Study on Status and Willingness towards Hepatitis B Vaccination among Migrant Workers in Chongqing, China: A Cross-Sectional Study

Background: Rural-to-urban migrant workers may serve as a bridge population for the cross-regional spread of hepatitis B vaccination (HBV) due to frequent shifts between their work areas and homelands, and they are less likely to be covered by the national hepatitis B (HB) immunization program. This study aimed to investigate the current inoculation status of HB vaccine among migrant workers and the willingness to be vaccinated among non-vaccinated ones. Methods: We conducted a cross-sectional survey using anonymous interviews with migrant workers selected by two-stage cluster sampling from July to December 2018. Binary logistic regression models were adopted to detect influencing factors associated with HB inoculation status and vaccination willingness. Results: 1574 respondents were recruited in the surveys, and 773 (49.11%) respondents reported that they had been inoculated with HB vaccine. Only 285 (35.58%) non-vaccinated respondents were willing to be inoculated. Logistic regression indicated that younger age, higher education level, less wearing of condoms, higher knowledge scores of HB, and higher risk perception of HBV infection were positively associated with inoculation of HB vaccine. Respondents who were more highly educated, and drinkers, with higher knowledge scores of HB and with higher risk perception of HBV infection were more willing to be vaccinated. Conclusions: the HB vaccination rate of migrant workers in Chongqing was relatively low and only a small section of non-vaccinated migrant workers had vaccination willingness. Health interventions and policies are needed to improve knowledge and cognition of HB among migrant workers, particularly for those who are older, less educated, poor in HB knowledge, less likely to wear condoms, and non-drinkers. Peer education, as well as the combination of traditional and new media, would be accessible and effective ways to disseminate HB related knowledge for migrant workers.


Introduction
Hepatitis B (HB) remains a challenge to public health, causing 500,000 to 750,000 deaths per year due to cirrhosis and liver cancer evolving from HB worldwide [1]. Currently, World Health

Study Sites and Participants
The present study was performed in Chongqing, the largest municipality directly under the Chinese central government. Chongqing, located in southwestern China, is referred to as a "miniature of China" because its geographic characteristics, social-economic profile, and urban-rural distribution are close to the national average [20]. The city area of Chongqing, one of the busiest regions for the inflow of China's migrant population, consists of nine administrative districts with 8.65 million residents, among which immigrants account for 23.5% [21]. In 2007, the estimated HBsAg positive rate was 8.6% among migrant workers in the city area of Chongqing [22]. According to the Health Statistic Yearbook of Chongqing, there were nearly 26,000 new infections of viral hepatitis in 2016 [23].
The target subjects in the study were migrant workers who (1) were 18 years and above, (2) had been in the city area of Chongqing for at least six months, (3) had not registered as Chongqing urban resident, (4) were engaged mainly in the secondary or tertiary industry, such as construction industry, manufacturing industry, wholesale and retail industry, transportation industry, hotel and catering industry, community services. Individuals who were not able to understand the questionnaire items or refused to be surveyed were excluded.

Sampling Methods
Field surveys were carried out from July to December 2018. Participants were selected by two-stage stratified cluster sampling-first, nine districts of the city area in Chongqing were categorized into three stratifications (very developed, medium developed, and less developed) according to economic background, geographic location and population density, and three districts were then randomly selected from each stratification respectively. Second, two or more enterprise units were purposefully selected within each district, including the manufacturing, construction, wholesale and retail industry, the transportation industry, the hotel and catering industry, community services, with the assistance of local Center for Disease Control and Prevention, Health supervision Institute, and Urban-rural Development Committee. All migrant workers meeting the inclusion criteria were sampled from each enterprise [13].

Study Instruments
The questionnaire was constructed based on the WHO fact sheet on HB immunization and the questionnaires adopted in the published studies on knowledge related to HB epidemics and prevention [24][25][26]. A pilot test was performed with 100 migrant workers in restaurants near to Chongqing Medical University (CQMU) and the framework and wordings of the questionnaire were modified for better understanding of target subjects. The final version consisted of four modules including socio-demographics (16 items), status, and willingness of vaccination (three items), knowledge of HB (14 items), perceived and behavioral risks (eight items). Detailed items and scales of the questionnaire can be found in the Supplementary Materials.

Statistical Analysis
Survey data were double-checked and entered by Epidata 3.1 (The EpiData Association, Odense, Denmark). All data were analyzed using SAS software, version 9.4 (SAS Institute, Cary, North Carolina, USA). Categorical variables were described by number and percentage. Continuous variables were described by the mean and standard deviation (normally distributed variables) or the median and interquartile range (abnormally distributed). Then, univariate analyses were performed using the Chi-square test or Wilcoxon rank sum test to assess the HB vaccination status and willingness of the respondents with variables in the modules of social demographics, knowledge of hepatitis B, and perceived and behavioral risk. Variables with P-values less than 0.10 were subsequently inputted into the binary logistic regression model to test the possible influencing factors for two outcomes-(1) HB vaccination status and (2) vaccinating willingness. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) of variables were computed and the test level was α = 0.05, β = 0.10.

Ethics Statement
The present study was approved by the Institutional Review Board of CQMU (2018016). Participants were reassured that all responses would be anonymous and written informed consent was secured from each respondent. Peak working hours were avoided to ensure the quality of interviews.

Results
A total of 1740 participants met the eligible criteria and 1574 (90.46%) repondents completed this survey. The median age was 32.06 ± 21.18 years, ranging from 18 to 68 while 1487 (94.47%) of respondents were ethnic Han and 87 (5.53%) respondents were ethnic minority. A number of 223 (14.17%) respondents had an education background of primary school or below, and 579 (36.79%) respondents had an education background of junior middle school. 561 (35.64%) respondents had a personal monthly income between 2500 and 4000 RMB, while 364 (23.13%) respondents earned less than 2500 RMB a month (Table 1).
A number of 801 (50.89%) respondents reported that they had not inoculated HB vaccine, among whom there were 516 (64.42%) unwilling to vaccinate. The 745 (42.01%) non-vaccinated respondents were mainly aged between 18 and 30, among whom there were only 285 (35.58%) willing to be vaccinated in the future ( Table 2). To explore the reason for not having HB vaccination, 206 (40%) respondents said that they had never heard of HB vaccination, 160 (31.11%) respondents did not know where to vaccinate, 84 (16.30%) respondents doubted the effectiveness or safety of HB vaccine, 55 (10.74%) respondents thought vaccination was inconvenient or inaccessible, and 55 (10.74%) respondents had followed their friends or family members in not vaccinating ( Table 3).
The HB knowledge score was 5.00 ± 6.00 and 7.00 ± 3.00 among respondents in non-vaccinated and vaccinated respondents respectively ( Table 1). The most popular access for the respondents to obtain HB knowledge and information is hearing from friends or family members (53.88%), followed by television or radio (38.12%), Internet or mobile phone (27.95%), and newspapers or magazine (19.70%) respectively (Table 4).
A number of 678 (43.07%) respondents thought that it was absolutely impossible/impossible for them to get HB infection. In the last six months, 184 (11.69%) respondents had had casual sex, 92 (5.84%) of them had commercial sex, and 52 (3.30%) respondents had homosexual behaviors or anal sex activities. There were 688 (43.71%) respondents who had had sex with a condom less than five times in the past half a year. In addition, 17 (1.08%) respondents had intravenous drug use by sharing injectors, 23 (1.46%) respondents had experienced illegal blood selling or transfusion, and 298 (18.93%) respondents had shared toothbrush or towels with others in the last six months ( Table 1).
As Table 2 shows, univariate analyses using χ 2 test or Wilcoxon sum rank test indicated that the willingness of HB vaccination for non-vaccinated respondents was significantly different with variables of gender, ages, ethnicity, education background, years of being a migrant worker, working hours per day, sending money to family, knowledge score of HB, perceived risk of HB, and condom use (P < 0.05). Binary logistic regression showed that migrant workers with higher education level (OR = 2.01, 95% CI: 1.24-3.28; OR = 2.21, 95% CI: 1.26-3.89), drinking (OR = 1.84, 95% CI: 1.23-2.76), with better knowledge scores (OR = 1.13, 95% CI: 1.08-1.19) and higher risk perceptions of HB (OR = 2.18, 95% CI:1.57-3.03) were more willing to be inoculated. Respondents of ethnic minority were unwilling to vaccinate compared with those of ethnic Han (OR = 0.46, 95% CI: 0.22-0.95).

Discussion
According to the political and economic context, HB vaccination rates vary across different locations in China. In the present study, 49.11% of migrant workers in the city area of Chongqing had been inoculated with HB vaccine, which was higher than those of migrant workers in Beijing (37.05%), Hebei province (23.58%), Heilongjiang province (21.05%), Jiangsu province (21.75%), Ningxia province (32.81%), and Hainan province (17.09%), but lower than Shenzhen (70.21%) [9,18]. In addition, the amount of non-vaccinated migrant workers, as much as 51.89% of respondents, was very much close to the vaccination rate (51.60%) reported by Gong et al. in 2006 [22], and 42.01% of non-vaccination respondents were mainly aged between 18 and 30 in our study. This suggested that in the past decades the implementation of the HB immunization program had been inefficient in the countryside of Chongqing and surrounding area, from where the migrant workers mainly flow into city area of Chongqing. Therefore, more specific and stronger policies and regulations are needed to enhance HB immunization in rural areas where the migrant workers mostly come from.
A negative association was detected between HB vaccination status and age, which is consistent with the study of Liu et al. [9]. On the one hand, aged respondents in our study might be beyond the required age for free HB immunization by the MoH. On the other hand, aged respondents might take more care of the health of children than themselves [6,9]. In accordance with the study of Yan et al., our findings showed that the more highly educated respondents had a higher vaccination rate and were more willing to be vaccinated if they had not been vaccinated before [27]. Evidence showed that good educational background leads to good cognition of disease prevention among adults, of which the vaccination status of migrant workers differs over educational deviations [28]. In addition, health education campaigns, carried out by staff associations and health service organizations, are also needed to improve migrant workers' attitude both at the workplace and in the community where they live.
HB knowledge scores were shown to be positively associated with both vaccination status and willingness of respondents, because good understanding of HB related knowledge probably results in concern of HBV infection and awareness of self-protection towards HB [9]. With regard to the ways of obtaining knowledge, friends or family members were the favorite source for migrant workers, which suggests that health interventions based on peer education might work to a large extent. Moreover, TV, newspapers, and magazines were still popular among respondents, and Internet/mobile phones also played an important role in information updates for migrant workers, which indicates that health education could be promoted by a combination of traditional and new media.
Consistent with the study by Xie et al., Ethnic Han was detected to be positively associated with willingness to vaccinate compared with the ethnic minorities [29]. Religious faith and briefs have been mentioned as influencing factors of immunization behaviors in previous studies [30], which could be an interpretation for the significantl difference in HB vaccination between migrant workers of ethnic Han and other minorities from nearby provinces like Yunnan or Guizhou.
Consistent with previous study, the perceived risk of HBV infection was positively associated with HB vaccination [9]. It makes sense that the more susceptible respondents perceive the risk of infection with HBV the more active they are to vaccinate [31]. Less condom use in the last six months was negatively associated with HB vaccination. A possible reason was that the vaccinated respondents actually dare to have unprotected sex [32]. Respondents who were alcohol consumers had a strong HB vaccination willingness. It is generally accepted that long-term drinking might be a potential cause of liver damage. Therefore, drinkers are concerned more about liver protection and are more willing to be inoculated.
Unlike some published studies, income was not detected to significantly affect either HB vaccination or vaccinating willingness as the balanced distributions within every income levels in our study showed [33]. Having said that, 8.15% of non-vaccinated respondents argued that too much self-payment was a big obstacle. Therefore, more subsidy or reimbursement targeting non-vaccinated adults is necessary to explore to reduce the financial burden of migrant workers [9].
In addition, only 35.58% of non-vaccinated respondents expressed willingness to vaccinate in the future. The top-three reasons for their missing of HB vaccination were not having heard of the vaccination, unclear where the vaccination sites were situated, and distrustful of the effectiveness or safety, which suggested that an urgent task of enhancing publicity of HB knowledge and dissemination of health and immunization services should be advanced not only by health education and promotion but also by joint effort and cooperation among the relevant administrative departments.

Limitations of the Study
Some limitations of the present study should be acknowledged. First, as it was a cross-sectional study, any causal inference should be made cautiously on the basis of the association observed in our study. Second, respondent non-random sampling was adopted due to the limitation of the availability and acceptance of subjects, which may cause a selection bias in the occupational distribution between sampled respondents and the overall migrant workers in Chongqing. Third, data collected by interviews may generate a recall bias both on HB vaccination status and risk behaviors. Therefore, a longitudinal study with a larger sample size by quota sampling, matching with vaccination records or serological examination, is needed to make the findings more robust.

Conclusions
Approximately one half of the migrant workers had been inoculated with HB vaccine in the present study. The non-vaccinated respondents were relatively low in their willingness to be vaccinated. Younger age, higher education levels, higher knowledge scores, higher perceived risk of HB, and less condom use were associated with HB vaccination for migrant workers. Migrant workers who were better educated, drank alcohol, with higher knowledge scores and higher perception of HB susceptibility were more willing to inoculate HB vaccine. Health education and campaigns related to HB prevention should be focused on non-vaccinated migrant workers who are older, of lower education, and poor in HB knowledge. Peer education works for the circulation of HB knowledge, and new social media like the internet and smart phone apps will play an important role in addition to traditional social media like television, radio, and magazines. Author Contributions: X.L. and X.T. conceived and designed the study. X.L., H.X., M.X., L.G., K.C., S.L. performed field investigation and data collection. H.X. and X.L. conducted data analysis. X.L. and H.X. drafted the Abstract with input from X.T. and X.L. reviewed and polished the Abstract. All authors have seen and approved the final version of the Abstract for publication.
Funding: This research was funded by National Natural Science Foundation of China, grant number 71603034.