1 Introduction

According to Earl Babbie, an American sociologist and author of many books in the field of social research, some paradigms appear obvious to the public and almost no one tries to discuss the roots of such perception (1). People tend to take things for granted but most of what we know is just an assumption or the result of faith. By analogy, the positive climate around phage treatment and its rapid global development has provoked a favorable perception of such a form of treatment among those who have or might have contact with it, including lay people. The potential of such an encouraging campaign for phage therapy was discussed by Hausler in a suggestively entitled article Phages Make for Jolly Good Stories (2). The number of appealing statements about phage therapy coming from the professionals, who tend to present it mainly in a positive light makes us think that people's perception of such novel therapy should be the least of our concern (3–8). However, has anyone ever tried to investigate this at a deep level? Recent events related to the COVID-19 pandemic have shown a quite surprising picture, in which even life-threating situations have not prevented the anti-vaccine movement from entering the mainstream. The problem has reached such proportions that even the World Health Organization labeled vaccine hesitancy as a major public health threat (9). By analyzing the history of the anti-vaccine movement, one might draw the conclusion that the current situation is the exact opposite of what most of us thought: the greater the incentives for vaccination from health care authorities, the greater the resistance from the public (10, 11). The complexity of social behavior in terms of human health was highlighted in 2020 by the U.S. Department of Health and Human Services by admitting that health and health-related behaviors are determined by influences at multiple levels, including personal (biological, psychological), organizational/institutional, environmental (both social and physical) and political levels (12). Similar conclusions were drawn by Charani et al. who saw major challenges in building health awareness in sociocultural disparities such as, among others, race, ethnicity, gender identity and cultural as well as religious practices or even migration status and geographical influence on health (13). For instance, it has been observed that smoking among adolescents is inevitably associated with similar habits in peers from the same environment, but this does not necessarily extend to quitting smoking (12). Such dependence could be a possible explanation for the above-mentioned vaccine hesitancy in particular social groups, for instance, ethnic minority groups (14).

Regardless of the above, phage therapy presents a rather unusual case in the context of public health. First, its public perception is still relatively unknown. Although phage therapy is becoming more and more popular, one cannot say that patients are persuaded to use this form of treatment as was the case on a global scale for vaccinations during the COVID-19 pandemic. Mostly, patients with infections caused by antibiotic-resistant (ABR) bacteria seek help on their own. Such an approach, where patients seek phage treatment by themselves, was recently emphasized by other authors (15). This is certainly a common situation in the Phage Therapy Unit (PTU) established at the Hirszfeld Institute of Immunology and Experimental Therapy (HIIET) in Wrocław, Poland (16). Moreover, following screening procedures, patients often do not qualify for such treatment. Over the 18 years of operation of PTU, less than half of the patients (nearly 823) qualified for phage treatment out of 2,286 patients registered at the PTU (as of December 2023) (16). In Belgium, between 1 January 2008 and 30 April 2022, 1,066 phage therapy requests were submitted to the Queen Astrid Military Hospital resulting in 100 patients treated with phages (17). Another determinant that makes the phage therapy landscape different from the COVID-19 pandemic, particularly in Poland, is the lack of successful clinical trials (due to pharma industry's hesitancy) and the lack of non-commercial clinical trials. Putting together the first non-commercial clinical trial in Poland was commenced at PTU in 2022 thanks to the funding from the Polish Medical Research Agency. It is worth mentioning that until a few years ago non-commercial clinical trials were almost non-existent in Poland (18). The main difference that could possibly influence the public's perception of phage treatment, contrary to the COVID-19 example, is the authorization process used for vaccines against SARS-CoV-2. Such a process is required by health authorities before any new drug is released to the market. We have all witnessed a historically rapid approval process for emergency use on an unprecedented scale, especially in the case of two mRNA-based COVID-19 vaccines (19). Such haste aroused distrust from a large part of the society. The following occurrence of adverse events, so-called adverse event following immunization (AEFI), which led to the suspension of certain vaccines in some countries, unequivocally increased public concern and perception of such treatment to an even greater extent (20, 21). Contrary to the rapid approval of new COVID-19 vaccines, phage therapy until recently encountered difficulties at a basic level (for instance, how to classify the natural bacteriophages in terms of medicinal products) even though antibiotic resistance is a major twenty-first century health challenge (22–26). In addition, adverse events following the administration of phage preparations are rare, mostly mild and, as mentioned above, phage therapy has never been the subject of acute safety concerns, including those discussed in the mainstream (16, 22, 27).

From the above description, a picture emerges suggesting that extensively focusing on phage therapy outcomes and legislative obstacles almost completely minimized the need for social research conducted in groups of people differing from each other, among others, by gender, age, education, employment and material status or experience with health care they received. There are only a few articles that have investigated people's opinions on phage therapy with a relatively limited sample size (28, 29). The need to investigate public understanding of phage therapy and research was recently emphasized by Hitchcock et al. (30). The authors believe that successful commencement and development of phage treatment in a clinical setting must be combined with listening to people about their concerns, experiences and expectations. Followed by a similar belief, we investigated how this situation looks in Poland, a country that has the largest in the EU database of patients who have undergone phage therapy (over 750 cases) in the form of compassionate treatment in accordance with current ethical and legal requirements (31).

There was an expectation that phage therapy and phage research awareness remain at the low level in Poland. Such assumptions were based on our long-term observation of the number of mainstream publications dealing with phage therapy and the time devoted to this topic in the Polish media. PTU, being the only phage therapy center in Poland and one of only a few across the globe, is not able to attract such attention as major oncology or cardiology centers to which all involved parties, i.e., the pharmaceutical industry, health care institutions, regulatory agencies, media and patients devote a lot of attention. In addition, recent investigation by McCammon et al. conducted with nearly 800 respondents clearly showed that phage therapy remains poorly understood by the public in the UK (32). The aim of the article was to present the general level of knowledge and perception of phage therapy and research in Polish society. To our knowledge this paper is the first analysis to be conducted on such a large group of people (over 1,000 individuals) regarding perception and understanding of phage research and therapy in humans.

2 Methodology 2.1 Overview

Surveys in health care must adhere to the same rules as any other typical questionnaire and the authors face similar problems that may affect the results, e.g., responders may wish to manipulate the answers, questions might be too subjective or the sample size too low (33). Since we encountered all these issues while working on our survey, it was important to spread the survey over time in order to maximize the potential of obtaining the most diverse and large group of respondents who understand the questions being asked. Due to the complex structure of our questionnaire consisting of up to 70 questions (Q), including conditional ones that were available to respondents depending on their previous answers, the number of social groups addressed in it and limited financial support, we decided to exploit a web survey. Such an approach is not only cheaper and faster than standard solutions (like mail-based polls) but also their popularity has consistently grown in recent years (34). We were aware that such an approach was not perfect, however, as it required access to the Internet, which to some extent excluded the less affluent as well as the older adults and disabled groups. Moreover, a less personalized approach typical in a web-based environment carried the risk of a reduced response from potential respondents, as confirmed by other authors (35, 36). Finally, given the scientific nature of the survey, there was a risk of general misunderstanding of the topic and terminology used across questions which could lead to incorrect answers, or the questionnaire not being completed (37). Nonetheless, a web survey was chosen as a methodologically acceptable way of conducting this kind of research.

2.2 Survey structure

Our questionnaire was prepared in two identical versions, separately for women and men, which was caused solely by grammatical complexities in Polish. The decision on which version of the questionnaire to use was made by the respondent at the very beginning, without the possibility of changing the choice later. A short description, invitation and a “thank you” message for the willingness to participate in the survey was placed on a dedicated website using HIIET servers. The collection of questionnaires started on 13 June 2022 and lasted until 18 April 2023 (data cut-off), i.e., for 10 months and 5 days. A link to the above-mentioned website (hirszfeld.pl/ankieta) was the master link sent to potential participants. The link led to a survey located on the servers of the Institute of Philosophy and Sociology, Polish Academy of Sciences in Warsaw, Poland. The questionnaire was completed anonymously and voluntarily. The information gathered did not put the participants at risk in any form. Further, they were able to end the questionnaire at any time, although this resulted in the loss of previously provided answers and the need to start the process from the beginning (if such a decision was made). Of note, if the questionnaire was not fully completed, the system did not consider it valid.

An original, structured, cross-sectional, self-administered questionnaire was designed with the CAWI (Computer Assisted Web Interview) technique, i.e., a questionnaire in which the respondent is asked to complete a questionnaire in electronic form. Among 70 single or multiple-choice questions, 36 were addressed to all respondents (first module), 11 were intended for health care professionals (second module) and 10 for people working in the scientific and research sector (third module). Persons declaring work in both above-mentioned sectors had to take part in both modules. It should be added that depending on the answers given, access to some subsequent questions was restricted. The last part, optional for each survey participant, was based on completing a quiz consisting of 12 questions about phage history, biology and phages' therapeutic connotations (fourth module). Although answering questions in the quiz required a bit more knowledge than the average, we had in mind when creating these questions that the level of difficulty should be adapted to lay people who may have never heard of phage therapy. Therefore, each question in the quiz offered the option to provide the answer I don't know in order to avoid randomly selected answers. Our goal was to examine the real level of understanding in the subject of phage therapy. We did not want to cause respondents to feel embarrassed or ashamed due to a lack of sufficient knowledge. The possibility of choosing the answer I don't know was intended to legitimize the lack of any proficiency in the examined matter. In addition, there was a separate question in which we asked for consent to participate in quiz. Thus, an active consent was required. At this point it should be added that the answers from the first three modules were still valid even if a respondent chose not to take the quiz (fourth module), which was communicated to each person before making the decision. The entire questionnaire with a full list of questions and answer options is provided in Appendix 1. Because English grammar is characterized by gender-neutral second person pronouns, there was no need to provide both versions of the questionnaire (for women and men) that was used in our research.

2.3 Recruitment of potential respondents

Seeking potential participants was probably the most challenging task we had to deal with. In the following weeks we started creating mailing lists based on publicly available contact details asking people, mostly professionals from health care and scientific sector, to fill out questionnaires. Such a request was accompanied by a brief explanation of the purpose of the survey. Each time (except for patient advocacy groups in which case e-mails were sent directly by the authors) the official email request was signed by the HIIET's authorities. A full list of entities we have contacted across Poland is enclosed in the Appendix 2. Shortly we reached 11 patients' associations, 67 institutes of the Polish Academy of Sciences (i.e., all existing institutes in Poland), 43 health care entities as well as 62 private and public universities. In addition, we spread the message across the official website of HIIET PAS, BINWIT social media channels (Facebook, ResearchGate, Twitter), we next reached out to our relatives, friends, colleagues and fellow workers. We were fully aware that such a strategy presented a lack of appropriate representativeness of the group hence this period was considered the first part of questionnaire collection. The increase in the number of completed questionnaires was tracked at irregular weekly, bi-weekly or monthly intervals (separately for male and female version of the survey) with the first follow-up on 22 September 2022, i.e., three and a half months after the survey was launched. A summary that gives a general idea of the number of responses we received after each mail distribution is shown in Supplementary Table 1.

After reaching half of the target number of respondents (~500) and an initial analysis of the results suggesting an expected lack of sufficient diversity we initiated cooperation with an experienced Warsaw-based research agency SW Research (swresearch.pl) whose clients include the world's largest brands. SW Research has completed over 3,600 research projects and is a signatory to the Interviewers' Work Quality Control Program as well as a member of the European Society for Opinion and Marketing Research (ESOMAR). The company's task was to provide access to its online panel of respondents (about 400 people, half women and men), which was assumed to be a group less associated with professionals and more representative of Polish society than the panel of people (more or less familiarized with the problem of antibiotic resistance and phage therapy) we reached on our own, hereinafter called Arm A. This group of interviewees received a form of compensation for completing the questionnaire directly from and on the terms of SW Research. Respondents from the SW Research database (hereinafter called Arm B) completed our survey between 10–14 March and 7–11 April 2023. In order to eliminate unreliable respondents from the analysis, questionnaires completed by respondents within the time frame belonging to the first decile were excluded. It ranged from 1 min 11 s to 5 min and 19 s. In the authors' opinion, this time was insufficient to provide thoughtful answers to all questions, even in the shortest version of the questionnaire.

2.4 Analysis design

The categorical variables were reported as frequencies and/or percentages and compared by sex, age, place of residence, level of education, employment status, experience with health care, profession and financial status. Both aforementioned datasets, Arm A and the Arm B, were analyzed jointly except for the key questions regarding knowledge and perception of phage therapy and research (due to expected significant differences among these two groups). The core design of analysis is presented in Supplementary Table 2.

2.5 Statistics

The data collected from both questionnaires (female and male versions) were merged into one file and analyzed using STATISTICA v. 13.3 (TIBCO Sotfware Inc., Palo Alto, CA, USA). Some calculations were made using Excel (Microsoft), version 16.72 for MacOS as weel as GNU PSPP Statistical Analysis Software (version 1.6.2 for MacOs; available at https://www.gnu.org/software/pspp/). GNU PSPP is an open-source software for analysis, intended as a free alternative for IBM SPSS Statistics while both are widely used in survey research (12, 37–40). Numbers and percentages were calculated for nominal and ordinal variables and included in contingency tables. It must be emphasized that despite our efforts the sample tested was neither random nor representative, hence all parametric tests based on the randomness of the sample must have been excluded from further analysis. Pearson's chi-square test was used to assess the homogeneity of distributions. Medians and quartile ranges were calculated for the ranked variables, and the significance of differences in the two groups was verified using the Mann-Whitney U test. In the case of three or more groups, the Kruskal-Wallis test was used. Post-hoc analysis was done using the Dunn's test. Spearman's rank correlation coefficients (rS) were calculated to assess the strength and direction of correlation between two ordinal variables. To determine whether two proportions are different from each other a two-proportion Z-test was used. In all statistical tests results were considered significant if p < 0.05.

To assess the level of knowledge on bacteriophages and phage therapy, the author's General Knowledge about Bacteriophages scale (GKB-12) was built based on the responses to Q59–70 and used after its validation. The value of the Cronbach's α reliability index was estimated.

3 Results 3.1 Characteristics of the sample size used in the study

The overall sample size for this study was 1,098 evaluable questionnaires after removing 121 cases from 1,219 collected questionnaires due to too short response time as described above. 681 questionnaires (62%) were collected as part of our efforts through mailing lists, social media, verbal requests (Arm A) while 417 questionnaires (38%) were collected from SW Research respondents (Arm B). As for the quiz, which was optional for everyone who completed the questionnaire, 1,020 individuals (92.9%) agreed to participate (614 from Arm A and 406 from Arm B).

From the beginning of the collection of the questionnaires, women filled out them over twice as often as men, regardless of the professional groups to which we distributed our requests which is a typical occurrence observed by other authors (41, 42). However, it must be emphasized that the difference between the number of women and men was even greater in Arm A (485 women vs. 196 men) while the Arm B was initially designed to collect the same number of both genders (ultimately there were 209 women and 208 men). The majority of participants lived in Poland which should not be surprising considering the language used in the survey. However, due to a significant number of Poles living and working abroad we decided to provide an additional answer option (i.e., living outside of Poland). Nearly half of the tested population (49.1%) lived in cities above 250,000 inhabitants, 65.4% whom had completed a higher education (from a bachelor's or engineer degree onwards) and the majority (66.5%) worked as full-time employees. Over three-quarters of the recorded answers reported a rather satisfactory or definitely satisfactory financial status and almost half of the participants was satisfied with the health care they received. Clearly, the surveyed population does not reflect the Polish society, and we suspect that it does not reflect the society of any country in the European Union or a country with a developed economy. For instance, according to Statistics Poland, Poland's principal government institution responsible for statistics and census data, the real number of Poles with completed higher education is twice as small (23.1% in 2021) (43). Further, it is easy to calculate that 11 Polish cities with more than 250,000 inhabitants account for <20% of the population (compared to 49.1% in our study). The main characteristics of the examined population is shown in Table 1.

Table 1. The main characteristics of the examined population (1,098 individuals).

As for the two professional groups, which were supposed to be studied separately, 104 (9.5%) interviewees described themselves as only health care professionals while 343 interviewees (31.2%) as only science and research professionals, whereas 91 (8.3%) allocated themselves to both groups. Such large proportion of scientists and individuals associated with the scientific sector is consistent with the abovementioned results regarding the level of education or place of residence (major research centers are mostly located in large cities). In fact, 72.5% of science and research professionals lived in cities above 250,000 inhabitants while only 9.7% in villages and even less (9.4%) in cities up to 50,000 inhabitants. 91 respondents worked in both sectors (health care as well as science and research) which, given the similarities between these two sectors, was an expected outcome. Of note, both questions on profession also refer to the past, hence there is a possibility that someone from the group of 91 respondents worked in both sectors in the past, not necessarily at the time the questions were answered. We certainly hoped for a greater participation in the survey of physicians, who account for only 2.6% of all respondents. Such a small number may be due to the specificity of the profession (overload of duties, lack of time) and difficulties in reaching the medical staff in hospitals (multi-departmental facilities employing hundreds of people), but also the lack of interest in phage therapy as a form of treatment with unconfirmed effectiveness in standard clinical trials, which is also not reimbursed. However, according to The World Health Organization, there are 2.4 physicians per thousand inhabitants in Poland, i.e., still 10 times less than participated in our survey (44). The participation of individuals from the science and research sector, in turn, was high (39.5%), and even greater if we consider only Arm A (61.2% vs. 4.1% in Arm B). This can be explained by scientists' interest in phage therapy as a field of science that still needs investigation and the number of people more or less directly associated with the authors. More detailed characteristics of both professional groups are presented in Tables 2, 3.

Table 2. Core characteristics of health care sector (195 individuals).

Table 3. Core characteristics of science and research sector (434 individuals).

3.2 Analysis of quiz results and assessment of general knowledge about bacteriophages and phage therapy

Ultimately, 1,020 respondents (92.9%) agreed to participate in a test assessing General Knowledge about Bacteriophages (GKB) and their use in the treatment of infections caused by ABR bacteria. We have tried as much as possible to ensure that the questions in the quiz are not directly related to the information contained in the questionnaire. In other words, the answers required the knowledge that the respondent had before completing the questionnaire. Achieving this goal was not always possible. For instance, references to bacterial infections or the experimental nature of phage therapy were inevitable. On the other hand, questions directly related to the questionnaire were a kind of test of whether the respondents answered the questions with attention and understanding. A summary of all responses is presented in Table 4. A full list of questions and a set of answers (with the correct answers underlined) is available in Appendix 1.

Table 4. Summary of basic knowledge of phage therapy and research (question 59–70) and results of the reliability analysis of the GKB-12 survey items regarding General Knowledge about Bacteriophages and their use in the treatment of infections caused by ABR bacteria.

There were only 4 questions (out of 12) with more than half correct answers (Table 4). In Q8 we checked whether the respondent understood the form of phage therapy conducted in Poland (experimental or standard of care). As many as 75.8% of participants gave the correct answer to this question (treatment conducted as a medical experiment). References to the experimental form of phage therapy were found in many places in the questionnaire, which certainly contributed to such a high percentage of correct answers. By analogy, in Q6, we asked about group of microorganisms (infectious agents) that can be treated with bacteriophages. This question generated the second highest percentage of correct answers (71.5%). Such high percentage should not be surprising as the entire questionnaire was devoted to the main target of phage therapy i.e., bacterial infections. As previously stated, this is also evidence that the respondents filled out the questionnaire carefully and understood the questions they were asked. In fact, we can only assume what would be the result without completing the questionnaire in a first place. Two questions that caused the most problems, Q5 and Q7 with only 11.8% and 12.4% of correct answers respectively, focused on phage therapy facilities and their activity. Such outcome is consistent with results of our survey described above where 76.2% out of 1,098 examined respondents have never heard anything about PTU in Wrocław, even though this facility was mentioned earlier in our questionnaire. Because PTU activity is not known, respondents also could not indicate correctly which infections are treated at PTU in Wrocław (Q7) as they are not familiar with its activity.

Noteworthy, an active consent required to take part in the quiz unintentionally became a selection factor. We can assume that the people who agreed to participate in the quiz had some knowledge on the subject, otherwise they would not be so eager to do it. In fact, the study population was screened three times. The first step was to agree to participate in the survey. The second stage was the completion of the entire questionnaire, and the third stage was to agree to take part in the quiz. Hence, there is a possibility that the general knowledge of the population is at a lower level than indicated by this quiz.

The results of the quiz were used for the total assessment of knowledge, and on their basis, a scale of General Knowledge about Bacteriophages was built. The possible number of points for giving a correct answer ranged from 0 to 12. The proposed scale was validated by analyzing the reliability of the scale items (Table 4). The value of the Cronbach's α reliability index was estimated, which ranges from 0 to 1 (the higher the value, the greater the reliability of the scale). In the literature, it is assumed that a reliable scale is one for which α > 0.7 (45). Looking at the results of the summary reliability analysis, we can see that the internal consistency reliability for this sum was estimated at 0.775. The magnitude of the Cronbach's α for a summary scale consisting of only 12 items is good. This value can be interpreted in such a way that approximately 78% of the variability of the total score is the variability of the true score, i.e., the true variability among respondents due to the concept (bias) common to all items.

All questionnaire items correlated positively with the final grade, as evidenced by the values of correlation coefficients r ranging from 0.277 to 0.532, whereas average inter-item correlation coefficient was 0.225. The estimated internal consistency of the GKB 12 scale indicates that it can be considered a reliable tool for use in scientific research.

In the group of 1,020 respondents, the mean knowledge score was 5.01 points (SD = 2.90) that corresponds to a rather average level of knowledge. The distribution of results deviated significantly from the theoretical normal distribution (see Supplementary Figure 1), therefore non-parametric significance tests were used to compare average knowledge in subgroups of respondents differing in sociodemographic characteristics.

Detailed analysis of median GKB-12 score in different age groups was shown in Figure 1. The lowest score (median = 3 points) was observed in the group of respondents aged up to 18 years (n = 54; 5.3% of the total) whereas 255 respondents (25%) aged 25–34 had the greatest knowledge (median = 6 points). It was similar in the group of people over 74 years of age, but there were only 7 of them (0.7%) and therefore in post-hoc tests the differences compared to other age groups were statistically insignificant.

Figure 1. Assessment of knowledge about bacteriophages on the GKB-12 scale in different age groups. Post-hoc analysis was done using the Dunn's test (the sizes of the analyzed groups are given in brackets).

3.3 Perception of experimental therapies

Prior to analyzing the public's awareness and perception of phage therapy and research we wanted to have an initial idea about opinions on experimental treatment in general. Our efforts were motivated by two reasons. First of all, work on the survey began during the COVID-19 pandemic, which has brought the importance of medical research and development of experimental therapies to the forefront of public awareness. Secondly, phage therapy in the European Union, including Poland, is conducted solely on an experimental basis and the number of clinical trials involving bacteriophages is constantly growing (46). Our goal was to identify the potential connection between the overall perception of experimental therapies and phage treatment.

Interesting outcomes regarding the perception of experimental therapies are shown in Figure 2. When asked about experimental therapies in general (Q2), respondents clearly acknowledged the need for such form of treatment (72.7%, Figure 2A). However, the question of whether experimental therapies prevail over the standard care (Q3) yielded quite the opposite response, i.e. the vast majority of respondents (86%, Figure 2B) did not agree with such a statement (the differences between Figures 2A, B are clearly visible). The distributions of answers to both questions are not uniform, which is confirmed by the result of the chi-square test (p < 0.001—see Supplementary Table 3). There is also a weak negative but statistically significant correlation (rS= −0.083, p < 0.05—see Supplementary Figure 2) between the ranks of answers to Q2 and Q3.

Figure 2. Responses to single-choice questions concerning perception of experimental therapies: (A) Q2; (B) Q3; (C) Q4; (D) Q7. Sample based on 1,098 responses.

Those responses were somehow confirmed in the following Q4 emphasizing the priority of standard therapies over experimental ones, where over three-quarters of the respondents (76.2%) would try standard of care in the first place (Figure 2C). In a Q7 designed to check the credibility of the previously given answers, again, the majority (75.3%) confirmed the need for experimental therapies if there is no other treatment option (Figure 2D). Such prioritization of approved therapies is consistent with our PTU regulations, which require ineffective cycles of antibiotic therapy in each patient to be documented before they qualify for phage therapy. The result of the chi-square test indicated the existence of differences between the distributions of answers to Q4 and Q7 (p < 0.001—see Supplementary Table 4) and the ranks of answers to these questions correlated weekly but significantly from zero (rS= 0.073, p < 0.05—see Supplementary Figure 3). Notably, 79.1% of interviewees were vaccinated against SARS-Cov-2 and only 10% of them did not plan to take any vaccine against coronavirus (Q5). These numbers are consistent with the above-mentioned results indicating that during the COVID-19 pandemic vaccines against coronavirus were considered the most effective form of prevention in the absence of standard therapies and, thus, most of the respondents chose to be vaccinated. According to the Polish Ministry of Health, 22.8 million people were fully vaccinated against COVID-19 in Poland as of May 2023 (47). Such a number translates into 60% of the total population in Poland i.e., much lower when compared to the population investigated in this study (79.1%). According to the responders themselves, media hype around the COVID-19 pandemic (Q6) increased interest in experimental therapies in 27.8% (answers definitely yes or probably yes) whereas more than half of them (56.4%) provided opposite answers (probably no or definitely no; p < 0.001). Similar results were obtained in the group of respondents who were employees of the science and research sector. In more than half of the respondents (56.3%), events related to the COVID-19 pandemic did not influence their interest in experimental therapies, and only 10.4% answered this question definitely yes (p < 0.001).

3.4 Awareness and perception of antibiotic resistance

Nearly half of the respondents (48%) encountered (or someone from their family) the ineffectiveness of antibiotics, while exactly one quarter of respondents (25%) did not know or did not remember (p < 0.001; Q8). The analysis of responses to Q9 revealed an unexpectedly high percentage of respondents who had heard about the phenomenon of bacteria acquiring resistance to antibiotics (94.4%) which certainly does not correspond with results obtained by other authors (38, 48, 49). Given the high proportion of science and research as well as health care professionals among responders in Arm A we checked whether such percentage was also achieved in Arm B consisting of more random people, likely not associated with science. It turned out that the result was only slightly lower although the difference was statistically significant (96.6% for Arm A compared to 90.6% for Arm B; p < 0.001; Table 5). We decided to investigate this phenomenon further in order to identify a group with the highest and lowest awareness of antibiotic resistance. Despite some differences among particular groups, the overall percentage of awareness remains high with no group scoring <66.7% (Table 5). Most of the respondents heard or read about antibiotic resistance on TV, radio, in the press or on the Internet (62.5%), 36.7% had heard about it at school or university (but this result was twice as high among students and amounted to 66.5%), 16.3% were familiar with antibiotic resistance because they underwent infection with ABR bacteria and 3.5% were physicians who had encountered that problem at their work. Among the 14% of individuals who chose the last option in Q10 (answer other) the majority were scientists who dealt with antibiotic resistance on a professional level or people who had relatives or friends dealing with infection caused by ABR bacteria.

Table 5. Awareness of antibiotic resistance between both tested datasets and selected groups (1,098 individuals in total).

When asked about potential source of infection with ABR bacteria (Q11) half of the respondents (49.9%) correctly identified that health care facilities are not the only source of such infections (answer definitely not true), however there were statistically significant differences between professionals and lay people (Figure 3). Concerns related to the possibility of getting infected with ABR bacteria (Q12) were expressed by 42.3% of responders in opposite to the 25.9% of those who did not afraid getting infected with such pathogens. We did not see any significant differences in this respect for health care professionals or science and research professionals vs. the rest of the population (Supplementary Table 5).

Figure 3. Comparison (in Mann-Whitney U test) of the awareness and perception of antibiotic resistance determined in Q11 (How true or untrue is, in your opinion, the following statement: You can get infected with bacteria resistant to antibiotics only in hospital departments or healthcare outpatient clinics?) between professionals (the employees of the health care and/or pharmaceutical sector) and the rest of responders (determined in Q33, the sizes of the analyzed groups are given in brackets below columns representing them).

3.5 Awareness and perception of bacteriophages and phage therapy

We started the analysis of the awareness of phage therapy by checking how much it corresponds with the awareness of antibiotic resistance of bacteria. It turned out that awareness of antibiotic resistance was closely associated with the score of general knowledge about bacteriophages and phage therapy (p < 0.001, Figure 4). Besides, between those who heard about antibiotic resistance one-third (30.7%) of respondents never heard anything about bacteriophages and 43.4% had never heard anything about phage therapy, whereas in a group of responders who never heard about antibiotic resistance those percentages were over twice higher (75.8% and 87.1% respectively) (Figure 5). Notably, awareness of antibiotic resistance (Q9; 94.4%) was twice as high as that of bacteriophages (Q13; 44.5%) and three times higher than that of phage therapy (Q14; 28.5%).

Figure 4. Assessment of knowledge about bacteriophages on the GKB-12 scale in groups of respondents differing in the answer to question Q9 (Have you ever heard that sometimes the bacteria that caused the disease can be resistant to antibiotics and that antibiotic therapy is then ineffective?) and the result of the Mann-Whitney significance test (the sizes of the analyzed groups are given in brackets).

Figure 5. Single-choice questions concerning awareness of bacteriophages (A) and phage therapy (B) in connection with knowledge of antibiotic resistance. Sample based on 1,098 responses. Questions are as follows:
Q13: Bacteriophages (also called phages) are bacteria-specific viruses. They infect them, multiply in them, and after multiplying they can destroy them. Have you ever heard of bacteriophages before?
Q14: Have you ever heard of phage therapy, which is based on the use of bacteriophages to treat infections caused, among others, by bacteria that have acquired resistance to various antibiotics?

The two above-mentioned questions concerning consciousness about bacteriophages (Q13) and about phage therapy (Q14) were summarized by sex, age, place of residence, level of education, employment status, profession, financial status (Tables 6, 7 respectively). Considering the objectives of the survey, these two questions are crucial and give a cross-sectional picture of the public awareness of phage research. Interestingly, there were statistically significant differences in the answers given to these two questions with females demonstrating a higher level of knowledge and awareness of both bacteriophages and phage treatment. Both males and females recognized more the term bacteriophage (60.6% and 70.3% respectively) than phage therapy (it remained unknown for around half of males and females). We observed an analogous relationship among people with higher education, whose knowledge about bacteriophages and their nature (58.9% in Table 6) was much higher than that of the essence of phage therapy (37.7% in Table 7). Similarly, we found statistically significant outcomes depending on the place of residence, level of education, employment status and, as expected, between lay people and health care/science and research professionals (Tables 6, 7). In addition, there were statistically significant differences in knowledge rate among people with different financial status with the highest knowledge rate about bacteriophages (56.2%) and phage therapy (37%) attributed to people describing their material status as definitely good. We did not notice statistically significant differences in awareness of bacteriophages and phage therapy (Q13–14) depending on people's experience with health care (Q36) (data not shown). In the age category (data not shown), the lowest percentages of positive responses were noted among adolescents up to 18 years old (60.7% had never heard about anything about bacteriophages and 66.1% had never heard about phage therapy), and people above 60 years old (44.6% and 57.1% respectively). This was in line with the median GKB-12 scores (Figure 1) which were the lowest in adolescents. Interestingly, the correlation between all respondents' ratings on the GKB-12 scale and their answers to Q13 and Q14 was moderately positive (Spearman's rho was 0.553 and 0.571 respectively; p < 0.001—see Supplementary Figure 4). This may be due to the fact that the test questions were not easy, and the substantive knowledge about bacteriophages and phage therapy is much weaker than the awareness about them.

Table 6. Self-assessment by responders of their knowledge of bacteriophages (n = 1,098).

Table 7. Awareness of phage therapy in the study population (n = 1,098).

Because respondents from Arm A could be more associated with us (e.g., by reaching them through e-mails or announcements on Institute's websites) than responders from Arm B (responses collected from an external source) additional analyses were conducted. We compared awareness of the existence of bacteriophages and phage therapy (Q13 and Q14) between lay people in Arm A and Arm B (Supplementary Tables 6, 7). Indeed, significantly more lay people from Arm A (n = 185) were familiar with term bacteriophages (64.9%) or phage therapy (51.4%) contrary to respondents (n = 375) from Arm B (35.7% and 28.3% respectively). Taking under consideration above differences and overrepresentation of health care as well as science and research professionals in our survey we can assume that the overall knowledge and awareness of bacteriophages and phage research in the Polish society is more similar to numbers attributed to Arm B.

The last analysis concerning the awareness of phage therapy involved question on whether research on phage therapy should be further developed in Poland (Q24). This idea got support from a very large percentage of all respondents (88.0%), but what is worth emphasizing also from lay people (83.2%). There was a clear correlation between awareness of phage therapy (Q14) and strong belief that research on phage treatment should be developed (Figure 6, Supplementary Figure 5). Twice as many people who have heard about phage therapy and know what it is (83.1%) support its development compar