Adequate health literacy is required to deal with the overwhelming amount of inaccurate health information. As defined by Sørensen et al. (11), health literacy refers to “knowledge, motivation, and competencies to access, understand, appraise, and apply health information in order to make judgments and take decisions in day-to-day life concerning healthcare, disease prevention, and health promotion to maintain or improve quality of life over the lifetime”. The World Health Organization defines health literacy as the cognitive and social abilities that affect an individual's motivation and capacity to obtain, absorb and use knowledge to maintain and develop their health (12). Health literacy empowers a person to be proactive in maintaining their health and confers the ability to take action and make well-informed decisions. Limited health literacy has been found to be significantly associated with poor health status, high use of healthcare services, low socioeconomic status, lower education and older age (13). Studies have shown that limited health literacy is indirectly and directly associated with poor health and clinical outcomes (14–16). Data from different developed countries have demonstrated that limited health literacy is associated with the declining use of public health services and information and disease prevention services and poorer self-management of diseases (17). Conversely, individuals with a higher level of health literacy have less frequent use of hospital services and better health status, due to greater disease knowledge, healthier behaviors, greater use of preventative treatment and greater drug compliance (18). Recent research has also demonstrated that health literacy may be seen as a social vaccine, as it enables individuals and communities to reduce the spread of viruses by comprehending and using the information supplied by governments and health authorities (19). Thus, adequate health literacy is essential to manage the current pandemic. Despite the concerns about and importance of COVID-19-specific health literacy, there is little empirical research on this topic, with only one study performed in Taiwan (20) and one in Germany (21). Thus, COVID-19-specific health literacy is a crucial research topic during this global health emergency.

In a recent study of COVID-19 vaccine hesitancy conducted in seven European countries, 18.9% of the sample of more than 7,600 respondents were uncertain of their intention to receive the COVID-19 vaccine, and 7.2% reported their unwillingness to be vaccinated (22). Studies have developed models, such as the 5C model (23), to integrate data regarding vaccination behavior. However, COVID-19 vaccines differ from previous vaccines in many aspects, such as the speed of their development, vaccine effectiveness and potential side effects (24). COVID-19 vaccine hesitancy has been shown to be dependent on context, place, time and the type of vaccine (25). A recent study of school principals in Taiwan reported that individuals with limited COVID-19-specific health literacy had a lower intention to be vaccinated against COVID-19 (20). Moreover, a study of Chinese college teachers and students found that the key factors influencing their reluctance to receive the COVID-19 vaccine were beliefs about the safety of the vaccines, attention to and awareness of vaccine-related news and chronic medical issues. However, students' vaccine hesitancy was unrelated to teachers' vaccine hesitancy (26). A recent French study conducted by Montagni et al. (27) also revealed that fake news detection scores of adults were associated with the intention to receive the COVID-19 vaccine. Altogether, these findings highlight that inadequate health literacy may affect individuals' hesitancy to receive the COVID-19 vaccine, as they may be concerned about its safety and affected by news about the vaccine. Misinformation may deeply influence their decision on whether to receive the vaccine. Thus, the association between inadequate health literacy and COVID-19 vaccine hesitancy is worth investigating.

Education is one of the social factors affecting the health literacy of the general population, especially children and adolescents, as they are at their most receptive period of life (28). The long-term development of children's and adolescents' health and well-being is expected to benefit from investments in health literacy (29, 30). Investing in health literacy system capabilities requires a systematic effort and transformation that may multiply and sustain over time and that is resilient to external trends and events, rather than relying solely on organizational and individual behavioral changes (31). Moreover, scholars have reported that improving health literacy across populations and systems is critical to achieving health equity (31). Consequently, it is important to recognize that health literacy in schools is a major public health issue (32). In this regard, schools have long been seen as a crucial setting in which to promote disease prevention and health promotion and develop the health literacy of students, teachers and administrators (33–36). The health literacy of children and adolescents has long been a popular research topic, but equal attention should be given to the health literacy of teachers, as they are responsible for enhancing the students' understanding of health-related topics. Peterson et al. (2001) defined teachers' health literacy as “the capacity of teachers to obtain, interpret and understand basic health information and services, with the competence to use such information and services in ways that enhance the learning of health concepts and skills by school students” (37). Teachers' health literacy may be as important as students' health literacy, as teachers are like health information providers, while students are the consumers (37). However, the prevalence of limited health literacy is concerning. A study of secondary school teachers in Sri Lanka and Iran found that the prevalence of problematic or inadequate health literacy was 31.5 and 50.6%, respectively (38, 39). In Turkey, a study of 500 teachers found that more than 70% of them had extremely low or low levels of health literacy (40). A survey of 1,000 educators conducted in Japan during the COVID-19 pandemic, using the European Health Literacy Survey Questionnaire (HLS-EU-Q47), revealed that educators had gained sufficient health literacy, with a score of 33.5 out of 50 (41). Under the context of COVID-19, a recent study of school principals in Germany revealed a prevalence of 29.3% for limited COVID-19-specific health literacy. These findings demonstrate that sufficient health literacy levels are place- and time-specific. Therefore, research on COVID-19-specific health literacy among school teachers in Hong Kong is urgently needed, especially as research on this topic in Hong Kong has been limited for many years.

Making appropriate health decisions based on relevant information may be beneficial in occupational and health settings (42). Studies have also shown that limited health literacy may be one of the barriers to the understanding and effectiveness of occupational training (43). A study of young adult workers also revealed that six components of their proposed health literacy structural model, namely, “self-perception,” “a proactive approach to health,” “self-control,” “self-regulation” and “communication and cooperation,” were associated with work ability (44). Although an Iranian study reported no significant relationship between job stress and health literacy (45), other studies have suggested that uncertainty related to COVID-19 has the potential to increase people's stress, anxiety, risk of burnout, fear and frustration levels (46–53). Educators have been accumulating negative psychological symptoms, such as stress, anxiety and depression, due to the closure of educational facilities globally and the need to adapt to new teaching methods since the beginning of the COVID-19 pandemic (54–56). A recent systematic review showed that the estimated prevalence of stress among teachers globally was 30% (57). In the context of Hong Kong, a local survey revealed that more than 85% of teachers felt stressed or over-stressed, and more than one third of people surveyed worked 61 hours or more per week (58, 59). These findings indicate that educators in Hong Kong are an occupational group experiencing a high level of stress since the beginning of the COVID-19 pandemic. In the context of a heavier workload and a greater need for self-management, researchers have hypothesized that self-endangering behavior, including work intensification, extensification and quality reduction, may be used as coping mechanisms (60). However, engaging in these behaviors may be harmful to one's physical health and long-term capacity for employment (60). Burnout syndrome, which includes psychosomatic complaints and exhaustion, and mild to severe mental disorders, such as anxiety and depression, may occur due to work overload or stress (61, 62). Therefore, investigating health literacy and its relationship to self-endangering work behavior and burnout syndrome may facilitate the development of interventions aimed at improving the mental health of teachers globally.

In Hong Kong, the health literacy level of the population is concerning. A recent study reported that 20.7% and 35.2% of the general population had inadequate or problematic health literacy levels, respectively (63). Inadequate levels of health literacy were also found to be prevalent among older adults in Hong Kong (64). However, the health literacy of teachers in Hong Kong has rarely been discussed, even though the unexpected digital revolution in teaching due to the COVID-19 pandemic has brought major teaching challenges, such as technical and motivational problems (65). According to a survey of 1,200 teachers in Hong Kong, conducted by the Hong Kong Federation of Education Workers in November 2020, 85% of teachers felt “relatively high” or “very high” levels of pressure at work during the pandemic. This prevalence is 5% higher than the prevalence reported in the previous year (66).

In this study, we aimed to examine COVID-19-related health literacy in teachers in Hong Kong and explore how different demographic variables are associated with health literacy and how health literacy levels predict the level of COVID-19-related information received, self-endangering work behavior and burnout symptoms. Moreover, we explored how COVID-19-related health literacy affected teachers' decisions regarding and attitudes toward vaccination.

The sample was composed of teachers in primary, secondary and special schools in Hong Kong. Participants were recruited through email invitations. English and Chinese versions of an online survey on the Qualtrics platform were sent to the principals of 1,130 schools, including 561 primary schools, 477 secondary schools and 36 special schools, registered with the Education Bureau in Hong Kong. The school principals were contacted via our established school network, with priori verbal agreement granted over the phone. Hardcopies of the surveys were mailed to 243 schools. In addition, we sent invitations via our personal and professional networks through social media platforms (e.g., WhatsApp), and respondents were asked to invite eligible participants to participate. Participation was anonymous and voluntary and no incentives were provided. Confidentiality of information was guaranteed. The study protocol was approved by the research ethics committee of Hong Kong Baptist University (REC/20-21/0465).

Demographic characteristics, including sex, age (years) and school type (primary school, secondary school or special school), were ascertained.

The 22-item HLS-COVID-Q22 scale (21) was used to measure the participants' capacity to understand and utilize COVID-19-related health information. The validated scale has high internal consistency (α = 0.940) and has been shown to be reliable for measuring COVID-19-related health literacy (21). The participants scored on a 4-point scale, from 1 (very difficult) to 4 (very easy), their perceived difficulties in accessing, understanding, appraising and applying health-related information in the context of the COVID-19 pandemic. Responses for the items were averaged to form the scale score, and the attainment of sufficient health literacy was indicated by an average score of 3. An average score less than or equal to 2.5 indicated inadequate health literacy, while an average score >2.5 but < 3 indicated problematic health literacy (21). A Cronbach's alpha value of 0.957 was obtained in the present study.

The adequacy and clarity of health information in the context of the COVID-19 pandemic were measured using the following two items: “On a scale of 1 (very informed) to 5 (insufficiently informed), how well-informed do you feel about the coronavirus or the COVID-19 pandemic?” and “On a scale of 1 (not at all confused) to 4 (very confused), do you feel confused about COVID-19-related information?” (21). The willingness to get vaccinated against COVID-19, if offered, was also measured on a 6-point scale, from “yes, certainly” to “certainly not”, or “I am already vaccinated”. Attitudes toward vaccination were assessed using five self-reported items. Examples of these items included “vaccinations are important to protect me and my family” and “vaccination is compatible with my attitudes”. The participants were asked to rate their agreement of each item using a 4-point scale, from 1 (totally agree) to 4 (do not agree at all). The responses for the items were averaged to calculate the scale score.

Health promotion and disease prevention in schools were measured using a 15-item questionnaire based on a study by Dadaczynski and Hering (67). The questionnaire measured a range of health issues addressed by schools in the context of the COVID-19 pandemic. The participants rated how they agree with the 15 statements (e.g., “students learned how to eat healthily despite the restrictions due to the coronavirus”), on a 4-point scale, from 1 (not true at all) to 4 (totally true). The responses for each item were summed and averaged to calculate the scale score. Higher scores indicated greater disease prevention and health promotion efforts at the school.

The three-item “exhaustion” subscale of the Burnout Assessment Tool was used to assess exhaustion related to work, while psychological discomfort was measured using the 5-item “psychosomatic complaints” subscale (68). The participants rated how often the statement applied to them on a five frequency-based response scale, from 1 (never) to 5 (always). The responses for each item were summed and averaged to calculate the total score. The Cronbach's alpha values of the exhaustion and exhaustion subscales in the present study were 0.932 and 0.801, respectively.

Self-endangering work behaviors were assessed using three subscales of self-reported self-endangering work behavior scales (60), including “extensification of work” (six items), “intensification of work” (three items) and “work quality reduction” (three items). The reliability of the scale has been shown previously (60). The participants rated how often the statements applied to them in the past 3 months, on a 5-point scale from 1 (never) to 5 (very often).

The number of teaching hours per week, the number of working hours per week, the change in the number of working hours per week due to the COVID-19 pandemic and the number of students at the school were collected. The participation status in health-promoting programmes at the participants' schools was also determined (never, < 1 year, 1 to < 2 years, 2 to < 3 years, 3 years or above).

The general health of the respondents was assessed through the self-reported question “How is your overall health?”, on a 5-point scale, from 1 (very good) to 5 (very poor). The participants were asked if they had any chronic or long-term health problems, answering 1 for “no” and 2 for “yes”, and to rate the degree to which their chronic disease impaired their activities of daily living, on a 3-point scale from 1 (no impairment at all) to 3 (severe impairment) (69).

Continuous variables were reported using the mean (M) and standard deviation (SD). Categorical variables were computed as the number (n) and percentage. Mean differences in COVID-19-specific health literacy among demographic groups (i.e., by age, sex, type of schools and number of students) were explored using independent Student's t-tests. Potential predictors of COVID-19-specific health literacy levels (i.e., age, sex and level of health promotion and disease prevention in schools) were explored using adjusted multiple linear regression models. Separate adjusted multiple linear regression models were computed to detect the predictive power of COVID-19-specific health literacy for COVID-19 vaccine hesitancy. Adjusted multiple linear regression models were used to explore how COVID-19-specific health literacy affected one's perceived general health. Mean score differences in self-endangering work behavior (i.e., extensification of work, intensification of work and work quality reduction), burnout symptoms (i.e., exhaustion related to work situation and psychosomatic complaints) and the level of knowledge of COVID-19- or pandemic-related information were compared between health literacy groups using an independent samples Student's t-test or one-way analysis of variance (ANOVA). All data analyses were performed using SPSS 27.0 for Windows (IBM, Armonk, NY, USA). Statistical significance was defined as a p-value < 0.05.

Of the 366 teacher participants (Table 1), 46.4% were male, and the average age was 38.3 (SD = 9.72) years. Almost half of the participants (45.1%) worked in secondary schools, 33.8% in primary schools and 21.2% in special schools. The average number of teaching hours per week was 21.9 (SD = 10.15), and the average number of working hours per week was 45.4 (SD = 16.18, ranging from 4 to 115 h). The number of working hours per week was higher during than before the COVID-19 pandemic for 51.2% of the participants, about the same for 36.0% of the participants and lower during than before the COVID-19 pandemic for 12.7% of the participants. The average number of students at the school was 516.16 (SD = 315.9).

There was no significant sex difference in HLS-COVID scores (males: M = 2.97, SD = 0.41; females: M = 2.98, SD = 0.43). A one-way ANOVA revealed that there was a statistically significant difference in HLS-COVID scores between participants who worked in primary schools, secondary schools and special schools (F_{[2, 351]} = 12.59, p < 0.001). Tukey's honestly significant difference (HSD) test for multiple comparisons showed that participants who worked in special schools had a significantly lower HLS-COVID score than those who worked in primary schools (p < 0.001, 95% confidence interval [CI] = −0.1699, −0.4036) or secondary schools (p < 0.001, 95% CI = −0.3467, −0.1235). There was no statistically significant difference in HLS-COVID scores between participants working in primary schools and those working in secondary schools (p = 0.296).

A one-way ANOVA revealed a statistically significant difference in HLS-COVID scores according to the change in the number of working hours per week during the pandemic (F_{[2, 349]} = 9.667, p < 0.001). Tukey's HSD test for multiple comparisons found that participants with a higher number of working hours per week during the COVID-19 pandemic had significantly lower HLS-COVID scores than individuals with a lower number of working hours per week during the COVID-19 pandemic (p = 0.03, 95% CI = −0.2797, −0.0115) and individuals with no change in the number of working hours per week (p < 0.001, 95% CI = −0.3003,−1.1117). An independent samples Student's t-test revealed a statistically significant difference in HLS-COVID scores between the group of participants who had 600 or fewer students at their schools and the group of participants who had more than 600 students at their schools (t_[354] = 2.274, p = 0.009). The results also showed that the participants who felt well informed or very well informed about COVID-19-related information had significantly higher health literacy scores (M = 3.15, SD = 0.45) than those who felt insufficiently or poorly informed about COVID-19-related information (M = 2.91, SD = 0.39; t_[350] = 5.38, p < 0.001). The participants who felt not at all or a little confused about COVID-19-related information (M = 3.04, SD = 0.39) had higher health literacy scores than those who felt very confused.

Table 2 provides the descriptive statistics of health-related variables. The average total HLS-COVID-Q22 score for the total sample was 2.98 ± 0.42. Sufficient health literacy was observed in 50.8% of the participants, whilst 47.4% participants had insufficient health literacy (including problematic or inadequate literacy levels). The majority of the participants (71.3%) had insufficient, poor or satisfactory pandemic-related information, and 77.3% of the participants felt not at all or a little confused about pandemic-related information. Almost half of the participants (49.0%) had already received a COVID-19 vaccine. The total mean scores for attitudes toward vaccination, health promotion and disease prevention in schools and perceived general health were 2.13 (SD = 0.65), 44.6 (SD = 6.49) and 2.65 (SD = 0.80), respectively. Most of the participants (82.7%) did not present with any chronic disease or long-lasting health problems. The majority of the participants (72.4%) were not at all impaired by health problems.

Table 3 shows the descriptive statistics for work satisfaction, self-endangering work behavior (i.e., extensification of work, intensification of work and work quality reduction) and secondary burnout symptoms (i.e., exhaustion related to work situations and psychosomatic complaints) for the entire sample and the differences between the two health literacy groups: sufficient health literacy (n = 186) and insufficient health literacy (n = 170). Respondents with sufficient health literacy (M = 2.95, SD = 0.87) had significantly lower intensification of work scores than those with insufficient health literacy (M = 3.23, SD = 0.88; t_[352] = 3.004, p = 0.003). After excluding participants who had already been vaccinated, those who had sufficient health literacy (M = 1.92, SD = 1.09) scored lower for their attitudes toward COVID-19 vaccination (i.e., felt more positive about COVID-19 vaccination) than those with insufficient health literacy (t_[180] = 4.168, p < 0.001). In addition, teachers with sufficient health literacy (M = 2.60, SD = 0.72) felt more informed about COVID-19-related information than those with insufficient health literacy (M = 2.97, SD = 0.62; t_[354] = 5.142, p < 0.001). Moreover, those with sufficient health literacy (M = 1.91, SD = 0.67) felt less confused about COVID-19-related information than those with insufficient health literacy (M = 2.24, SD = 0.63; t_[350] = 4.68, p < 0.001).

Multiple linear regression analyses were performed to explore the association between health literacy level and other potential influential factors (i.e., age, sex and number of working hours and teaching hours per week). Table 4 provides the regression coefficients for predicting the health literacy level. The model was adjusted for age and sex. A significant regression equation was found (F_{[4, 312]} = 6.500, p < 0.001) with an R-squared value of 0.077. The results showed that age and the number of teaching hours per week were significant predictors of HLS-COVID scores. Specifically, age positively predicted the HLS-COVID score, and the number of teaching hours per week negatively predicted the HLS-COVID score. Separate multilinear regression models adjusted for age, sex and the presence of chronic illness were generated to predict attitudes toward vaccination (Table 5). The level of COVID-19-specific health literacy was negatively associated with attitudes toward COVID-19 vaccination, while the level of confusion about COVID-19-related information was positively associated with attitudes toward COVID-19 vaccination (i.e., less agreeable to vaccination; F_{[6, 158]} = 5.865, p < 0.001), with an R-squared value of 0.182.