Factors of influenza vaccine inoculation and non-inoculation behavior of community-dwelling residents in Japan: Suggestions for vaccine policy and public health ethics after COVID-19

Abstract

The aim of this study is to provide basic information that contributes to vaccine inoculation policy after COVID-19. We used the secondary data of the influenza vaccine inoculation behavior survey for community-dwelling adults conducted in 2011, before the COVID-19 pandemic, but after the 2009 novel influenza A (H1N1) pdm 09 pandemic. All factors such as socio-demographic characteristics, health-related behaviors, family environment, physical and social environment, and area of residence were adjusted, and factors related to vaccine inoculation behavior were analyzed. Those living with pregnant women had a significantly higher odds ratio of inoculation; this was self-evident in that those people considered infection to their family. Regarding the social environment, those aged 20–64 years with a significantly higher adjusted odds ratio of inoculation were those with “at least five people with which they interacted in the neighborhood”. This result can be interpreted in two ways relating to altruism in Japan. Finally, we indicated the importance of learning from the past, including the case of 2009.

Introduction

Various vaccination strategies were discussed during the COVID-19 pandemic, such as prioritizing inoculations (resource allocation), issues of inoculation refusal, and public health significance of vaccines (i.e., preventing infection of others). There were several similar discussions approximately 10 years ago. This was during the 2009 novel influenza A (H1N1) pdm 09 pandemic. The prioritization of inoculations was discussed. The World Health Organization (WHO) [1] stated the following objectives: 1) integrated functioning of the health care system and protection of the basic foundations of each country, 2) reduction of morbidity and mortality, and 3) reduction of virus transmission in society. In this context, they stated that health care workers should be prioritized. In addition, the US Centers for Disease Control and Prevention (CDC) [2] stated that health care workers and emergency personnel should be prioritized given their impact on the medical system, after which pregnant women and those with concerns about exacerbation of illness should be prioritized. At the time, the CDC prioritized younger people aged 6 months to 24 years due to their collective actions and movement (i.e., from the public health belief that they would spread infection throughout society). The national awareness survey on H1N1 vaccine inoculation in Japan is of interest (2009, N = 600) [3]. At this point, H1N1 vaccine inoculation was decided to be voluntary. As a result, 1) 60% wanted inoculation, but 30% did not. The reason for wanting inoculation was “for self-protection,” and not wanting inoculation was “safety concerns”. 2) There was strong resistance against international vaccines (safety concerns), 3) there were concerns about vaccine effectiveness, and 4) parents actively recommended their children aged less than 15 years to get the vaccine. Prioritization of vaccine provision to healthcare professionals was supported by over 90% of respondents.

Therefore, society has had a considerable debate regarding inoculation policies at the time of the novel influenza A (H1N1) pdm 09 pandemics. Based on current knowledge (as of June 2022), the differences between H1N1 and COVID-19 were: 1) COVID-19 is more lethal than H1N1, but its lethality is incomparably low when compared to Ebola hemorrhagic fever or SARS; 2) COVID-19 appears to have higher infectivity, and 3) COVID-19 has a faster appearance of mutant variants. In this way, although there are differences (not yet confirmed as scientific fact), the discussions at the time of the H1N1 pandemic appear to be very helpful for the current COVID-19 pandemic.

Therefore, this study considered further learning from conventional influenza vaccine inoculation policy and people’s inoculation behavior. This present study aims to use the data of the influenza vaccine inoculation behavior survey for community-dwelling adults conducted in 2011, before the COVID-19 pandemic but after the 2009 novel influenza A (H1N1) pdm 09 pandemic. Secondarily, all factors such as socio-demographic characteristics, health-related behaviors, family environment, and physical/ social environment of the area of residence were adjusted; and factors related to vaccine inoculation behavior re-analyzed. Basic materials contributing to vaccine inoculation policy after COVID-19 and several recommendations will be provided. The present study may be useful for combating mild to moderate but highly contagious emerging infectious diseases rather than highly lethal ones such as Ebola hemorrhagic fever and SARS.

Summary of global status of influenza

Morbidity and mortality: Influenza affects 5–10% of adults and 20–30% of children worldwide early, with an estimated 3–5 million illnesses and 250,000–500,000 deaths yearly [4]. Meanwhile, 80% of influenza-related deaths are among older adults [5], with high complications risks, hospitalizations, and deaths among older adults  ≥65 years [6], [7], [8], [9], [10], people with underlying illnesses [11], [12], [13], [14], [15], and children <five years. In addition, pregnancy has been reported to increase the risk of complications due to influenza morbidity among healthy adult females [16]. Thus WHO has indicated older adults, people with underlying illnesses, pregnant women, and children aged 6–23 months as high-risk individuals. Inoculation is recommended for those who are in regular contact with these high-risk individuals [17].

Factors related to influenza vaccine inoculation behavior: Previous studies

In 2011, the inoculation rate among Japanese people was 28.6% in those aged 13–64, years [18]. Research on factors related to vaccine inoculation behavior includes older adults [19], [20], [21], [22], people with underlying illnesses and outpatients [23], [24], [25], adults [26], [27], [28], [29], [30], health care workers [31], [32], and pregnant women [33], [34]. Furthermore, surveys have been conducted to examine the relationship between inoculation and socio-demographic characteristics (e.g., age, sex, educational history, and annual household income), health conditions (e.g., presence or absence of underlying illnesses), and health behaviors (e.g., exercise, drinking, and smoking habits). However, no study has examined the relationship between inoculation and the presence or absence of pregnant women in the same home. The physical environment of the residential area is also important. It was reported that the inoculation rate increases with secured access to medical institutions in the United States [19], [20]. The relationships between inoculation behavior and the degree of involvement with friends and neighbors have been investigated as social factors. A study in Italy showed that inoculation rates were high in environments where neighbors assisted when in need [35]. Inoculation is impacted by information provided by social networks, and the relationship between social participation and inoculation has also been investigated [36].

In this way, vaccine inoculation behavior is influenced by factors such as age, sex, educational history, work, illnesses while an outpatient, and health-related behaviors; simultaneously, the family environment and the physical and social environments of the residence are also involved. In Japan, the Immunization Act was revised in 1976, and up to 1994, to prevent exacerbation of the illness among the elderly, mass influenza vaccine inoculations were required for elementary and junior high school students, whose infection was likely to increase infection rates in the community [37]. However, the government suffered a series of lawsuits for damages and health impairment in elementary and junior high school students caused by the mass inoculations. The revision of the Immunization Act in 1994 excluded influenza from the target diseases for regular inoculation [37], [38]. After the abolition of mass inoculations for elementary and junior high school students, the number of deaths among community-dwelling older adults increased during the influenza epidemic [37]. Given the movement of recommending inoculation to older adults overseas, the Immunization Act was revised in 2001, and vaccine inoculation became recommended for older adults. Currently, according to the Immunization Act, inoculations are recommended for “≥65 years; alternatively, those ≥ 60 years and > 65 years, with disabilities that extremely restricted their activities of daily living due to the functions of their heart, kidneys, or respiratory organs; or those who have a degree of impaired immune function from human immunodeficiency virus that makes activities of daily life nearly impossible” [39].Given this background, we divided subjects into two age groups (20–64 years and ≥ 65 years) in this study and analyzed the data with the presence or absence of vaccine inoculations the objective variable and other factors as the explanatory variables.

Methods

• 1.Survey area and participants

This study used secondary data of the “Survey on Citizens’ Health Awareness, Health Condition, and Health Behaviors” conducted by a city in the Tohoku region. The survey involved distributing questionnaires in October 2011, retrieved by participants visiting the health and hygiene promotion committee within two weeks. Compared to the national survey results [40], the survey region had many families with three or more generations living together and a large amount of snow cover. Participants were 5,002 randomly selected people and stratified according to districts, sex, and age out of 99,653 citizens aged 20 years or older and under 80 years in the Basic Resident Register as of March 31, 2011. There were 26 districts in the city, and the extraction rate was changed to ensure equal distribution per district, and the number of men and women selected from each district was the same. A vaccine inoculation ticket (postcard) was mailed to the subjects at the time of the survey. The out-of-pocket cost of inoculation for influenza vaccine recommended for individuals (≥65 years, or ≥60 years and <65 years, and with underlying illnesses as stipulated by the Immunization Act) was JPY 1,500 (about USD 11) and 50% subsidy by city.

• 2.Survey items

Socio-demographic characteristics included age, sex, educational history, annual household income, type of work, and illness during hospital visits. Health-related behaviors evaluated in the survey included general health examinations and cancer screenings in the past year. Exercise, eating, drinking, and smoking habits, family environment (including living with pregnant women), as well as the physical and social environment of the area of residence were assessed. The social environment included four-choice items on the number of people that the participants interacted with in the neighborhood (20 people or more, 5–19 people, four people or fewer, no acquaintances/interactions) was divided into two (≥5 people, ≤ 4 people) for analysis. Subjects were asked to answer either “yes” or “no” regarding whether the influenza vaccine inoculation was received in the past year.

• 3.Ethical considerations

Self-administered anonymous questionnaires were used for the study. Consent to the survey was obtained by submission of the questionnaire. This study was approved by the Institutional Review Board of the University of Tokyo School of Medicine (approval number 1568-(4)).

• 4.Statistical analysis

Percentages or mean values (standard deviation, SD) of socio-demographic characteristics, health-related behaviors, family environment, and the area of residence were calculated to understand the characteristics of the participants. In this study, participants were divided into ages 20–64 and ≥65 years. The presence or absence of influenza vaccine inoculation was set as the objective variable, and other items were set as the explanatory variables. The crude odds ratio was used to investigate relationships between inoculation behavior and each item. Furthermore, multiple logistic regression analyses we conducted by adjusting for all items of socio-demographic characteristics, health-related behavior, family environment, and environment of the area of residence.

In logistic regression analysis, the adjusted odds ratio of vaccine inoculation for those aged 20–64 years and vaccine non-inoculation for those ≥65 years were obtained. In the Japan, influenza vaccination is recommended for people aged ≥65 years and people with certain diseases aged 60–64 years. Therefore, we have decided to focus on the characteristics of those 1) who are non-vaccinated in the ≥65 years age group, and 2) healthy adults under 64 years old who are vaccinated although vaccinations are not recommended. Stata Ver.15 were used for analyses, and the significance level was set to 5% on both sides.

Results

• 1. Survey subjects and analysis subjects ( Table 1 )

Table 1.

Socio-demographic characteristics of the analyzed participants.

		20–64 years		65–79 years
		n=3,162		n=1,103
	Age	45.6	(12.7)	71.9	(4.2)
	20–39 years	1,078	(34.1)
	40–64 years	2,084	(65.9)
Sex
	Female	1,622	(51.3)	559	(50.7)
	Male	1,540	(48.7)	544	(49.3)
Educational history (final educational history)
	Junior high school	250	(8.3)	546	(53.4)
	High school	1,729	(57.6)	358	(35.0)
	Vocational school / junior college / university / graduate school	1,024	(34.1)	119	(11.6)
Household income
	Less than 2 million yen	475	(20.8)	243	(33.5)
	2–6 million yen	1,341	(58.8)	407	(56.1)
	6 million yen or more	466	(20.4)	75	(10.3)
Work
	Work (all day)	1,723	(54.9)	67	(6.3)
	Agriculture and forestry-industry / self-employed	469	(15.0)	246	(23.2)
	Work (part-time)	366	(11.7)	41	(3.9)
	Unemployed	580	(18.5)	706	(66.6)
Illnesses for hospital visits
	Not going to hospital	2,328	(77.5)	295	(33.6)
	Lifestyle-related illnesses†	555	(18.5)	509	(58.0)
	Orthopedic illnesses‡	122	(4.1)	74	(8.4)

For age, results show mean (standard deviation); for others, the number of people (%).

†:Includes diabetes, angina, myocardial infarction, asthma, hypertension, stroke, hypercholesterolemia, hypertriglyceridemia, gout, hyperuricemia, stomach/duodenal illnesses, hepatitis, liver cirrhosis.

‡:Includes lower back pain, knee joint pain, and osteoporosis.

Questionnaires were distributed to 5,002 participants, and 4,570 returned theirs (91.3%). Those whose age, sex, presence or absence of influenza vaccine inoculation was unknown or those outside the target age range (80 years or older) were excluded; Thus, 4,265 (85.2%) participants were analyzed.

• 2.Influenza vaccine inoculation rate in Japan

The overall inoculation rate for those aged 20–79 years was 40.2% (1,714), with 33.8% for men and 46.3% for women. The overall inoculation rate was 31.7% (342) for those aged 20–39 years, with 22.6% for men and 40.6% for women (P < 0.001); 31.1% (648 people) for those aged 40–64 years, with 23.3% for men and 38.4% for women (P < 0.001); and 65.6% (742 people) for those aged 65–79 years, with 64.2% for men and 67.1% for women (P = 0.306). The inoculation rate for women was higher than for men in all age groups.

• 3.Factors relating to influenza vaccine inoculation behavior

3.1. Relationship between 20–64 years vaccine inoculation behavior, ≥65 years vaccine non-inoculation behavior, and all survey items (univariate analysis, Table 2).

Table 2.

Associations between influenza vaccine inoculation/non-inoculation behavior and socio-demographic characteristics, health-related behavior, family environment, and physical/social environment of the area of residence (univariate analysis).

		20–64 years			65–79 years
		Vaccine inoculation behavior			Vaccine non-inoculation behavior
		OR	(95%C.I.)	p-value	OR	(95%C.I.)	p-value
Age
	20–39 years	1
	40–64 years	0.97	(0.83, 1.14)	0.717
Sex
	Female	1			1
	Male	0.47	(0.40, 0.54)	<0.001	1.14	(0.89, 1.46)	0.306
Educational history (final educational history)
	Junior high school	1			1
	High school	1.32	(0.97, 1.81)	0.081	1.07	(0.81, 1.42)	0.624
	Vocational school / junior college / university / graduate school	2.23	(1.61, 3.08)	<0.001	1.08	(0.71, 1.64)	0.726
Household income
	Less than 2 million yen	1			1
	2–6 million yen	1.44	(1.14, 1.83)	0.002	0.82	(0.59, 1.14)	0.228
	6 million yen or more	1.89	(1.43, 2.50)	<0.001	0.61	(0.34, 1.07)	0.083
Work
	Work (all day)	1			1
	Agriculture and forestry-industry / self-employed	0.58	(0.46, 0.74)	<0.001	0.69	(0.40, 1.19)	0.179
	Work (part-time)	0.80	(0.63, 1.03)	0.082	0.97	(0.44, 2.11)	0.929
	Unemployed	0.94	(0.77, 1.15)	0.530	0.61	(0.37, 1.01)	0.057
Illnesses for hospital visits
	Not going to hospital	1			1
	Lifestyle-related illnesses†	1.35	(1.11, 1.64)	0.002	0.41	(0.31, 0.56)	<0.001
	Orthopedic illnesses‡	1.35	(0.93, 1.98)	0.119	0.43	(0.25, 0.74)	0.002
General medical examinations
	Present	2.23	(1.80, 2.75)	<0.001	1
	Absent	1			3.82	(2.78, 5.27)	<0.001
Cancer screenings
	Present	2.09	(1.78, 2.44)	<0.001	1
	Absent	1			2.52	(1.91, 3.32)	<0.001
Exercise habits♯
	Present	1.09	(0.90, 1.31)	0.395	1
	Absent				1.47	(1.10, 1.97)	0.009
Consideration for eating habits
	Present	1.66	(1.42, 1.93)	<0.001	1
	Absent	1			1.16	(0.90, 1.50)	0.260
Alcohol drinking habits§
	Present	0.70	(0.59, 0.84)	<0.001	1
	Absent	1			0.81	(0.60, 1.10)	0.186
Smoking habits
	Never smoked	1			1
	Quit smoking	0.90	(0.75, 1.08)	0.252	1.06	(0.79, 1.41)	0.699
	Smoking	0.47	(0.39, 0.57)	<0.001	2.48	(1.66, 3.70)	<0.001
Family composition
	Couple	1			1
	Single	0.63	(0.40, 0.98)	0.040	1.69	(1.06, 2.67)	0.026
	2-generation household	0.84	(0.65, 1.10)	0.210	1.38	(0.99, 1.92)	0.057
	3+ generation household	1.01	(0.76, 1.33)	0.949	1.01	(0.69, 1.49)	0.955
	Other	0.86	(0.62, 1.20)	0.373	1.43	(0.91, 2.26)	0.120
Cohabitation with pregnant woman
	Present	1.72	(1.19, 2.49)	0.004	1
	Absent	1			0.64	(0.19, 2.10)	0.457
Cohabitation with preschool infant
	Present	1.56	(1.30, 1.89)	<0.001	1
	Absent	1			0.76	(0.47, 1.21)	0.249
Cohabitation with elementary school student
	Present	1.67	(1.40, 1.98)	<0.001	1
	Absent	1			1.37	(0.97, 1.93)	0.074
Environment of area of residence: physical
There are sidewalks on almost all roads
	Yes	1.03	(0.88, 1.21)	0.691	1
	No	1			1.09	(0.83, 1.44)	0.517
There are train stations / bus stops in neighborhood
	Yes	1.03	(0.88, 1.21)	0.685	1
	No	1			1.12	(0.85, 1.48)	0.411
There is a hospital that can be reached immediately in event of sudden illness
	Yes	1.38	(1.18, 1.61)	<0.001	1
	No	1			1.45	(1.10, 1.91)	0.008
Environment of area of residence: social
Number of people with whom participant is interacting in neighborhood
	5 or more	1.38	(1.18, 1.61)	<0.001	1
	4 or fewer	1			1.34	(1.01, 1.76)	0.039
There are neighbors who lend a helping hand when needing help
	Yes	1.20	(0.98, 1.46)	0.082	1
	No	1			1.40	(0.95, 2.06)	0.091
There is participation in community activities
	Yes	1.27	(1.07, 1.51)	0.004	1
	No	1			1.20	(0.88, 1.65)	0.253

OR:crude odds ratio.

C.I.: confidence interval.

†:Includes diabetes, angina, myocardial infarction, asthma, hypertension, stroke, hypercholesterolemia, hypertriglyceridemia, gout, hyperuricemia, stomach/duodenal illnesses, hepatitis, liver cirrhosis.

‡:Includes lower back pain, knee joint pain, and osteoporosis.

♯:At least twice a week, at least 30 min per session, except in winter.

¶:Following all apply: pay attention to nutritional balance, eat vegetables every day, no excess salt intake.

§:At least three times a week, at least one go (∼0.18 L) per session.

Socio-demographic characteristics: those aged 20–64 years with a significantly higher percentage of influenza vaccine inoculation included women, those with educational history of vocational school or above, those with a high annual household income, those who worked all day compared to people in agriculture, forestry, or fisheries, and those who visited the hospital for lifestyle-related illnesses. Those aged ≥65 years with a significantly lower percentage of vaccine non-inoculation was those who visited the hospital for lifestyle-related or orthopedic illnesses.

Health-related behavior: Those aged 20–64 years with a high influenza vaccine inoculation rate were found to have undergone general medical examinations and cancer screenings, made efforts to eat healthy without drinking alcohol, and exhibited a low rate of smoking. Those aged ≥65 years with a high rate of influenza vaccine non-inoculation were those who did not undergo general medical examinations or cancer screenings and smoked at a higher rate.

Family environment: Those aged 20–64 years with a high rate of influenza vaccine inoculation had a couple of households with three generations or more as opposed to single households in terms of family composition. Among families living together, those living with a pregnant woman, preschool infants, or elementary and junior high school students had a high rate of inoculation. Those aged ≥65 years with high rates of vaccine non-inoculation were those who lived alone. No significant differences were also seen among families living together.

Physical and social environment of the area of residence:

Physical environments among those aged 20–64 years with a significantly higher rate of vaccine inoculation included immediate healthcare access in the event of a sudden illness; social environments included those that interacted with five or more in the neighborhood and participation in community activities. Physical environments among those aged ≥65 years with a significantly higher rate of vaccine non-inoculation were those without immediate healthcare access in the event of a sudden illness. Social environments included those who interacted with four or fewer in the neighborhood.

3.2. Independent associations with influenza vaccine inoculation/non-inoculation behaviors (multiple logistic regression analysis, Table 3).

Table 3.

Associations between influenza vaccine inoculation/non-inoculation behavior and socio-demographic characteristics, health-related behavior, family environment, and physical/social environment of the area of residence (multivariate analysis).

		20–64 years			65–79 years
		Vaccine inoculation behavior			Vaccine non-inoculation behavior
		AOR	(95%C.I.)	p-value	AOR	(95%C.I.)	p-value
Age
	20–39 years	1
	40–64 years	0.71	(0.54, 0.94)	0.017
Sex
	Female	1			1
	Male	0.51	(0.38, 0.67)	<0.001	0.80	(0.40, 1.59)	0.520
Educational history (final educational history)
	Junior high school	1			1
	High school	1.71	(0.99, 2.98)	0.056	1.16	(0.67, 1.99)	0.597
	Vocational school / junior college / university / graduate school	2.71	(1.53, 4.78)	0.001	1.17	(0.53, 2.57)	0.699
Household income
	Less than 2 million yen	1			1
	2–6 million yen	1.24	(0.91, 1.69)	0.179	1.15	(0.65, 2.03)	0.642
	6 million yen or more	1.30	(0.90, 1.88)	0.168	0.81	(0.33, 1.96)	0.633
Work
	Work (all day)	1			1
	Agriculture and forestry-industry / self-employed	0.54	(0.35, 0.75)	0.001	0.95	(0.32, 2.84)	0.928
	Work (part-time)	0.70	(0.49, 1.01)	0.057	1.19	(0.23, 6.08)	0.833
	Unemployed	0.90	(0.65, 1.25)	0.538	0.60	(0.21, 1.70)	0.340
Illnesses for hospital visits
	Not going to hospital	1			1
	Lifestyle-related illnesses†	1.59	(1.19, 2.12)	0.002	0.60	(0.26, 0.74)	0.002
	Orthopedic illnesses‡	1.95	(1.18, 3.22)	0.009	0.44	(0.11, 1.22)	0.102
General medical examinations
	Present	1.60	(1.11, 2.29)	0.011	1
	Absent	1			1.89	(0.84, 4.27)	0.126
Cancer screenings
	Present	1.32	(1.01, 1.73)	0.039	1
	Absent	1			2.35	(1.18, 4.67)	0.015
Exercise habits♯
	Present	1.02	(0.78, 1.33)	0.900	1
	Absent	1			1.36	(0.80, 2.31)	0.261
Consideration for eating habits¶
	Present	1.22	(0.97, 1.53)	0.093	1
	Absent	1			0.61	(0.36, 1.04)	0.070
Alcohol drinking habits§
	Present	0.86	(0.65, 1.13)	0.282	1
	Absent	1			0.81	(0.43, 1.53)	0.523
Smoking habits
	Never smoked	1			1
	Quit smoking	1.43	(0.70, 1.93)	0.017	1.49	(0.75, 2.97)	0.254
	Smoking	0.95	(0.44, 1.28)	0.722	1.63	(0.66, 4.02)	0.292
Family composition
	Couple	1			1
	Single	0.86	(0.44, 1.67)	0.652	1.59	(0.63, 3.99)	0.326
	2-generation household	0.89	(0.61, 1.31)	0.551	1.71	(0.91, 3.21)	0.097
	3+ generation household	0.87	(0.57, 1.33)	0.522	1.30	(0.55, 3.05)	0.552
	Other	0.94	(0.58, 1.52)	0.807	1.02	(0.40, 2.62)	0.959
Cohabitation with pregnant woman
	Present	1.92	(1.13, 3.25)	0.015	1
	Absent	1			1.65	(0.08, 35.7)	0.751
Cohabitation with preschool infant
	Present	1.19	(0.88, 1.60)	0.263	1
	Absent	1			0.60	(0.22, 1.67)	0.328
Cohabitation with elementary school student
	Present	1.75	(1.35, 2.27)	<0.001	1
	Absent	1			3.08	(1.30, 7.33)	0.011
Environment of area of residence: physical
There are sidewalks on almost all roads
	Yes	1.23	(1.00, 1.53)	0.048	1
	No	1			1.29	(0.77, 2.16)	0.335
There are train stations / bus stops in neighborhood
	Yes	0.98	(0.78, 1.23)	0.876	1
	No	1			1.20	(0.72, 2.01)	0.480
There is a hospital that can be reached immediately in event of sudden illness
	Yes	1.41	(1.13, 1.77)	0.002	1
	No	1			1.55	(0.92, 2.62)	0.102
Environment of area of residence: social
Number of people with whom participant is interacting in neighborhood
	5 or more	1.36	(1.06, 1.74)	0.017	1
	4 or fewer	1			0.99	(0.56, 1.74)	0.970
There are neighbors who lend a helping hand when needing help
	Yes	0.95	(0.71, 1.28)	0.753	1
	No	1			1.01	(0.48, 2.13)	0.980
There is participation in community activities
	Yes	0.99	(0.76, 1.28)	0.940	1
	No	1			1.01	(0.57, 1.79)	0.972

AOR:adjusted odds ratio.

C.I.:confidence interval.

†:Includes diabetes, angina, myocardial infarction, asthma, hypertension, stroke, hypercholesterolemia, hypertriglyceridemia, gout, hyperuricemia, stomach/duodenal illnesses, hepatitis, liver cirrhosis.

‡:Includes lower back pain, knee joint pain, and osteoporosis.

♯:At least twice a week, at least 30 min per session, except in winter.

¶:Following all apply: pay attention to nutritional balance, eat vegetables every day, no excess salt intake.

§:At least three times a week, at least one go (∼0.18 L) per session.

Socio-demographic characteristics

Independent associations with vaccine inoculation behavior in those aged 20–64 years showed a significantly lower odds ratios of inoculation in those aged 40–64 years compared to 20–39 years, men in terms of sex, forestry and fishery industry, and self-employed people compared to full-day workers. In addition, significantly higher odds ratios of inoculation were observed in those with vocational school backgrounds or higher compared to those with a junior high school background and those visiting the hospital for lifestyle-related and orthopedic illnesses compared to those without illnesses for which they visit the hospital. Independent associations with influenza vaccine non-inoculation behavior in those aged ≥65 years showed significantly lower odds ratios of non-inoculation (i.e., were more likely to be inoculated) among those visiting the hospital for lifestyle-related illnesses compared to those not going to the hospitals.

Health-related behavior

Independent associations with vaccine inoculation behavior in those aged 20–64 years showed that those with significantly higher odds ratios of inoculation underwent general medical examinations, cancer screenings, and those who quit smoking. Independent associations with vaccine non-inoculation behavior in those ≥65 years revealed significantly lower odds ratios of non-inoculation (i.e., were more likely to be inoculated) among those who underwent cancer screenings.

Family environment

Independent associations with vaccine inoculation behavior in those aged 20–64 years showed that those with significantly higher odds ratios of inoculation lived with pregnant women or elementary or junior high school students. In contrast, vaccine non-inoculation behavior among those ≥65 years revealed that those with significantly higher odds ratios in the non-inoculation group did not live with elementary or junior high school students.

Physical and social environments of the area of residence

In terms of physical environments, the independent associations with vaccine inoculation behavior in those aged 20–64 years showed significantly higher odds ratios of inoculation in those with sidewalks on almost all roads and immediate access to the hospital in the event of a sudden illness. Regarding social environments, people who interacted with at least five neighbors showed independent associations with vaccine non-inoculation behavior among those ≥ 65 years, although no significant differences were found for any items.

Discussion

The greatest strength of the present study is its use of data from the 2011 influenza vaccine inoculation behavior survey for community-dwelling adults, which was conducted before the COVID-19 pandemic but after the 2009 novel influenza A (H1N1) pdm 09 pandemic. The results obtained in this study were almost entirely consistent with what was indicated in previous research. In other words, the significantly lower odds ratio for the socio-demographic characteristics can be explained for men in terms of sex and those aged 40–64 years compared to those aged 20–39 years in terms of age (i.e., in a more important and responsible position, not enough time due to parental care and child-rearing). Compared to those who work all day, agricultural and forestry-industry/self-employed people are thought to potentially be able to obtain opportunities for inoculation (workplace inoculations) depending on the workplace. This result may indicate that the inoculation rate will increase if workplace inoculations are promoted with COVID-19.

Meanwhile, the adjusted odds ratio was significantly higher for highly-educated people in terms of educational history. It can be explained that people who went to the hospital for lifestyle-related and orthopedic illnesses had a high awareness of the illness. Those with significantly higher odds ratios in terms of health-related behaviors included those who underwent general medical examinations, those who underwent cancer screenings, and those who quit smoking; these are all people with high health awareness. In terms of the family environment, previous research [26] has investigated spouses, but the present survey was novel in that it considered cohabitation with pregnant women. However, the results showed that those with a significantly higher odds ratio of inoculation were those who lived with pregnant women and or with elementary or junior high school students. In terms of the physical environment, those with a significantly higher odds ratio were those with sidewalks on nearly all roads and immediate access to a hospital in the event of sudden illness; these results showed access was important.

The only interesting result in the present study is that regarding the social environment. Those aged 20–64 years with a significantly higher adjusted odds ratio of inoculation were those with “at least five people they interacted in the neighborhood.” This result can be interpreted in two ways. The first interpretation is that having many acquaintances and friends in the neighborhood resulted in the transmission of information related to the vaccine and the experience of actually being inoculated, which then impacted inoculation behavior. Secondly, people who interact and engage in social participation may get inoculated due to their awareness of the potential increased risk of infecting others. For example, according to Nagata et al., vaccinated persons are more confident of the vaccine's effectiveness and value its benefits to their families and communities [41]. As will be described later, this perspective is important for public health ethics and will play a major role in future policies.

Commentary on future vaccine policy in relatively wealthy Japan

#1 If men or people busy with work cannot get inoculated, what needs to be done for them to do so? Workplace inoculations, establishing inoculation sites on weekends and at night, and the introduction of mobile vaccine inoculation stations may be effective, but a shortage of public health nurses and general nurses was an issue in the current COVID-19 pandemic. In both cases, a shortage of human resources and funds is expected. Thus, it is important to introduce the practice of subcutaneous injection and intramuscular injection of the vaccine in the education of all medical professionals. The law should be revised to reflect this. In doing so, medical professionals can be mobilized in an emergency.

#2 The convenience of access and workplace inoculation systems have previously been indicated: paving more roads is surely not the sole solution. Particularly for regions with high snow cover or islands, what is the optimal solution? Is having a family doctor an effective strategy for normal healthy people? At the very least, it is strongly recommended that public locations (e.g., public halls, government offices, schools, parks) be organized to become inoculation sites in the event of an emergency. Specifically, such plans should be incorporated into installation standards.

#3 If educational history is relevant, then how can the health literacy of citizens be raised? Is raising health literacy for all diseases difficult, and it is sufficient to do so only if vaccine inoculation is necessary. Interventions at the primary and secondary educational levels to raise health literacy for all diseases are considerably labor intensive. It is perhaps possible in health education to pinpoint the importance of inoculation and the importance of gargling, and hand washing.

#4 Public awareness of individual freedoms and public interest (public awareness on “Public Health Ethics”).

The extremely valuable data obtained in the present analysis were the results regarding the social environment. The result that “those who interact with many people” had a significantly higher odds ratio of inoculation could be interpreted that such people inoculated themselves because contacting with a large number of people by engaging in social participation inevitably increases the risk of infecting the others (i.e., due to the public interest or a certain type of altruism). This is a crucial point in public health ethics. Namely, in an emergency, the question is, how should we consider the balance between the “rights of individual freedoms” and “public interest of the whole”, and to what extent are paternal interventions by the government acceptable.

Such ideas of public health ethics have been rarely discussed in Japan. Punitive measures have been considered for the behavior of some younger people during the present COVID-19 pandemic, but there was little understanding of public health ethics. Government-run COVID-19 vaccine inoculation campaigns for children were also extremely insufficient. The instruction manuals attached to the inoculation tickets given to households only emphasized the point of “preventing infection to their children.” In fact, the perspective of “preventing children from becoming vectors of infection to others and from spreading infection throughout the community (public interest),” which was a major objective of inoculation among children, was entirely missing from the instruction manuals. The explanation that emphasized individual interests only to increase inoculation rates was significantly lacking the perspective of public health.

Why Japan could not lock down is likely a political factor and is beyond the scope of this paper. However, during the present COVID-19 pandemic, some people self-quarantined with the thought that they should not infect the other people. The objective of this paper is not to evaluate the vaccine strategy for the present COVID-19 pandemic. We dug up old data from 10 years ago to indicate the importance of learning from the past. It is undeniable that we had known so much during the H1N1 pandemic. Once again, this statistical secondary analysis of epidemiological data indicate that Japanese have a temperament of certain type of altruism. We hope this paper will serve as a basic material for considering vaccine policy after the COVID-19 pandemic.

Limitations of this study

The first limitation is that this was a cross-sectional study, and the causal relationship between inoculation behavior and each factor was only speculated after considering the effects, and they were not identified. The second limitation is that the surveyed region was just one regional city, and care is needed when generalizing these results to metropolitan areas. The third limitation is that antiviral drugs were widespread at the time, which is thought to have had a large impact on ideas regarding influenza treatment and inoculation behavior, but such aspects cannot be discussed in the present paper.

Conclusions

Influenza is one of the oldest infectious illnesses in history. Global pandemics have repeatedly occurred since the modern era, with the Spanish flu that became widespread in 1918, the discovery of the influenza virus as the pathogen, the Asian flu that started in 1957, the Hong Kong flu in 1968, Russian flu in 1977, and the novel influenza A (H1N1) pdm 09 in 2009 [42]. In each instance, humankind should have been able to learn something regarding infection countermeasures. This issue will be revisited in the future. How did Japan learn from the novel influenza A (H1N1) pdm 09 in 2009? There were already major changes in people’s behavior with COVID-19. We have also found from epidemiological data that Japanese seems to have certain type of altruism. Further analysis and enlightenment of this aspect might be useful to the future vaccination strategy. Once the current COVID-19 pandemic has settled, a new strategy for vaccine inoculation is needed. We must learn from the past and share the wisdom of preparing for the next pandemic at the national and regional levels.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

The authors do not have permission to share data.