of the study (Table 4, Table 5).

The quarantine and isolation
As quarantine and isolation cause a separation from the family, loss of freedom, and boredom, it is understandable [27] that people would like to avoid it and that at the beginning of the epidemic they were scared of it. However, the quarantine and isolation is crucial in preventing the pandemic from spreading [1]. Due to their concerns quarantine and isolation some people might not comply with the lockdown rules, which could result in higher incidence rate in the population. After a month, the quarantine became widespread and many people themselves or their family members had to undergo quarantine. This familiarity may be the reason why, after a month of the lockdown, the respondents were less afraid of being quarantined.

Worry about family members
At the beginning many people were worried about the health of their family, similarly to the results noted in other studies [28,32,33]. Such concerns increased as the government imposed physical distancing, quarantine and isolation, especially for the elderly who have the greatest mortality risk [29]. Over time, the number of cases of COVID-19 in Poland increased [8]. Additionally, many healthcare systems could not manage COVID-19 without being overwhelmed (lack of hospital beds and medical personnel) which resulted in a high death toll, for example in Italy [9].
These may be among possible reasons why concerns about the health of respondents' families considerably increased and were the most frequently selected reason for worry in the second wave of the survey.

Financial crisis
It was expected that the prices for products in shops would rise or be instable, as an effect of decreased and prolonged delivery services, access, utilization and stability [34,35]. Such thinking caused many people to stockpile for the upcoming pandemic [36]. Moreover, a global economic recession is projected [37]. The fear of job loss is experienced by 26% of the Poles [38]. As a result, the threat of an economic crisis registered the greatest increase among all reasons for worry listed in the survey.

Education crisis
Relatively low fear rate at the beginning of pandemic in three surveyed "student" cohorts was observed. With introduction of the lockdown restrictions, students could no longer attend their educational facilities. With time, students started to worry about their future and their fear rates increased in the wave two of the study. Similar trends are observed globally [14,39,40]. Students of medicine-oriented faculties reported this worry most frequently, likely because their courses included practical classes like laboratory classes or clinical workshops that cannot be taught online.

Fears among medical professionals
Being the rst line of defence against COVID-19 for over a month, there was no respondent among nurses and physicians who did not pick at least one reason for worry. This shows that as the pandemic spread, medical professionals felt more and more endangered, possibly due to rising stress levels and tiredness. Moreover, in the second wave of the study medical professionals were the only group who did not indicate their absence at work as a worrisome problem. There is a likelihood that they would rather not go to work if they had any choice. This interpretation is consistent with the results from other studies, where physicians reported reluctance to work and even considered resignation [41] or showed poorer mental health than the population's average [18]. Moreover, doctors experience high levels of posttraumatic stress [42], which suggests a need to pay more attention to psychological problems of the medical professionals, especially as members of this population suffer from vicarious traumatization [12] and seek increased social support during the pandemic [43].
The signi cant advantage of this survey is the fact that the rst phase of the survey was conducted at the onset of the pandemic in Poland [3]. This allowed for monitoring of the change in level of knowledge and attitudes to COVID-19 among the Poles by repeating the same survey after a month. Furthermore, an online survey is an optimal method for collecting data during this dynamically changing situation, since it is faster and requires no contact between people, thus being more convenient than traditional methods.
Due to the shortness of time available for creating an ad hoc spontaneous survey, the interpretation of the ndings from the study was subject to methodological limitations. It should be mentioned that some facts about the virus have not yet been scienti cally proven and they change over time. Moreover, the samples in wave one and wave two of the study were not matched, and their demographic and professional composition differed to some extent, as the researchers did not conduct a follow-up survey and, as a result, anyone could ll in the survey in either or both waves -thus it is possible that some of the responses in wave two were from respondents who completed wave one, and some responses were from new respondents. Regardless of this methodology limitation, the samples were treated as independent samples coming from the same population. To increase the likelihood of obtaining a comparable sample, the survey was distributed by the same information channels in both waves. It is also problematic to assess whether respondents looked up answers to any of the knowledge questions while completing an online questionnaire.
In addition to that, the mean age of the sample was young, so in order to get wider and stronger results the survey should have included the whole adult population.

Conclusions
Due to their good initial knowledge about symptoms, transmission, and preventive behaviours regarding COVID-19, the Polish population appears to have been prepared for the peak of pandemic's cases. People had more short-term concerns, like the possibility of rising grocery prices or worries about coping with quarantine and isolation. Over time, the knowledge and the fears among the respondents changed and at the end of the study they re ected long-term pandemic management issues. After a month, more people reported being scared of the pandemic in general, as well as economic and healthcare crises. The quarantine was not perceived as being as stressful as often as in the rst wave. Students concerns were related to the education crisis. Medical professionals reported higher level of fear of the pandemic than other groups included in the study. Over time, a transition from the use of popular mass-media to more scienti c sources of information was observed among the non-medical sub-population.
Compared to other studies, this study used before-and-after approach, which highlights the changes in people's

Declarations
Ethics approval and consent to participate All participants had to read a written consent form for purposes of this study before ful lment of the survey. Ful lment of the survey by the participant was an equivalent to his or her consent to take part in this study. The Bioethics Committee of the Medical University of Gdańsk had given its approval to carry out this study. Bioethics Committee of the Medical University of Gdańsk approval number: NK BBN/157/2020.

Not applicable
Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Not applicable
Authors' contributions WN and KN created the survey and wrote abstract, results, discussion and conclusions. JL wrote the introduction section and helped in survey development. AP wrote the methods and material section, helped in survey development, and applied for approval of the Bioethics Committee of the Medical University of Gdańsk. AS took part in development of the study concept, study survey and substantively revised the manuscript. MB analysed the data statistically and reviewed the manuscript. All authors read and approved the nal manuscript.