Preplanned Studies: Occurrence and Reduction of Hepatitis B Vaccine Hesitancy Among Medical University Students — Shanxi Province, China, 2020

Hepatitis B virus (HBV) infection is a significant global public health concern (1). According to the World Health Organization (WHO), in 2019, an estimated 296 million individuals worldwide had chronic hepatitis B, with 1.5 million new cases reported annually and approximately 820,000 deaths (2). In China, universal hepatitis B vaccination for newborns was introduced in 1992, resulting in a high vaccination rate among university students. However, the antibody response to HBsAg (anti-HBs) decreases over time after vaccination (3). Additionally, university students are at a higher risk of HBV infection due to various factors such as tattoos, unprotected sexual contact, pierced ears, and accidental blood exposure to medical students. The WHO recommends booster vaccination for high-risk populations (4). However, there is a growing reluctance among individuals to receive recommended vaccinations (5). Despite this, there have been limited efforts to evaluate hepatitis B vaccine hesitancy among medical university students. This study was conducted among the class of 2017 at a medical university from April to June 2020, with a total of 1,003 students from different professions completing a questionnaire. The results illustrated that hepatitis B-related knowledge, attitudes, and practices had a negative impact on hepatitis B vaccine hesitancy. The findings from this study provide insights for developing a health education-based hepatitis B vaccination strategy.

A cross-sectional study was conducted in the class of 2017. Out of the 27 professions, nine were randomly selected for recruitment, and all students in those nine professions were included in the study. Prior to enrollment, written informed consent was obtained from all participants. A self-designed questionnaire, consisting of 15 questions on HBV knowledge, 15 questions on HBV attitudes, and 10 questions on practices (Supplementary Table S1), was distributed through the Wenjuanxing platform. The questionnaire was administered on-site by trained investigators. Students were asked to independently complete the questionnaire. The internal consistency of the questionnaire, measured by Cronbach’s α, was determined to be 0.855, indicating good reliability.

Hepatitis B knowledge was categorized into four dimensions based on their relevance to infectious disease epidemics among humans: etiological knowledge (EK), transmission route knowledge (TRK), prevention measures knowledge (PMK), and epidemic status knowledge (ESK). Each correct answer was scored 1 point, while incorrect or unclear answers received 0 points. Hepatitis B attitudes were assessed using a 5-point Likert scale, ranging from “very consistent” to “very inconsistent,” with scores of 5 to 1, respectively, indicating personal inclination. Questions 3, 4, 5, 12, 13, and 14 were reverse-scored. Hepatitis B practices were divided into two dimensions based on behavioral characteristics: health-seeking practices (HSP) and high-risk practices (HRP). A score of 1 point was given for “yes” responses to hepatitis B practice items, while “no” or “unknown” responses received 0 points. Knowledge, attitudes, and practices related to hepatitis B were classified into two groups based on the mean score.

In our study, “vaccine hesitancy” (VH) was defined as the participants’ response of “unsure”, “basically inconsistent”, or “very inconsistent” when asked about their willingness to be vaccinated against hepatitis B in the absence of detectable anti-HBs. This term was used to describe the hesitation towards receiving the hepatitis B vaccine.

The database was established using Microsoft Excel (version 2016; Microsoft Corp, Redmond, USA), and analyses were conducted using SAS (version 9.4; SAS Institute, Cary, NC, USA). The chi-square test was used to assess the difference between the groups with vaccine hesitancy and those without. Logistic regression analysis was employed to examine the associations between the independent variables and hepatitis B vaccine hesitancy. Independent variables with P<0.05 in the univariate analysis were included in the logistic regression model. Structural equation modeling (SEM) was performed using Amos software (web version 24.0; IBM, New York, USA) to elucidate the interrelationships between latent variables. The model parameters were estimated using maximum likelihood estimation.

A total of 1,012 students were contacted, and 1,003 students responded, resulting in a response rate of 99%. The respondents were predominantly female (75.37%), with a sex ratio of 0.33∶1. Among the respondents, the majority (n=961) were of Han ethnicity. The participants consisted of 747 medical professionals (74.48%) and 256 nonmedical professionals (25.52%) (Table 1).

The Kaiser-Meyer-Olkin coefficient was found to be 0.785, indicating satisfactory sampling adequacy. Additionally, Bartlett’s test showed statistical significance (P<0.001), supporting the factor analysis. Based on eigenvalues, a total of four factors were extracted, explaining 69.54% of the cumulative variance. These factors were identified as attitude toward the source of infection (ASI), self-protection attitude (SPA), life contact attitude (LCA), and vaccine effect attitude (VEA).

A total of 199 respondents expressed hesitancy towards the hepatitis B vaccine, resulting in a vaccine hesitation rate of 19.84% (199/1,003). In the univariate analysis, significant associations were observed between vaccine hesitancy and factors such as knowledge, attitudes, and health-seeking practices (P<0.05) (Table 1).

The logistic regression model revealed that university students with lower TRK and PMK scores had a higher likelihood of hesitating to receive the hepatitis B vaccine (OR=1.947, 95% CI: 1.309, 2.897; OR=2.501, 95% CI: 1.682, 3.719). Participants who held stigmatizing attitudes toward the source of infection were 1.854 times more likely to hesitate in getting the hepatitis B vaccine compared to those with more inclusive attitudes (95% CI: 1.280, 2.684). Vaccine hesitancy was also associated with concerns about vaccine efficacy (OR=6.529, 95% CI: 4.417, 9.650). Furthermore, individuals who adopted fewer health-seeking behaviors were more likely to hesitate to receive the vaccine and had a 2.119-fold increased likelihood of adopting such behaviors (95% CI: 1.357, 3.308) (Table 2).

SEM was developed using the knowledge, attitude, and practice theory model as a basis. The initial model included knowledge-related latent variables as independent variables, attitude and behavior-related latent variables as mediators, and the observed variable VH as the dependent variable. Non-significant paths between latent variables in the hypothetical model (P>0.05) were removed and corrected using the correction index. Model fit measurements indicated that the revised model demonstrated a satisfactory fit, and the model was professionally interpreted. The reference standards for the fit indicators, as well as the fit indicators for the final model, are provided in Supplementary Table S2. A diagram of the standardized model can be seen in Figure 1. The values assigned to each variable in the revised model are presented in Supplementary Table S3.

Figure 1. 

Revised model (SEM).

Note: The ellipse and arrow in the structural model represent the effects of factors. One-way straight arrows in the structural model depict the paths of each factor with standardized regression weights, and one-way straight arrows in the measurement model show the measurement loadings of each item in the scale.

Abbreviation: PMK=prevention measures knowledge; ASI=attitude toward the source of infection; PMK=prevention measures knowledge; VH=vaccine hesitancy; LCA=life contact attitude; TRK=transmission route knowledge; HSP=health-seeking practices.

Supplementary Table S4 presents the associations between different factors (TRK, PMK, ASI, HSP, LCA) and vaccine hesitancy for hepatitis B. TRK had an indirect effect on vaccine hesitancy [standard indirect effect (SIE)=−0.301]. PMK, on the other hand, influenced vaccine hesitancy through both direct and indirect effects (standard direct effect (SDE)=−0.168, SIE=−0.101). ASI (SDE=−0.267) and HSP (SDE=−0.204) only had direct effects on hepatitis B vaccine hesitancy. Additionally, LCA had an indirect effect on vaccine hesitancy (SIE=−0.066), and this relationship was partially mediated by ASI. The analysis also revealed a negative correlation between hepatitis B-related knowledge, attitudes, and practices, and vaccine hesitancy. Further details regarding the structural model parameters can be found in Supplementary Table S5. These findings indicate the importance of health education that focuses on enhancing knowledge, attitudes, and practices related to hepatitis B, as part of an effective hepatitis B vaccination strategy.