Introduction

The COVID-19 pandemic profoundly impacted global public health, prompting unprecedented efforts to combat its spread, including the widespread distribution of vaccines. The Chinese government was at the forefront of these efforts, supplying vaccines for both domestic use and distribution to developing countries. As a result, the vaccination coverage rate in Mainland China has exceeded 90% (China Daily, 2022). Despite these extensive vaccination efforts, vaccine hesitancy still poses a significant challenge in achieving widespread immunisation (Robinson et al. 2022; Teng et al. 2022). Accurate knowledge and understanding of vaccines are crucial for addressing this challenge and promoting vaccination acceptance.

Recent research highlights the positive association between vaccination knowledge and willingness with the concept of eHealth literacy, which encompasses the ability to access, comprehend, and apply online health-related information to promote health behaviours (Lwin et al. 2020; Olisarova et al. 2021). However, existing measures of vaccination knowledge have not been sufficiently comprehensive or validated in prior research. This calls for the development of a robust vaccination knowledge scale that promotes eHealth literacy and health status in China.

This study achieved two primary objectives. First, it developed a comprehensive vaccination knowledge scale tailored to the Chinese population by considering factors such as eHealth literacy and health status. Second, it elucidated the relative importance of each component within the overarching framework, including vaccination knowledge, willingness and eHealth literacy. The insights gained from this study have the potential to inform policymakers, public health professionals, vaccine developers, communication strategists and media practitioners on how to enhance public confidence in COVID-19 immunisation efforts.

Theoretical perspective in understanding willingness to vaccinate

The COVID-19 pandemic sparked global urgency to accelerate vaccination efforts, with various countries implementing vaccination campaigns to curb its spread. In China, the Sinopharm BIBP vaccine was authorised for emergency use in June 2020, marking a significant milestone in the country’s response to the pandemic (Huang et al. 2023). The rollout of COVID-19 vaccines was accompanied by efforts to monitor vaccination clinics and evaluate adverse reactions among vaccine recipients. A follow-up cross-sectional online survey was conducted to assess the willingness, hesitancy and coverage of the COVID-19 vaccine in China (Wang et al. 2021), which revealed that while there was a relatively high level of trust in the vaccine, delivery system and government, vaccine hesitancy remained prevalent among participants, with only 34.4% reporting vaccination uptake. These findings underscore the importance of addressing vaccine hesitancy and enhancing public confidence in vaccination efforts to effectively control the spread of the disease.

Previous research has emphasised the critical role of high vaccination coverage to achieve herd immunity and halt the spread of the virus. For instance, Tomietto et al. (2022) provided valuable insights into generational patterns in vaccine acceptance by examining the willingness to vaccinate among healthcare students and professionals, revealing that older healthcare professionals in China demonstrated a higher willingness to receive the COVID-19 vaccine compared to their younger counterparts. However, concerns about vaccine safety, reliability and effectiveness have contributed to vaccine hesitancy, prompting individuals to postpone or refuse vaccination (Betsch et al. 2017; DeRoo et al. 2020; Leng et al. 2021; Mathieu et al. 2021). Therefore, it is essential to eliminate vaccination reluctance and boost interest in COVID-19 vaccination through targeted public health interventions (Dror et al. 2020; Lazarus et al. 2022).

Vaccination knowledge plays a crucial role in shaping individuals’ attitudes towards vaccination and their willingness to receive it. Accurate information and understanding of vaccines are essential to increase immunisation acceptance and protect public health (Fakonti et al. 2022). Previous studies have demonstrated a positive correlation between vaccination knowledge and willingness to receive vaccines (Schulz and Hartung, 2021; Zheng et al. 2022). Despite the importance of vaccination knowledge, existing scales for measuring general vaccination knowledge remain underdeveloped and require further validation. While some studies have attempted to assess vaccination knowledge, limitations such as reliance on literature reviews, lack of generalisability and insufficient consideration of social and cultural factors may negatively impact the reliability and restrict the applicability of the findings (Gaglia et al. 2007; Johnson et al. 2008; Ridda et al. 2008).

Understanding the willingness to vaccinate is essential for guiding effective vaccination strategies and achieving population immunity against COVID-19. Addressing vaccine hesitancy, enhancing vaccination knowledge and considering sociocultural factors are important steps in promoting vaccine acceptance and preventing the spread of COVID-19 to pandemic levels in the future. Further research is needed to develop comprehensive measures of vaccination knowledge and explore effective interventions to improve vaccine acceptance rates globally.

eHealth literacy and vaccination intention

eHealth literacy, defined as the ability to access, understand and utilise online health-related information to make informed decisions and adopt healthy behaviours (Lwin et al. 2020; Olisarova et al. 2021), plays a pivotal role in shaping individuals’ attitudes towards vaccination and their willingness to engage in preventive health practices. Britt et al. (2017) demonstrated a positive association between eHealth literacy levels among college students in the United States, their inclination to seek vaccination information and future vaccination willingness. A study in Kenya (Muturi, 2020) revealed that young adults with higher eHealth literacy had greater knowledge of the human papillomavirus and its vaccine. Qin et al. (2021) examined eHealth literacy and vaccination knowledge among Chinese university students and found that individuals with higher eHealth literacy were more knowledgeable about the COVID-19 vaccination and more likely to adhere to preventive measures.

Several studies have underscored the important impact of eHealth literacy on healthcare decision-making, particularly in promoting infection-prevention behaviours (Dib et al. 2022; Li and Liu, 2020). A recent study conducted in China examined the interplay between health literacy, stress and COVID-19 vaccine hesitancy, focusing on the potential moderating effects of stress on this relationship (Zhang et al. 2022).

These findings highlight the significant role of health literacy in shaping vaccine acceptance and underscore the importance of considering psychological factors, such as stress, alongside health literacy when designing interventions to promote COVID-19 vaccination. The study’s emphasis on the Chinese context provides valuable insights into the unique factors influencing vaccine acceptance in this population. Moreover, the findings shed light on the complexity of vaccine hesitancy and underscore the need for multifaceted approaches that address both cognitive and emotional factors driving individuals’ vaccination decisions. Additional research could identify and explore further moderators that may influence the relationship between health literacy and vaccine hesitancy. Moreover, there is a need to investigate the efficacy of interventions specifically designed to address stress management in diverse populations to bolster vaccine acceptance.

Montagni et al. (2021) conducted a study in France that revealed a negative correlation between health literacy and vaccine acceptance among adults with low health literacy levels. Liu et al. (2022) found notable disparities in COVID-19 vaccine acceptance rates among Chinese medical students based on their varying levels of eHealth literacy. These studies underscore the intricate interplay between eHealth literacy, vaccination knowledge and willingness. Specifically, higher levels of eHealth literacy appear to indirectly influence vaccination willingness by fostering enhanced knowledge and cultivating positive attitudes towards vaccination.

In summary, eHealth literacy plays a significant role in determining vaccination intention, with consequences for both encouraging vaccination uptake and reducing vaccine reluctance. To improve public health outcomes, research needs to clarify the underlying mechanisms and pathways by which eHealth literacy affects vaccination attitudes and behaviours. This would facilitate the development of targeted treatments. We developed five hypotheses and three research questions (RQs) for the purpose of addressing this research gap.

RQ1 is as follows: How scalable is the vaccination knowledge measurement process, and how does the hierarchical arrangement of these items change?

Correspondingly, the following hypothesis (H) is proposed to explore RQ1, which provides the foundation for comprehending the architecture and scalability of vaccination knowledge assessment—a crucial aspect of developing successful interventions and instructional initiatives:

H1: The vaccination knowledge assessment procedure will exhibit scalability, indicating consistent and dependable measurement across various vaccination knowledge thresholds.

Another research question investigates how eHealth literacy affects people’s willingness and knowledge about vaccinations. Our comprehension of the measuring procedure and hierarchical organisation of vaccination knowledge items is enhanced by this shift, which enables us to delve deeper into the individual-level factors influencing vaccine-related attitudes and actions. The following RQ2 and two hypotheses are put forth:

RQ2: What impact does eHealth literacy have on people’s understanding of vaccinations and their willingness to vaccinate?

H2: Vaccination knowledge will be positively correlated with higher levels of eHealth literacy, meaning that those with higher levels of eHealth literacy will know more about vaccines.

H3: Higher levels of eHealth literacy will be positively associated with individuals’ willingness to get vaccinated, suggesting that individuals with greater eHealth literacy will be more inclined to receive vaccinations.

Our findings regarding RQ2 shall provide insights into the direct impact of eHealth literacy on both vaccination knowledge and willingness. However, to gain a deeper understanding of the mechanisms underlying these relationships, we explore the role of vaccination knowledge as a potential mediator and the moderating effect of health status. A third research question allows us to explore the underlying processes driving individuals’ vaccination decisions, considering both their level of eHealth literacy and health status. This expanded perspective enhances our understanding of the complex interplay between eHealth literacy, vaccination knowledge and willingness, contributing valuable insights to health communication strategies aimed at promoting vaccination uptake. The following RQ3 is raised:

RQ3: Considering the moderating effect of health status, how does vaccination knowledge mediate the association between eHealth literacy and vaccination willingness?

In pursuit of RQ3, we investigate the potential mediating role of vaccination knowledge and its interaction with health status in shaping individuals’ vaccination willingness. We accordingly propose the following hypotheses:

H4: The association between eHealth literacy and vaccination willingness will be mediated by vaccination knowledge, such that higher levels of eHealth literacy will be associated with greater vaccination knowledge, which in turn will be associated with increased vaccination willingness.

H5: Health status will moderate the indirect effect of eHealth literacy on vaccination willingness through vaccination knowledge, such that the indirect effect will be stronger for individuals with poorer health status compared to those with better health status.

Figure 1 presents the research questions and hypotheses that have been organised in alignment with the objectives of this study. This schema comprises three research questions and five hypotheses that specifically target the two primary research objectives.

Fig. 1
figure 1

Schematic diagram of the research questions and hypotheses of this study.

Full size image

Methods

To address the research questions concerning the fundamental mechanisms influencing the relationship between individuals’ knowledge of vaccines and eHealth literacy, a survey targeting Chinese adults included the following components: assessment of eHealth literacy, evaluation of vaccination knowledge, perception of vaccine information, attitudes towards vaccination and demographic information, including age, gender, educational level, occupation and health status. This information would establish how these factors influence the relationship between eHealth literacy and vaccination knowledge.

Procedure

A web-based survey was employed to assess eHealth literacy and vaccination awareness among Chinese individuals aged 18 and older. The study received approval from the Ethics Committee for Social Sciences and Humanities Research from the University of Macau (approval code: SSHRE22-APP020-FSS; approval date: 13 May 2022), ensuring adherence to the ethical guidelines for human subject research. The survey was administered through the Qualtrics XM platform, with distribution facilitated using popular local social media platforms, such as WeChat and QQ, from May to October 2022. Participants who understood Mandarin Chinese and were residing in Mainland China, Macao and Taiwan since the onset of the COVID-19 pandemic in spring 2020, were recruited for the study. Recruitment utilised exponential non-discriminative snowball sampling, whereby initial respondents referred additional participants from their social networks, creating a chain referral process until an adequate sample size of 950 participants was attained. Participation was voluntary and anonymous.

Measurements

Vaccination knowledge

In this study, nine items were adopted from previous studies to assess vaccination knowledge and predict perceived general knowledge about vaccines and infectious diseases gleaned (Ruiz and Bell, 2021; Zingg and Siegrist, 2012). The questionnaire contained four correct and five incorrect statements for measuring vaccination knowledge. Each item contained three categories: 0 for “don’t know,” 1 for “incorrect,” and 2 for “correct.” To some extent, a high total score reflected a high level of vaccination knowledge. Supplementary Table S1 in the Appendix shows the detailed items of vaccination knowledge.

Willingness to get vaccinated

One question assessed respondents’ readiness for immunisation when the coronavirus vaccine became available (Ruiz and Bell, 2021). Responses ranged from 1 (very unlikely) to 5 (highly likely). A high score indicated that the participant was willing to get vaccinated.

eHealth literacy

Eight questions from previous studies were used to gauge the respondents’ eHealth literacy or capacity to handle and access health-related information online (Chang and Schulz, 2018; Lwin et al. 2020). A high total score indicated a high level of eHealth literacy on a 5-point Likert scale, with 1 denoting “strongly disagree” and five denoting “strongly agree.”

Covariates

Four variables were considered covariates: gender, age, education level and health status (Ruiz and Bell, 2021; Zheng et al. 2022). Health status is a measure of how people perceive their health and is rated as poor, fair, good, very good, or excellent. Supplementary Table S2 in the Appendix provides detailed information on the measurements of the variables.

Data processing

RStudio with the R 4.2.0 programming language (RStudio PBC, Boston, United States) was used to process the data throughout this study. The sociodemographic variables of the respondents and the relationships between eHealth literacy, health status, vaccination knowledge and willingness were investigated using descriptive analysis and a correlation matrix, respectively. The degree of scalability of a set of items was examined using Mokken scale analysis, a nonparametric technique (Van Abswoude et al. 2004; Van Schuur, 2003; Zhang and Li, 2020). This method is used to determine which items of vaccination knowledge should be included in the scale and to what extent the items are positioned hierarchically based on the construct being measured and the level of vaccination knowledge.

The link between eHealth literacy and vaccination willingness can be examined using causal mediation analysis (Tingley et al. 2014) to identify the strength and direction of the relationship between these variables. Using multiple linear regressions under the 95% bootstrap confidence intervals (CI) with the bias-corrected and accelerated method by adopting the 2022 random seeds and 1000 simulations as parameters, we investigated the mediation effect of vaccination knowledge on the relationship between eHealth literacy and vaccination willingness. Additionally, the overall effect of vaccination information was used to determine the relative magnitude of the impact sizes.

Directed acyclic graphs (DAGs) were used to illustrate the mediating effects. For instance, previous studies indicated that DAGs serve as a concise guide for researchers in the social sciences, psychiatry epidemiology and psychological sciences by distinguishing structural equation models (SEMs) and causal-directed acyclic graphs (DAGs)—two common visual diagramming methods used in causal-inference research. While SEMs are both conceptual and statistical tools where models are drawn and tested, DAGs are purely conceptual tools aiding in the development of analytic strategies and result interpretation, used explicitly for causal inference (Kunicki et al. 2023; Lipsky and Greenland, 2022; Matthay and Glymour, 2020).

By viewing DAGs as qualitative schematics for some SEMs and SEMs as algebraic systems for DAGs (Kunicki et al. 2023), the study highlights the complementary nature of DAGs. As psychological and psychiatric epidemiology increasingly integrate both causal modelling and latent-variable concepts, the ability to understand and combine the tools of DAGs becomes valuable in social sciences communication. Therefore, the DAGs revealed vaccination willingness as the dependent variable (Y), vaccination knowledge as the mediator (M) and eHealth literacy as the independent variable (X) in this study. The direct impact of eHealth literacy on vaccination willingness is indicated by the arrow pointing from X to Y. The indirect effect of eHealth literacy on vaccination willingness through vaccination knowledge is indicated by the arrows pointing from X to M to Y.

Results

Sample characteristics

A total of 950 participants answered questions regarding their vaccination-related knowledge, attitudes and intentions. After excluding 267 incomplete surveys (28.1%), the analysis contained complete information from 683 participants who answered the first nine questions on vaccination knowledge. Of the respondents, 487 (71.3%) were from Mainland China and 196 (28.7%) were from Macao and Taiwan. More than half the respondents were female (60.9%, n = 416), aged between 18 and 29 years (67.8%, n = 463) and had a college degree (68.5%, n = 468). A total of 97.7% of participants (n = 667) reported having continuous health concerns. Of the respondents, 19.9% reported poor and fair health (n = 136), 44.7% reported good health (n = 305), 24.7% reported very good health (n = 169) and 8.3% reported exceptional health (n = 57). Supplementary Table S3 in the Appendix exhibits the sociodemographic analysis of the respondents in 2022.

Vaccination knowledge analysis

RQ1 addressed the scalability of the vaccination knowledge measurement process and how the hierarchical arrangement of these items varied, based on vaccination knowledge thresholds. The Mokken scale analysis (MSA) is a nonparametric technique used to assess the degree of scalability of a set of items intended to measure a unidimensional construct. Scalability coefficients—also known as Loevinger’s coefficients (Hi)—are computed for each item, with values above 0.30 typically considered satisfactory (Van Schuur, 2003; Zhang and Li, 2020). These coefficients indicate the degree to which the items on a scale consistently order respondents based on the level of the latent trait being measured.

According to the findings, 534 (78.2%), 508 (74.4%) and 449 (65.7%) respondents correctly answered Items 1, 3 and 2 regarding the need for vaccinations, the efficacy of vaccinations and the importance of smallpox involvement, respectively. However, the highest percentage of incorrect responses to questions concerning vaccine components was for Item 9 (33.4%, n = 228). This was followed by the question concerning the immune system from Item 6 (25.9%, n = 177) and the possibility of allergies from Item 8 (33.2%, n = 227). Item 6 concerning the immune system (46.1%, n = 315) and Item 9 regarding the occurrence of allergies (41.9%, n = 286) were the ones for which the “do not know” option was the most frequently selected.

In MSA, the unidimensionality tests were performed three times to eliminate irrelevant vaccination knowledge items. Item 2, “smallpox” (Hi = 0.274, standard error (SE) = 0.028) and item 9 “allergy occurrence” (Hi = 0.272, SE = 0.027), were excluded from the first-time test, while item 8 “immunisation components” (Hi = 0.289, SE = 0.027) were also excluded from the second-time re-test. Therefore, a revised six-item scale of vaccination knowledge was developed, and it showed a reasonable average scaling coefficient (Hia = 0.348, SE = 0.023); the Hi of individual items ranged between 0.320 and 0.368. The hypothesis H1 was supported.

Furthermore, the Molenaar Sijtsma statistic finding (ρ = 0.701) provided an appropriate measure of single-administration reliability for the vaccination knowledge scale, which allowed it to be used as an unbiased estimate of test-score reliability (Van der Ark et al. 2011). This number is higher than the minimal acceptable level observed in prior research (ρ = 0.70) (Sijtsma, 2009). The distribution of responses to the nine questions on the MSA-based vaccination knowledge scale is shown in Fig. 2.

Fig. 2
figure 2

Distribution of vaccination knowledge among Chinese adults based on the Mokken scale analysis in 2022.

Full size image

eHealth literacy, vaccination knowledge and willingness

RQ2 addressed the degree of association between eHealth literacy, vaccination knowledge and willingness to vaccinate among Chinese individuals. A correlation matrix with Pearson’s coefficient (r) was employed to identify potential relationships among the primary variables. The findings demonstrated substantial positive correlations between eHealth literacy and vaccination knowledge (r = 0.117, p = 0.002) and between eHealth literacy and vaccination willingness (r = 0.261, p < 0.001), supporting H2 and H3, respectively. In addition, vaccination knowledge and willingness were also significantly positively linked (r = 0.247, p < 0.001). Self-reported health status changed simultaneously with eHealth literacy (r = 0.202, p < 0.001), vaccination knowledge (r = 0.150, p < 0.001) and vaccination willingness (r = 0.147, p < 0.001), compared to other covariates. Table 1 shows the zero-order correlation analysis of eHealth literacy, vaccination knowledge, willingness to vaccinate and subjective health status.

Table 1 Zero-order correlation analysis of each pairwise primary variable.
Full size table

Mediating role of vaccination knowledge

RQ3 indicated that the mediating role of vaccination knowledge in the relationship between eHealth literacy and willingness to vaccinate was moderated by health status. The unstandardised coefficient (B) was chosen to represent the influence of each independent variable on the outcome using multiple linear regression. Specifically, eHealth literacy significantly affected vaccination knowledge (B = 0.060, SE = 0.023, t = 2.641, p = 0.008). Vaccination knowledge was positively associated with vaccination willingness (B = 0.061, SE = 0.011, t = 5.449, p < 0.001). After controlling for the influence of vaccination knowledge, eHealth literacy had a positive direct effect on vaccination willingness (B = 0.034, SE = 0.007, t = 5.138, p < 0.001). Table 2 shows the multiple linear regression analysis of eHealth literacy, vaccination knowledge and vaccination willingness, including the potential confounders of the relationship between eHealth literacy and COVID-19 vaccination willingness.

Table 2 Multiple linear regression analysis of the relationship among eHealth literacy, vaccination knowledge and vaccination willingness.
Full size table

Model-based causal mediation analysis revealed an average causal mediation effect (ACME) of 0.004, with a 95% CI of [0.001, 0.010], p = 0.008. This analysis identified vaccination knowledge as a mediating variable between eHealth literacy and vaccination willingness. Thus, H4 was supported. In addition, eHealth literacy demonstrated a significant overall effect on willingness, independent of vaccination knowledge, with a B-value of 0.037 and 95% CI of [0.024, 0.050], p < 0.001 (see Table 2). Based on this, the proportion mediated was calculated to be 0.098 (9.8%), with a 95% CI of [0.029, 0.210], p = 0.008, indicating statistical significance for the ratio of the indirect effect through vaccination knowledge to the overall effect.

Moreover, the relationship between eHealth literacy and vaccination knowledge was significantly moderated by health status (B = −0.057, SE = 0.020, t = −2.798, p = 0.005). The low (mean minus 1 SD = 2.30), medium (mean = 3.20) and high (mean plus 1 SD = 4.10) health status groups served as post hoc tests to determine the magnitude of the mediation effects through vaccination knowledge, which varied by health status. The findings revealed that the indirect effects were not significant for individuals with high health status (ACME = 0.001, 95% CI [−0.002, 0.000], p = 0.59), but were significant for those with low (ACME = 0.007, 95% CI [0.003, 0.010], p < 0.001) and medium-level (ACME = 0.004, 95% CI [0.002, 0.010], p = 0.002) health status. Furthermore, a positive difference in mediation effects (ACMEd) between groups with low and high health statuses was observed (ACMEd = 0.006, 95% CI [0.001, 0.012], p = 0.032). Thus, H5 was supported.

DAGs depict the potency and orientation of the suggested causal route, in which eHealth literacy impacts vaccination knowledge, subsequently affecting vaccination willingness. It was suggested that subjective health status was posited to moderate the relationship between eHealth literacy and vaccination knowledge. The generated DAGs reflect the intensity and direction of the postulated causal links, showing how eHealth literacy influences vaccination knowledge and subsequently affects vaccination willingness. Furthermore, the impact of health status on the connection between eHealth literacy and vaccination knowledge underscores the importance of considering different health conditions in understanding these cause-and-effect dynamics. Thus, considering different levels of health conditions, Fig. 3 illustrates the relationship between eHealth literacy, vaccination knowledge and willingness to be vaccinated.

Fig. 3
figure 3

Mediating effect of vaccination knowledge on the relationship between eHealth literacy and vaccination willingness, varying across health status.

Full size image

Discussion

Principal findings and comparison with prior work

The high prevalence of incorrect responses observed in questions related to various aspects of vaccination, allergies and the immune system underscores the imperative need for targeted educational initiatives and awareness campaigns aimed at enhancing public understanding of these critical areas (Dror et al. 2020; Lazarus et al. 2022). These findings indicate a lack of self-assessment among the Chinese respondents regarding the fundamental aspects of vaccination, suggesting potential gaps in health literacy and awareness, which is consistent with prior studies conducted in China (e.g., Wang et al. 2021; Zhang et al. 2022). One possible explanation is that these people think they know something about vaccines without realising that they actually do not know it, leading to misunderstandings (Browne et al. 2015).

Notably, the frequent selection of the “do not know” option for questions pertaining to the immune system, allergy occurrence and timing of vaccine administration underscores areas where Chinese individuals may feel uncertain or uninformed, emphasising the necessity for clearer communication and dissemination of information regarding these topics. Such uncertainties or lack of knowledge may contribute to vaccine hesitancy or misinformation, ultimately hindering vaccination efforts (Leng et al. 2021; Mathieu et al. 2021). The likely reason for this is that although the majority of the Chinese public is supportive of vaccination, they do not have sufficient knowledge about the specifics and scientific principles of vaccines (Wang et al. 2024), and there is more room for development of pathways ways to popularise expertise about vaccines.

The study findings also highlight the significant role of eHealth literacy in promoting vaccination willingness among the Chinese population, which is consistent with previous studies (e.g., Britt et al. 2017; Chang and Schulz, 2018; Qin et al. 2021). With the positive correlation observed between eHealth literacy and willingness to vaccinate, it is evident that eHealth literacy can serve as a crucial strategy in addressing public health crises in China. By enabling Chinese individuals to access, comprehend and discern relevant health information online, eHealth literacy enhances public knowledge about vaccines, comprehension of their benefits, and awareness of the potential consequences of vaccine refusal or delay (Britt et al. 2017; Lwin et al. 2020). Furthermore, eHealth literacy enhances Chinese individuals’ ability to make scientifically informed judgments regarding the risks associated with vaccination, thereby fostering greater confidence and willingness to vaccinate (Chang and Schulz, 2018; Robinson et al. 2022).

Moreover, this study provides valuable insights into the mechanisms underlying the relationship between eHealth literacy, vaccination knowledge and willingness to be immunised. While model-based causal mediation analysis uncovered the mediation effect of vaccination knowledge between eHealth literacy and vaccination willingness, a graphical tool for analysing casual diagrams offer a powerful visual tool to elucidate this causal pathway. Vaccination knowledge emerged as a significant mediator in the relationship between eHealth literacy and vaccination willingness, which indirectly supports the findings of previous studies on the pathway from eHealth literacy to vaccination knowledge (e.g., Qin et al. 2021) and from vaccination knowledge to willingness (e.g., Zheng et al. 2022). This suggests that Chinese individuals with higher eHealth literacy levels tend to possess greater vaccination knowledge, which, in turn, positively influences their willingness to receive vaccines. These findings underscore the importance of enhancing accurate vaccine-related information and understanding as a means of promoting vaccination willingness, particularly among Chinese populations with varying levels of eHealth literacy (Fakonti et al. 2022; Schulz and Hartung, 2021).

Additionally, this study elucidated the moderating role of health status on the relationship between eHealth literacy and vaccination willingness through vaccination knowledge. In particular, Chinese individuals with low health status demonstrated a significantly stronger association between eHealth literacy and vaccination knowledge, indicating that interventions aimed at improving eHealth literacy may have a more pronounced empowerment on vaccination awareness and intentions within this vulnerable population (Jiao et al. 2023; Lu et al. 2024). Conversely, Chinese individuals with higher health status exhibited a less pronounced association between eHealth literacy and vaccination knowledge, suggesting a potential saturation effect, wherein individuals with higher health status may already possess a sufficient level of vaccine-related knowledge. This is an important breakthrough discovery that previous research has not shown. One possible reason for this is that those who are successfully self-managed their health and are in better health are themselves sufficiently knowledgeable about special diseases and the health outcomes that follow (Eyuboglu and Schulz, 2016; Lu et al. 2022; Marciano et al. 2019).

Contribution to the body of knowledge

This study’s findings contribute novel insights into the current body of knowledge regarding eHealth literacy, vaccination knowledge and vaccination willingness, specifically about cultural influences, small effect sizes in statistics, and the hierarchical organisation of vaccination knowledge.

Impact of cultural factors on eHealth literacy and vaccination attitudes

As the global community navigates the evolving landscape of healthcare, the role of eHealth literacy has become increasingly crucial. Vaccination, a fundamental public health intervention, has also garnered significant attention, with vaccination attitudes playing a pivotal role in shaping individual and community-level decisions. While numerous studies have examined the relationship between eHealth literacy and vaccination attitudes, a deeper analysis of the cultural factors that may influence these dynamics is warranted. One key cultural factor that can shape eHealth literacy and vaccination attitudes is the prevalent belief systems and worldviews within a given community (Cheng et al. 2020; Pavić et al. 2023).

For instance, some communities may hold strong beliefs in traditional or alternative forms of medicine, which can inform their perceptions of the efficacy and necessity of vaccination (Hornsey et al. 2020). Similarly, cultural norms and values surrounding the role of individual autonomy, community cohesion and trust in public health authorities can also significantly influence eHealth literacy and vaccination attitudes. Additionally, socioeconomic status, access to healthcare resources, and educational attainment can all have an impact on a population’s level of health literacy, which in turn can affect attitudes towards vaccinations and eHealth literacy. Furthermore, the growing reliance on digital platforms for health information and vaccine-related decision-making can magnify the impact of cultural factors on eHealth literacy and vaccination attitudes. Digital divides, whereby certain segments of the population have limited access to or familiarity with digital technologies, can exacerbate the impact of cultural factors on eHealth literacy and vaccination attitudes.

This study demonstrates a positive relationship between the intention of Chinese individuals to receive vaccinations and their level of eHealth literacy. It highlights the significant role of eHealth literacy in shaping people’s attitudes towards vaccination. As technology advances, eHealth information is becoming increasingly popular and convenient in China. This allows Chinese individuals to obtain and comprehend health information more effectively, enabling them to make better-informed decisions on vaccinations. Chinese culture is predominantly collectivist, prioritising social harmony and collective well-being, while traditional Chinese culture prioritises “prevention first,” aligning with the contemporary notion of immunisation (e.g., Huigang et al. 2020). This cultural context may foster a greater inclination among Chinese individuals to receive COVID-19 vaccinations, as a means to safeguard the overall health of the community (Chen et al. 2021; Liu et al. 2023).

Insights to understand small effect size

When conducting causal mediation analysis, researchers often encounter scenarios where the estimated average causal mediation effect (ACME) is statistically significant but represents a relatively small effect size (e.g., ACME = 0.004, p = 0.008). In such cases, it is important to recognise that the practical significance of the findings should not be dismissed solely based on the magnitude of the effect. There are several reasons why a small but statistically significant mediation effect can still be considered acceptable and worthy of further investigation.

First, even small effects can have important practical implications, especially in the context of public health and policy interventions (Chen et al. 2021; Liu et al. 2023). The impact of a small effect can accumulate and lead to meaningful changes at the population level. Additionally, a small mediated effect may be indicative of a complex causal process where multiple mediators work together to influence the outcome.

Second, the interpretation of effect size in mediation analysis should be considered within the specific research context and the theoretical framework guiding the study. The traditional focus on large effect sizes may not always be appropriate, as the goal of mediation analysis is to understand the underlying mechanisms rather than simply maximise the effect size. Researchers should consider the theoretical and practical significance of the findings rather than relying solely on arbitrary effect size thresholds (Nguyen et al. 2021).

Moreover, the reporting of precise effect sizes, along with their associated confidence intervals, is crucial in mediation analysis. This allows readers to evaluate the magnitude of the effect and make informed judgments about its practical relevance (Chang and Jiao, 2020; Rizzo et al. 2022). It is also important to consider the context in which the study is conducted and the specific research question being addressed. In some cases, a small but statistically significant mediated effect may be meaningful and valuable, particularly when the outcome is difficult to modify or when the intervention is targeting a complex, multifaceted issue (Rijnhart et al. 2021; Wang et al. 2022; Yin et al. 2022).

In summary, while a small effect size may initially seem less impactful, it does not necessarily mean that the findings are unimportant or unworthy of further investigation. Researchers should carefully consider the theoretical and practical significance of their results, the context of the study and the potential for meaningful impact, rather than dismissing small but statistically significant causal mediation effects.

Hierarchical organisation of vaccination knowledge

Implementing a hierarchical organisation of vaccine information can provide substantial advantages in educational interventions and public health campaigns. Two supplementary ideas that could improve this strategy involve utilising technology and customising messaging for distinct sociodemographic groups. Integrating technology into hierarchical vaccination knowledge frameworks has the potential to enhance their effectiveness (Honora et al. 2022). Specialised mobile applications, for example, could offer users a single platform to efficiently acquire and explore organised vaccine-related information. These digital solutions have the potential to provide individualised assistance, reminders for appointments and access to reliable resources, enabling individuals to make well-informed decisions regarding vaccination.

Customising educational campaigns and public health messages based on sociodemographic variables is essential for optimising their impact (Hong, 2023; Omer et al. 2021). It is crucial to identify and address the particular beliefs, attitudes and obstacles that affect vaccination rates in diverse populations to execute more focused and effective treatments. By comprehending the distinct requirements and viewpoints of diverse communities, educators and public health experts can create messages that have a stronger impact and specifically target the individual issues of each group.

Practical implications

This study has revealed some practical implications that could be relevant to other communities and global health policies, based on a survey conducted in China. This study posits that the large population of China may result in global audiences having restricted awareness and comprehension of China’s vaccine immunisation system, allergies and the timing of vaccines. Hence, the conclusions elucidated in the previous sections have substantial implications for worldwide health policies that transcend the Chinese populace. The identified trends in the acceptance of vaccination and levels of health literacy can be applied to more groups, particularly in low- and middle-income countries.

The issue of vaccine reluctance and insufficient health literacy is not limited to China but rather reflects a wider worldwide problem. Research has indicated that the willingness to receive vaccines is typically greater in low- and middle-income countries in comparison to high-income nations (Hong, 2023; Omer et al. 2021; Solís Arce et al. 2021). These findings indicate that focusing on distributing vaccines to countries in the Global South could result in significant progress in achieving global vaccination coverage. Simultaneously, it is crucial to tackle the obstacles posed by vaccination reluctance and inadequate health literacy by implementing focused interventions.

Addressing knowledge disparities and enhancing health literacy are essential measures in transforming attitudes towards vaccination (e.g., Honora et al. 2022; Wollburg et al. 2023). Radio broadcasts have been recognised as an effective means of addressing specific audiences with educational initiatives, according to Wollburg and others in 2023. Moreover, harnessing the confidence bestowed upon healthcare professionals might serve as a potent tactic to foster the acceptance and use of vaccines (Solís Arce et al. 2021). Tackling these variables can have extensive consequences for the effectiveness of vaccination campaigns and the broader world health situation.

Limitations and future directions

This study has several limitations that need to be addressed. In the future, it is recommended that simple random sampling be implemented as a straightforward approach to obtaining a viable sample group. Assigning numbers to each group member or sample and subsequently randomly selecting those numbers through an automated process to determine who will be included in the sample are the simplest methods of simple random sampling. By ensuring that each adult member of the population has an equal chance of being selected as a participant in the study, simple random sampling significantly reduces the likelihood of sampling bias.

Another approach is quota sampling for the Chinese to provide researchers with representative results that facilitate the development of robust data-driven decisions. This technique is a nonprobability sampling approach in which researchers establish a sample of adult individuals who are representative of a particular population. By employing quota sampling, we can guarantee that the results of our survey closely resemble the demographic of our target audience. Quota sampling is a highly effective method for obtaining actionable data and insights from a specific audience and in a populated country, although it entails a similar risk-reward as nonprobability sampling.

In addition, this study focused solely on assessing general vaccination knowledge within the Chinese context and did not explore the participants’ knowledge specifically regarding the COVID-19 vaccine. As the landscape of vaccination evolves, future research should consider incorporating more comprehensive and tailored measures to assess knowledge related to emerging vaccines. Furthermore, certain confounding variables that may have influenced the observed associations were not considered. For instance, reasons for vaccine hesitancy, trust in the healthcare system and previous vaccination experience were not accounted for despite their potential impact on eHealth literacy and vaccination willingness. Future investigations should comprehensively explore these factors to provide a more nuanced understanding of attitudes towards vaccination.

Conclusions

Following the COVID-19 pandemic, various factors associated with the pandemic experience and aftermath have influenced Chinese people’s decisions regarding widespread vaccination. The following two points demonstrate how China’s vaccination decisions have changed in the post-COVID-19 era. First, the COVID-19 pandemic has highlighted the importance of immunisation in preventing the spread of infectious diseases. Consequently, the Chinese public now understands the importance of immunisation against COVID-19 and other communicable diseases. Owing to the pandemic’s stark reminder of the potential consequences of diseases that can be treated with vaccines, a greater number of individuals are prioritising vaccination as a preventive measure. Second, the population’s trust and faith in vaccinations has grown because of the COVID-19 vaccine’s successful development and introduction in China. People have become more likely to trust the efficiency and safety of vaccines in general after observing how vaccinations helped contain the spread of COVID-19 and reduce its severity. Beyond COVID-19, routine vaccinations may become more widely accepted and adopted because of this growing confidence in vaccines.

This study underscores the critical importance of targeted educational interventions, enhanced communication strategies and tailored health-promotion campaigns to address gaps in vaccination knowledge, promote eHealth literacy and foster greater willingness to vaccinate among diverse populations in China. By leveraging the insights provided by this study, government organisations, healthcare providers, and public health authorities can design more effective interventions and communication strategies to promote vaccination acceptance and mitigate vaccine hesitancy, ultimately contributing to improved public health outcomes and disease prevention efforts.

In addition to exploring the association between eHealth literacy and vaccination knowledge in a sample of the general public, this project sought to design and verify a new scale for measuring general vaccination knowledge. The findings of this study showed that health status had a substantial impact on the association between eHealth literacy and vaccination knowledge. The results of the moderation study showed a substantial interaction effect, indicating that a person’s health status influences how strongly their eHealth literacy and vaccination knowledge are related. By improving our knowledge of the variables influencing vaccine-related decision-making, this study has implications for advancing public health research as well as practice.