key: cord-0854939-raoeascj
authors: Ishimaru, T.; Okawara, M.; Ando, H.; Hino, A.; Nagata, T.; Tateishi, S.; Tsuji, M.; Matsuda, S.; Fujino, Y.
title: Gender differences in the determinants of willingness to get the COVID-19 vaccine among the working-age population in Japan
date: 2021-04-20
journal: nan
DOI: 10.1101/2021.04.13.21255442
sha: 84842cde217e79f134080a3fc7b9d495623860e0
doc_id: 854939
cord_uid: raoeascj

Many factors are related to vaccination intentions. However, gender differences in the determinants of intention to get the coronavirus disease 2019 (COVID-19) vaccine have not been fully investigated. This study examined gender differences in the determinants of willingness to get the COVID-19 vaccine among the working-age population in Japan. We conducted a cross-sectional study of Japanese citizens aged 20-65 years using an online self-administered questionnaire in December 2020. Logistic regression analysis was performed. Among 27,036 participants (13,814 men and 13,222 women), the percentage who were willing to get the COVID-19 vaccine was lower among women than among men (33.0% vs. 41.8%). Age and education level showed a gender gap regarding the association with willingness to get the COVID-19 vaccine: men who were older or had a higher level of education were more willing to get the vaccine, whereas women aged 30-49 years and those with a higher level of education showed a relatively low willingness to get the vaccine. For both men and women, marriage, higher annual household income, underlying disease, current smoking, vaccination for influenza during the current season, and fear of COVID-19 transmission were linked to a higher likelihood of being willing to get the COVID-19 vaccine. These findings give important insight into identifying target groups in need of intervention regarding COVID-19 vaccination, especially among women. Providing education about COVID-19 and influenza vaccination in the workplace may be an effective strategy to increase COVID-19 vaccine uptake.

Vaccines are expected to be a key measure against the coronavirus disease 2019 60 pandemic. 1 Vaccine development usually takes decades, and, although a 61 variety of COVID-19 vaccines are currently being developed at an unprecedented rate, 62 evidence has not yet established their long-term safety. 2 In Japan, the COVID-19 63 vaccination program has been administered since February 2021, with eligibility 64 proceeding in the following order: healthcare workers, older adults (aged 65 years or 65 older), and people with underlying diseases and caregivers working at care facilities for 66 older adults. 3 Other people will become eligible for vaccination at a later date, following 67 the populations listed above, which are the main sources of infection. 4 To reach herd 68 immunity, most citizens need to be vaccinated. 5 69 70 Vaccination intentions are recognized as the most important issue in the rollout of 71 vaccination programs. 6 However, Japan is known for its lack of public trust in vaccines 7 72 because negative campaigns against the vaccination-highlighting scandals and severe 73 adverse effects, for instance-have increased hesitation concerning some vaccinations 74 over the past several decades. vaccine. 14-18 In Japan, willingness to get the COVID-19 vaccine was previously 83 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 examined in a single study of 1,100 adults, with results showing that vaccine hesitancy 84 was higher in Japan than in other countries, particularly among women. 19 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 Criteria for eligibility for the present study were being aged 20-65 years and currently 108 working. Healthcare workers and caregivers were not invited to participate. We 109 recruited study participants from the panelists of Cross Marketing Inc. (Tokyo, Japan), 110

an Internet research company. These panelists regularly respond to self-administered 111

Internet surveys distributed by the company, receiving tokens that can be redeemed for 112 products and services as compensation. 113 114 For this survey, we used quota sampling by gender, residence (five districts), and 115 occupation (office worker or other). Information on these characteristics was retrieved 116 from the panelists' Cross Marketing Inc. registration information. Cross Marketing Inc. 117 sent an invitation email to panelists who met the eligibility criteria, and those who were 118 interested in the study proceeded to the survey via a hyperlink. The participants 119 provided informed consent prior to beginning the online questionnaire. We set a quota 120 of 1,650 participants for each of the 20 strata and ceased recruitment when the target 121 number was reached. In total, 33,087 participants from all over Japan completed the 122 questionnaire. After excluding invalid responses, a total of 27,036 participants 123 remained. All participants received standard incentives through Cross Marketing Inc. 124 (i.e., a few dollars' worth of tokens). The study was conducted according to the 125 guidelines of the Declaration of Helsinki, and it was approved by the Ethics Committee 126 of the University of Occupational and Environmental Health, Japan (Approval number: 127 R2-079). 128 129

The survey questions included items on demographic and health characteristics, 131 influenza vaccination status during the current season, fear of COVID-19 transmission, 132 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 and willingness to get the COVID-19 vaccine. The demographic and health 133 characteristics were gender, age, education, marital status, annual household income (1 134 USD was equal to 106.78 JPY, using 2020 conversion rates), 23 underlying disease, and 135 smoking status. In line with a previous study, 18 we developed the following item to 136 assess willingness to get the COVID-19 vaccine: "If a COVID-19 vaccine becomes 137 available, I will get it." The possible answers to this question were yes and no. 

We calculated frequencies and proportions for all variables. Descriptive statistics were 143 also calculated by gender to reveal more detailed background factors. Logistic 144 regression analysis was performed to calculate adjusted odds ratios (aORs) and 95% 145 confidence intervals (CIs) for all variables to evaluate associations with willingness to 146 get the COVID-19 vaccine. Because we hypothesized that gender was an important 147 determinant in the current study, the analysis was stratified by gender. We adjusted for 148 age; no adjustments were made for the other covariates because the purpose of this 149 study was to detect target populations for future intervention programs. All P-values 150 were two-sided, and statistical significance was set at P < .05. We used Stata/SE 16.1 151 (StataCorp, College Station, TX, USA) for all analyses. 152

153 Results 154 Table 1 shows the characteristics of the study population. Data on a total of 27,036 155 participants (13,814 men and 13,222 women) were analyzed. We found that 43.0% of 156 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. This study provides evidence on gender differences in the determinants of willingness 195 to get the COVID-19 vaccine. In the present study, the percentage of people who were 196 willing to get the COVID-19 vaccine was lower among women than among men 197 (33.0% vs. 41.8%). This trend was similar to the findings of previous studies in 198 developed countries. 14, 19, 24, 25 Notably, we found a gender gap in the associations of age 199 and education level with willingness to get the COVID-19 vaccine: men who were older 200 or had a higher level of education were more willing to get the vaccine, whereas women 201 aged 30-49 years and those with a higher level of education were less willing to get the 202 vaccine. Other factors showed similar trends for men and women: for both men and 203 women, a higher likelihood of being willing to get the COVID-19 vaccine was observed 204 in married participants, those with a higher annual household income, those who had an 205 underlying disease, those who currently smoked, those who had been vaccinated against 206 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101/2021.04.13.21255442 doi: medRxiv preprint influenza during the current season, and those with a fear of COVID-19 transmission. 207

These findings give important insight into identifying target groups for improving 208 COVID-19 vaccination intentions, especially among women. 209

The most notable finding of the present study was a gender gap in the associations of 211 age and education with willingness to get the COVID-19 vaccine. As mentioned above, 212 men who were older or had a higher level of education were more willing to get the 213 vaccine, whereas women aged 30-49 years and those with a higher level of education 214

were less willing to get the vaccine. For men, this finding is consistent with previous 215 studies; older age and higher level of education are known facilitators for health 216 protection, which should correlate with COVID-19 vaccination intentions. 24, 26, 27 217 Conversely, the current results for women vary from the findings presented in these 218 studies. This difference may be explained by a higher level of concern about adverse 219 effects of the vaccine among women aged 30-49 years and among those with a 220 relatively high level of education in Japan during the current study period. We 221 considered that recent reporting on adverse effects following receipt of the human 222 papillomavirus vaccine in major Japanese newspapers may have had a negative impact 223 on the vaccination intentions of these groups. 28 These findings may imply the need for 224 intervention in this population regarding COVID-19 vaccination. 225

The current study found that influenza vaccination status and fear of COVID-19 227 transmission were strongly associated with willingness to get the COVID-19 vaccine 228 among both men and women. These results are in line with previous studies. 16, 17, 29, 30 In 229 the case of the 2009 influenza pandemic, perceived risk of infection translated into 230 preventive behaviors, including vaccination. 31 Similarly, the experience of receiving a 231 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. In the current study, a higher likelihood of being willing to get the COVID-19 vaccine 239 was observed among those who were married, those with a higher annual household 240 income, those with an underlying disease, and those who currently smoked. Vulnerable 241 populations, such as people with low socioeconomic status, are often found to have 242 relatively poor health and to need support regarding their willingness to be vaccinated. 16 243

In addition, a previous study on influenza vaccination suggests that single people are 244 less willing to be vaccinated compared with their married counterparts because single 245 people are not at risk of transmission from coresident family members. 11 Our finding 246 regarding the presence of underlying disease may be related to the fear of COVID-19 247 transmission because underlying disease a known risk factor for clinical severity of 248 COVID-19 infection. 24 Only our finding on tobacco use was inconsistent with previous 249 studies, which have found that smokers often tend to refuse vaccines regardless of the 250 risk of clinical worsening associated with smoking. 10, 11, 33 Further studies should focus 251 on confirming the relationship between smoking and COVID-19 vaccination intention. 252 253 Overall, we found that 37.5% of the working-age population in Japan was willing to get 254 the COVID-19 vaccine. This percentage is not high enough to achieve herd immunity 255 through the vaccination program. 5 This acceptance rate is lower than that in other 256 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101/2021.04.13.21255442 doi: medRxiv preprint countries, such as China (91.3%), France (58.9%), the US (56.9%), and Russia 257 (54.9%). 6 Furthermore, the acceptance rate is lower than the rate reported in a previous 258 study in Japan (65.7%) that was conducted in September 2020. 19 One possible reason 259 for the low vaccination intention in the current study may be tied to the research 260 period. 15 Our study was conducted in December 2020, right after the start of the 261 COVID-19 vaccination program in the UK and the US, and the media coverage of 262 adverse effects was exaggerated. 34, 35 Another reason is that people of working age have 263 been found to be less willing to get the COVID-19 vaccine compared with older 264 adults, 19 which would affect the present results. These findings have important 265 implications for vaccination intentions, although vaccination acceptance may change 266 after the vaccine is available for administration among the working-age population. 267

The main strengths of this study are the large sample size and the use of a sample from 269 throughout Japan. A limitation of this study is that we recruited study participants from 270 an Internet research company's list of panelists. A previous study on the anti-vaccine 271 movement reported that anti-vaccine messages were more prevalent on the Internet than 272 in other sources. 36 Therefore, the participants in the current study may have been 273 particularly likely to access anti-vaccine websites, and this should be taken into account 274 when interpreting the results of our study. Additionally, we conducted the current study 275 before the administration of the COVID-19 vaccination program in Japan; therefore, we 276 could not provide participants with detailed information, such as the vaccination 277 schedule, which may have affected their willingness to get the vaccine. 278

In conclusion, the current study revealed a gender gap in the associations of age and 280 education level with willingness to get the COVID-19 vaccine. In particular, women 281 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101/2021.04.13.21255442 doi: medRxiv preprint Table 1 . Characteristics of the study population. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 20, 2021. ; https://doi.org/10.1101 https://doi.org/10. /2021 

Influenza vaccination uptake among the working age 345 population of Japan: results from a national cross-sectional survey

Safety 348 and perception: What are the greatest enemies of HPV vaccination programmes? 349

Current knowledge of and attitudes toward 352 human papillomavirus-related disease prevention among Japanese: A large-scale 353 questionnaire study

Determinants of COVID-19 356 vaccine acceptance in the US

Yeoh 359 EK. Change of Willingness to Accept COVID-19 Vaccine and Reasons of Vaccine 360 Hesitancy of Working People at Different Waves of Local Epidemic in Hong Kong, 361 China: Repeated Cross-Sectional Surveys. Vaccines (Basel) 2021

Acceptability of a COVID-19 vaccine among 364 adults in the United States: How many people would get vaccinated?