Consistent with the non-migrant vaccination studies in Korea, Brazil and China [17 18 36 37], Table 3 reports that migrant workers’ vaccination intention declined with age and sex was not statistically significant. Compared with the unmarried group, married migrant workers were more likely to take the HB vaccine. For self-rated health, the good health group was more likely to take up the HB vaccine than the poor health group, which is also consistent with previous non-migrant HB vaccination behavior studies [17]. This suggests that HBV prevention and education policy should focus on the migrant workers who are in the older and unmarried, and poor self-rated health status, groups.
Taking the migration-industry variables, only migrant location was significantly associated with vaccination intention, with the Tianjin migrant group more willing to choose HB vaccination than non-Tianjin migrants. This reinforces the recommendation that HBV health education and HB vaccination management policy should concentrate on migrant workers from other provinces. There were no significant differences whether the migrant worker was accompanied by family members and between the three industry groups. We expected industry differences since migrant workers in the retail and service industry were required by the government to undergo a physical examination to make sure they were not an HBV carrier before starting work. This absence of industry differences might reflect a high level of inadequate HBV and HB vaccine knowledge of migrant workers irrespective of industry sectors. If migrant workers were ill-informed about HBV, then health officials missed an opportunity to educate migrant workers in services and retail industries on the benefits of the HB vaccine during the physical examination process. One place to improve migrant worker HBV education is during the existing testing of workers. We also recommend an expansion of the testing regime to all workplaces.
Validating our PMT approach, Model 3 shows that only severity, self-efficacy and response cost were the significant cognition factors determining HB vaccination intention. A meta-analysis of the literature on PMT found that not all PMT variables are able to predict a given behavior or intention with the same strength, with the role and influence of PMT variables varying across different vaccination domains and research populations [25 26]. One of the most researched areas, influenza vaccination intention, found that response efficacy and self-efficacy were significant predictors [38 39]. One study showed that compared to threat appraisal concepts (severity, vulnerability), coping appraisal concepts (response efficacy, self-efficacy and response costs) usually have stronger relationships with the adaptive intention to vaccinate [25]. Our results are consistent with these other PMT studies.
While a partial PMT migrant worker model was estimated by Liu et al (2019) [20], our full PMT model is the first to estimate vaccination cost data, socio-demographic and migration-industry characteristics for migrant worker HBV vaccination intention in China. The smaller the HB vaccine costs, the higher the probability migrant workers intended to vaccinate (response cost factor OR = 0.612). Response costs, self-efficacy and severity from threat appraisal concepts were significant for migrant workers, which is consistent with non-migrant worker HBV studies. The odds-ratio (OR) value of severity was 1.493, which indicates that the probability of intention to take the HB vaccine would rise 1.493 units as the severity factor increased one unit. This means that when migrant workers were more likely to realize the serious and unrecoverable consequence of HB, there was a higher probability of them vaccinating. Also, the willingness to receive the HB vaccine would increase 1.738 units when the self-efficacy factor increased one unit (self-efficacy factor OR = 1.738).
Our results differ from Liu et al’s [20] partial PMT model, where vulnerability and response-efficacy were found to be the significant PMT factors determining migrants’ HB vaccination intention. The different results can be explained by the different study places, different measurement tools and different model specification, with Liu et al PMT model missing variables. Our findings that severity, self-efficacy and response cost were significant PMT factors suggests that migrant populations need improved knowledge about the side effects of the HB vaccine and require information on HB vaccinations, which should form a key content in health education. Management policy for improving the HB vaccination rate should eliminate these barriers to facilitate migrant worker’s vaccination rates
By including migration-industry variables, our paper also expanded the migrant worker HBV literature. As a vulnerable group, the migrants’ origins should be considered when measuring PMT. Our study follows previous non-migrant worker PMT subpopulations research, for instance, adolescent’s drug use and smoking, women’s preferences for contralateral prophylactic mastectomy and selective estrogen reuptake modulators and travellers' self-protections, by dissecting the migrant worker into Tianjin and non-Tianjin migrants [40–44]. As several PMT meta-analysis [25 26] showed, the specificity of the PMT measurement for special populations are crucial, especially when PMT will be applied to provide guidance at the operational level. Our study identified the crucial differences in the migrant population and the need to target subgroups of migrant workers.
Our findings have important implications for health promotion, education design and immunization management. The social-demographic characteristics and migration-industry characteristics identified the target migrant worker population for health education and immunization management policy: those who are older, unmarried, have poor self-rated health status and from outside Tianjin. Our threat evaluation results suggest that the content of HB education messages should emphasize the severity of HB, including identifying HBV symptoms, the heavy economic burden of HBV, the worries and pain of family members, the barriers to acquiring employment and potential social discrimination. We found coping evaluation, self-efficacy and response cost were strong predictors of migrant workers’ vaccination intention, which informs both health education content design and immunization management policy planning. Considering migrant workers’ characteristics, better health outcomes will depend on improving migrants’ self-efficacy, reducing the HBV response cost and expanding the accessibility to health and vaccination services for the migrant workers. Based on the significant role of self-efficacy in Table 3, health authorities should tap into work organizations as a location for vaccinations where other workers being vaccinated encourages individuals to vaccinate. According to the influence of response cost factor in Table 3, the side effects of the HB vaccine should be included in the health education content and health authorities should provide more information on where, when and how to get the HB vaccine. Our study highlights the importance in health policies identifying subgroups, Tianjin and non-Tianjin migrants, within migrant worker populations. Our results suggest that health managers should establish more vaccination sites, especially workplaces, and visit existing injection service sites [32].
Maladaptive response rewards were not included in our PMT model. In some previous vaccination studies, maladaptive response rewards were measured by saving money or time, avoiding the side effects of the vaccine and acquiring natural immunity to subsequent infection [39 45]. Since HB is incurable, acquiring natural immunity does not exist, and saving expenses and worries about the side effects have been included in response cost, so maladaptive response rewards may not be a serious omission in our PMT model. However, the abandonment of maladaptive response rewards warrants empirical study in future HBV research.
Second, our study did not assess vaccination behaviour directly. While it has been shown that vaccination behavior in empirical studies can be predicted by previous intention in a wide range of contexts [46 47], future studies should include vaccination behavior directly. Third, as a result of convenience sampling, our results and findings need confirmation through studies of other migrant worker populations, regions and industries.