Improved parental attitudes and beliefs through stepwise perinatal vaccination education

ABSTRACT

This study examined the effects of providing vaccination education during the perinatal period on Japanese parents' knowledge, attitudes, and beliefs about childhood vaccination. A cluster-randomized controlled-trial method was used on a sample of 160 pregnant women recruited from 9 obstetrical sites in Niigata, Japan. The treatment group received a stepwise interactive education intervention, while the control group received a general vaccination leaflet. Changes in parental attitudes toward and beliefs about infant vaccination were assessed on the child's one-month and 6-month birthdays using paper questionnaires. Of the initial 188 participants, 160 (90.4%) completed the final post-survey questionnaire. Scores on injunctive social norms (a morally neutral perception of the behavior of the majority) and descriptive social norms (a moral perception of what individuals should do) significantly increased in the treatment group (p = .02 and p = .01, respectively). There was a significant difference between the 2 groups over time in terms of perceived benefit (efficacy of available preventive actions) (p = .03), but no significant differences in perceived severity (seriousness of a disease outcome), perceived susceptibility (likelihood of getting a disease), perceived benefits, perceived behavioral control, or descriptive social norms between the groups at any time point or in the patterns of change over time (p > .31). Thus, stepwise perinatal vaccination education was found to positively influence maternal attitudes and beliefs about infant vaccination. This study suggests the importance of vaccination education during the perinatal period.

KEYWORDS:

• Cluster-randomized controlled trial
• infant vaccination status
• maternal attitudes and beliefs
• perinatal vaccination education

Introduction

The environment surrounding vaccination policy in Japan has been subject to remarkable changes since 2008,¹ and the availability of vaccines has increased considerably. However, these changes in Japan's vaccination system have increased the level of confusion among providers and recipients, who need to know how and when to deliver and receive vaccines in a timely manner. Additionally, because of the lack of a unified vaccination education system, providers vary in the vaccination education that they offer, for which reason Japanese parents do not receive standardized information on vaccination and, in many cases, do not have a clear sense of their children's need to receive vaccines at recommended times. An ideal educational approach would provide optimal and timely education on these points, while taking maximum advantage of the limited opportunities to do so. To this end, further improvements are required to achieve a standardized vaccination education program in Japan with optimal timing and content that encourages appropriate infant vaccination.

In Japan, it is parents who make the decision regarding whether or not to immunize their infants. Many studies have found that the main factors influencing parents' decisions to vaccinate their children strongly relate to their attitudes and beliefs about vaccine-preventable diseases (VPDs). When their attitudes and beliefs discourage vaccination, parents might delay or reduce vaccine uptake because they are skeptical about its effectiveness,² and react instead to adverse opinions and safety concerns. To increase vaccination use, it is first necessary to understand parents' beliefs about vaccination and what motivates them to immunize their children or not, and, second, to deliver effective and adequate information and explanations about vaccination.

Perceptions of risk include factors that influence decision-making regarding vaccinations, such as perceived threats.³ In this regard, the most common concerns are the risks of vaccination, overstimulation of the immune system, and receiving multiple injections in one visit.⁴ Fears and concerns regarding infant vaccination contribute to scheduling delays⁵ as parents with concerns about vaccine safety and efficacy are more likely to delay vaccines.⁶ One study reported that parents of unvaccinated children were more likely than parents of fully vaccinated children to perceive vaccines as having low safety and efficacy.⁷ In this context, it is important to understand how parents' attitudes and beliefs might change in response to educational interventions in relation to vaccination. The objective of this study was, therefore, to identify the effects of a periodic education intervention on Japanese parents' knowledge, attitudes, and beliefs about childhood vaccination.

Results

Study participants

A flowchart of participant entry into and progress during the study is shown in Fig. 2. Five hospitals and 4 clinics agreed to participate in this study. We approached 490 pregnant women, of whom 188 (38.3%) agreed to participate in the study at one of the 9 sites: 81 (43.1%) at hospitals and 107 (56.9%) at private clinics. Altogether, post-survey questionnaires were returned (response rate = 90.4%) (Table 2).

Attitudes and beliefs at baseline and at child's one-month birthday

Attitude scores at baseline and at child's one-month birthday were compared between the treatment and control groups. Scores on injunctive social norms and descriptive social norms were significantly higher at one month in the treatment group (p = .02 and p = .01, respectively), but not in the control group (p = .11 and p = .43, respectively) (Table 3). Perceived benefits and perceived behavioral control were significantly different in both groups (p = .001 in the treatment group and p < .03 in the control group). There were no significant differences for either group between the baseline and the child's one-month birthday in perceived severity, perceived susceptibility, or perceived barriers (p > .09).

Attitudes and beliefs at baseline, child's one-month birthday, and child's 6-month birthday

When we analyzed the total score of each subscale's item, no significant differences were found in any of the constructs (perceived benefit, perceived severity, perceived susceptibility, perceived behavioral control, and descriptive social norms) at any point or in the pattern of change over time (p > .06). However, when we separately analyzed the subscales' items, a difference was found between the 2 groups over time on the perceived benefit item “vaccines are safe for children” (p = .03) (not shown). This difference regarding perceived barriers and injunctive social norms was found at the one-month follow-up survey in the treatment group compared with the control group (p = .04 and p = .02, respectively) (Table 4).

Vaccination status

We already have presented the following results in the previous report.⁸ Although the percentages of children who completed 3 doses of Hib vaccine and PCV13 at 6 months of age did not differ significantly between the intervention (98.6% and 97.8%, respectively) and control (92.9% and 92.9%, respectively) groups, (p = .67 and p = .20, respectively), the percentage of children who completed 3 doses of IPV-DTaP vaccine at age 6 months was significantly higher in the intervention (97.3%) than in the control (88.4%) groups (p = .04). Additionally, mean duration to completion of the third dose of the Hib, PCV13, and IPV-DTaP vaccines was shorter in the intervention than in the control group: Hib vaccine, 128 vs. 133 days; PCV13, 128 vs. 133 days; and DPT-IPV vaccine, 159 vs. 164 days, respectively (p < .001, p < .002, and p = .01, respectively).⁸

Discussion and conclusions

The results indicated that stepwise maternal vaccination education increased parents' perceptions of benefit, increased the effect of social norms, and decreased the perceptions of barriers to infant vaccination practice. The use of educational content targeted at the perinatal stage to match maternal needs was thus found to be helpful to mothers in Japan, where vaccination education is not routinely or systematically delivered.

Regarding the questionnaire items on attitudes and beliefs, the percentages of participants who answered “yes” to the statement “vaccines are safe for children” differed significantly between the treatment and control groups. Furthermore, participants who stated that vaccines were safe were more likely to vaccinate their infants at the appropriate times. This result is consistent with previous findings that parents concerned about vaccine safety and efficacy were relatively more likely to delay vaccination.⁶ Other studies have reported that parents of unvaccinated children are more likely than parents of fully vaccinated children to report that vaccines are not safe and effective.⁷ The findings of this study thus emphasize the importance of vaccine education, particularly regarding the provision of information on VPDs and vaccine safety.

We found that injunctive and descriptive social norm scores in the treatment group had increased at the child's one-month birthday. This result implies that social norms play a key role in influencing mothers' attitudes by sensitizing them to the behaviors expected of them as parents and socializing them to care about what other parents are doing. Previous studies have found that social norms are important to parents' vaccination decision-making.^9-11 Our intervention included educational content that promoted opportunities to discuss infant vaccination in the family, including with participants' spouses and parents, because mothers' decisions on vaccination are often influenced by the attitudes of people who are important to them.¹⁰ We further found that scores on perceived barriers significantly decreased in the treatment group compared with the control group after the final (6-month) treatment. As described in the Introduction, previous studies have suggested that parents' hesitation to immunize their children is a major determinant of delayed vaccination^12,13 due to fears over infant vaccination. The content of the tested education program was thus designed to fulfill mothers' needs at each perinatal period and to address these concerns. We believe that the educational content helped to lessen the participating mothers' fears, along with the treatment structure, by providing numerous education opportunities that emphasized the importance of immunizing children.

However, several factors might have limited the effectiveness of the treatment, and these findings should therefore be interpreted in light of these concerns. First, it is possible that the study participants obtained information on vaccination between treatments from other sources, which would mean that our treatment was less effective than our findings indicate because only some of the effects would have been caused by our treatment. Second, some participants may have participated in the study as they were particularly interested in vaccination, which could have biased the results. Third, during the study period, the Japanese media sensationally covered a story about the human papilloma virus (HPV) vaccine. This story increased public awareness of the safety of vaccination and may have influenced some of the participants' decisions to participate as well as some of their questionnaire responses.¹⁴ Fourth, the generalizability of the study may be limited as the participants were from the same region of the country. Fifth, although our findings showed that all required vaccines, including the Hib vaccine, PCV13, and the IPV-DTP vaccine, were given earlier in the intervention group compared with the timing in the control group,⁸ additional studies are necessary to clarify the improvement in actual vaccination behavior as it relates to improved vaccination rates. Last, the data were self-reported, and we were unable to verify the clinical practice of the vaccination education. Therefore, response bias may have influenced the validity of the results.

In conclusion, the results of this study have found that stepwise perinatal vaccination education positively influenced maternal attitudes and beliefs regarding infant vaccination. Our findings thus support the importance of vaccination education during the perinatal period. Future efforts should develop a systematic vaccination education system that meets mothers' needs during the perinatal period to improve infant vaccination rates.

Methods

Study design and setting

The study sites comprised 9 obstetrical hospitals/clinics in the metropolitan area of Niigata (population: approx. 800,000). The participating institutions were Niigata University Hospital, its 4 affiliated hospitals, and 4 private obstetrics clinics. The annual number of deliveries at each hospital or clinic ranged between approximately 100 and 800. Study participants were recruited to participate over an approximately 4.5-month period from September 15, 2014, to January 31, 2015. This cluster-randomized controlled trial investigated the effectiveness of an infant vaccination education program during the pregnancy and postpartum periods.

Eligibility and enrollment

Pregnant women aged 18 y or older in gestational weeks 24 through 30 who could communicate in Japanese were recruited at antenatal classes by a midwife or obstetrician or at their prenatal checkups at the sites. During recruitment, expectant mothers with cognitive impairments or those for whom the study was deemed a physical or mental burden (determined by medical professionals at each site) were excluded. Demographic information collected from the participants included employment status, infant's date of birth, mother's education level, number of children, and family structure. After signing an informed consent form, participants completed a self-administered handwritten baseline questionnaire at home, which took approximately 10 minutes. The participants were instructed not to review any materials while completing the questionnaire, which occurred between September 15, 2014, and February 28 of 2015.

The 9 participating institutions were organized into 5 strata (4 pairs and one unmatched hospital) according to their characteristics, and the randomization procedure assigned each institution to the treatment or control group for each stratum. Because the treatment comprised an education program, blinding the study staff and participants was not possible once the groups had been assigned. Sample size was calculated based on the vaccination rate of the previous report.⁸

Ethical considerations

This study was approved by the Institutional Review Boards of the University of Tokyo (#10384), Niigata University Hospital (#1882), and Saiseikai-Niigata-Daini Hospital (#E14–02).

Treatment

Before applying the treatment, which was performed by clinical educators at the participating sites, we provided the educators with the study protocol and guidance on the intervention procedures. Vaccination education was provided to the treatment group on 3 occasions: (1) during the prenatal period (gestational weeks 34–36) in an outpatient setting, (2) during the postpartum period (3–6 d after delivery) in an inpatient ward, and (3) at a one-month well-baby check-up in an outpatient setting. At each treatment session, one educator covered specific subjects focusing on the content expected to be of greatest interest to parents at each perinatal stage, based on the results of a pilot study.⁸

The content of the treatment was developed by combining parents' needs with Health Belief Model goals, such as increasing parents' recognition of VPD severity and susceptibility, conveying the benefits of vaccination, and reducing perceived barriers, such as anxiety about vaccination. We developed 3 educational leaflets corresponding to and reflecting the content of each of the treatments and these leaflets were used in all sessions. Each session presented a scenario with relevant questions and answers, developed in advance and lasting approximately 5 minutes.

The control group received written information in an educational pamphlet, which was more extensive than the leaflets provided to the treatment group, but without oral explanation. The pamphlet covered a wide range of subjects related to vaccination and was designed as a supplement to the basic vaccination information provided at routine check-up visits.

Post-treatment surveys were sent to all participants at 2 time points, i.e., when their children were approximately one month and 6 months old. The participants were asked not to review any educational or informational materials before completing the questionnaire, and they each received a JPY 1000 (approximately USD 9) gift card as an incentive to participate.

Conceptual Framework

Previous studies have used a variety of behavioral theories to explain parental behaviors related to vaccination. Becker's Health Belief Model (HBM) is the most widely used socio-cognitive theory used to explain parental behavior related to vaccination.⁸ The HBM includes the following: (1) perceived susceptibility (likelihood of getting a disease), (2) perceived severity (seriousness of disease outcome), (3) perceived benefit (efficacy of available preventive actions), and (4) perceived barriers (pain, side effects, or inconvenience of preventive action).^{8, 9} The HBM is an appropriate theory to explain vaccination behaviors and might be the most appropriate lens through which to observe the behaviors of parents regarding their children.^10-12 An early application of the HBM aimed to learn about barriers to polio vaccination and ways to increase vaccination coverage.⁶

Evidence from previous studies suggests that behavioral changes and the decision to vaccinate one's children relate to perceptions of increased threat when parents recognize the severity and susceptibility of VPDs and/or perceive that the benefits of vaccination outweigh the risks, such as side effects. The present study adopted educational content on vaccination that specifically targeted the perceived threat of disease (susceptibility and severity), the difference between the perceived virulence of the disease and the perceived efficacy of the vaccination, and knowledge of preventative vaccinations (Fig. 1).

Figure 1. Conceptual framework. Abbreviation: HBM, Health Belief Model; IBM, Integrated Behavioral Model.

Figure 2. Flowchart of participant entry and progress in the study.

The HBM and Integrated Behavioral Model (IBM)¹³ were used to develop the educational materials and determine their content. The materials incorporated the constructs of “behavior control” and “injunctive social norms” (a morally neutral perception of the behavior of the majority) and “descriptive social norms” (a moral perception of what individuals should do), which might influence parents' intentions to vaccinate their children.¹⁴

Maternal attitudes and beliefs

Psychosocial factors were assessed using responses to questionnaire¹⁵ items based on the HBM¹⁶ and IBM,¹⁷ which were used in previous studies.¹⁸ Based on an initial investigation of the reliability and validity of the psychosocial measures, some questionnaire items were modified to reflect the current vaccination situation in Japan. Attitudes and beliefs about VPDs and vaccination assessed by the HBM¹⁶ and IBM¹⁷ include perceived severity, perceived susceptibility, perceived benefits, perceived barriers, injunctive social norms, descriptive social norms, and perceived behavioral controls. All of the items provided response options on a 5-point Likert-type scale ranging from 1 = strongly disagree to 5 = strongly agree (Table 1). The range of each construct reflected the number of items in that construct. For example, “perceived barriers” comprised responses to 6 questions to give a range from 6 to 30.

Table 1. Attitudes and beliefs regarding vaccine-preventable diseases (VPDs) and immunization.

Statement/Question	Possible Responses
• Perceived severity
1. VPDs are serious diseases.
2. VPDs are serious diseases for babies.
• Perceived susceptibility
3. My baby is not very likely to get VPDs.
• Perceived benefits
4. Vaccines for babies will prevent my baby from getting sick with VPDs.
5. If my baby receives his/her vaccines, it will help protect my friends and family from getting VPDs.
6. The vaccine will prevent my baby from missing daycare because of VPDs.
7. If my baby receives his/her vaccines, it will prevent us (parents) from missing work to care for our baby.	1. Strongly disagree
8. Vaccines are safe.	2. Disagree
• Perceived barriers	3. Neither disagree nor agree
9. I feel uncomfortable because vaccines are painful or uncomfortable for my baby.	4. Agree
10. Vaccines are too expensive.	5. Strongly agree
11. It is better for my child to develop immunity by getting sick than by getting a shot.
12. Children receive too many vaccines.
13. I am concerned that my child's immune system could be weakened by too many vaccines.
14. I am concerned that my child might have side effects from a vaccination.
• Perceived behavioral control
15. I have control over whether or not my baby gets vaccines.
• Social norm (injunctive)
16. Most people important to me think I should have my baby vaccinated.
17. My parents think my baby should be vaccinated.
18. My friends think my baby should be vaccinated.
• Social norm (descriptive)
19. I know other people my age who had their babies vaccinated.
20. Most of my friends had their babies vaccinated.

Abbreviation: VPDs, vaccine-preventable diseases.

Table 2. Baseline characteristics of study participants.

	Intervention group		Control group
	ITT (n = 100)	Per-protocol (n = 88)	ITT (n = 88)	Per-protocol (n = 72)
Hospital, n	2	2	3	3
Clinic, n	2	2	2	2
Participants, n	100	88	88	72
Age (mean ± SD)	31.9 ± 4.4	31.6 ± 4.4	31.5 ± 5.1	32.3 ± 4.9
Education level
Middle/high school	27(27.0)	25(28.4)	20(22.7)	14(19.4)
Junior college	43(43.0)	39(44.3)	51(58.0)	43(59.8)
College/graduate school	30(30.0)	24(27.2)	17(19.3)	15(20.8)
Number of siblings
0	48(48.0)	45(51.3)	40(45.5)	32(44.4)
1	35(35.0)	30(34.1)	30(34.1)	24(33.3)
≥ 2	17(17.0)	13(14.8)	18(20.5)	16(22.2)
Maternal employment status
Unemployed	73(73.0)	71(80.7)	60(68.2)	57(79.2)
Employed	27(27.0)	17(19.3)	28(31.8)	15(20.8)
Household annual income (thousands of yen)
0–3000		10(11.4)		6(8.3)
3,000–4,999		38(43.2)		32(44.4)
5,000–6,999	N/A	23(26.1)	N/A	25(34.7)
7,000–9,999		11(12.5)		6(8.3)
≥ 10,000		6(6.8)		2(2.7)

Numbers in parentheses show percentages for each group.

Abbreviations: ITT, intention to treat; SD, standard deviation; N/A, not available.

Table 3. Scores for attitudes and beliefs at baseline and 1 month in the intervention and control groups.

	Intervention (n = 100)			Control (n = 88)
	Baseline	1 month	p-value^a	Baseline	1 month	P-value^a
Attitudes and beliefs
Perceived severity (HBM)	7.9 ± 1.6	8.3 ± 1.6	.09	7.9 ± 1.9	7.8 ± 1.5	.69
Perceived susceptibility (HBM)	1.9 ± 1.0	1.9 ± 1.1	.59	1.8 ± 1.0	1.9 ± 1.1	.85
Perceived benefit (HBM)	16.6 ± 5.1	17.7 ± 3.9	.001^*	16.0 ± 4.9	17.2 ± 4.5	.03^*
Perceived barriers (HBM)	18.3 ± 5.7	17.8 ± 3.6	.77	17.8 ± 4.5	18.8 ± 3.8	.15
Perceived behavioral control (IBM)	3.0 ± 1.2	3.6 ± 1.3	.001^*	3.1 ± 1.0	3.7 ± 1.2	.02^*
Social norm (injunctive)	12.0 ± 2.6	12.7 ± 2.3	.02^*	11.9 ± 2.9	12.2 ± 1.6	.11
Social norm (descriptive)	8.3 ± 1.6	9.2 ± 1.5	.01^*	8.3 ± 1.8	8.6 ± 1.2	.43

The data are means ± standard deviations.

Abbreviations: HBM: Health Belief Model; IBM: Integrated Behavioral Model.

^a One-sample Wilcoxon rank-sum test.

* P < .01

Table 4. Scores for knowledge, attitudes, and beliefs at baseline, 1 month, and 6 months of age between the intervention and control groups.

Outcome	Score	Intervention		Control
Timing	Range	Mean	SD	Mean	SD	P-Value	ES
Perceived severity (Pgroup × time = 0.88)	2–10
Baseline		7.9	1.6	7.9	1.9	.97	0
1 month		8.3	1.6	7.8	1.5	.06	0.32
6 months		8.3	1.8	8.3	1.6	.79	0
Perceived susceptibility (Pgroup × time = 0.47)	1–5
Baseline		1.9	1	1.8	1	.16	0.1
1 month		1.9	1.1	1.9	1.1	.59	0
6 months		1.9	1	2.1	1.1	.08	0.76
Perceived benefit (Pgroup × time = 0.31)	5–25
Baseline		16.5	5.1	16	4.9	.88	0.1
1 month		17.7	3.9	17.2	4.5	.59	0.12
6 months		17.6	4.1	17.6	4.1	.83	0
Perceived barriers (Pgroup × time = 0.46)	6–30
Baseline		18.3	5.7	17.8	4.5	.81	0.09
1 month		17.8	3.6	18.8	3.8	.04*	0.27
6 months		18.4	4	19.1	3.7	.28	0.18
Perceived behavioral control (Pgroup × time = 0.69)	1–5
Baseline		3	1.2	3.1	1	.47	0.09
1 month		3.6	1.3	3.7	1.2	.95	0.08
6 months		3.7	1.2	3.6	1.3	.33	0.08
Injunctive social norm (Pgroup × time = 0.94)	3–15
Baseline		12	2.6	11.9	2.9	.26	0.03
1 month		13.7	1.9	12.2	1.6	.02*	0.86
6 months		12.9	2.1	12.3	1.9	.09	0.3
Descriptive social norm (Pgroup × time = 0.92)	2–10
Baseline		8.3	1.6	8.3	1.8	.73	0
1 month		8.8	1.3	8.6	1.2	.54	0.24
6 months		8.9	1.3	8.8	1.3	.62	0.23

Abbreviation: SD, standard deviation; ES, Effect Size

Data analysis

The analyses were performed using SPSS version 19.0 (Chicago, IL, USA). Descriptive statistics were assessed for the distributions of the background and outcome variables. Fisher's exact test was used to perform bivariate analyses of data distributions and associations between variables. Continuous variables were analyzed in a hierarchical linear model with fixed effects of group, time, and interaction of group by time to test patterns of change over time. Participants were included as random effects to account for covariance attributable to repeated measures. The Wilcoxon rank-sum test was used to compare the mean scores to determine secondary outcomes in the treatment and control groups. All tests were 2-tailed and the results were evaluated at a significance level of .05.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Funding source

This study was supported by a grant from St. Luke's Life Science Institute.

Financial disclosure

None of the authors has a financial or personal relationship to disclose that could inappropriately influence this study.

Acknowledgments

We are grateful to all of the study participants. We are also grateful to Toshio Yagimoto at Yagimoto Pediatric Clinic, Takeshi Hirohashi at Hirohashi Obstetric Clinic, Akiteru Tokunaga at Tokunaga Jyo-sei Clinic, Masato Arakawa at Arakawa Ladies' clinic, Akira Honda at Honda Ladies' clinic, Isao Takeyama at Takeyama Hospital, Takumi Kurabayashi and Yoshihisa Nagayama at Niigata City Hospital, Isao Hasegawa at Saiseikai-Niigata-Daini Hospital, Shinya Watanabe at Kameda-Daiichi Hospital, and Koichi Takakuwa and Takayuki Enomoto at Niigata University Hospital for recruiting the study participants. We also thank Fukiko Sato and Setsuko Mitome at Niigata University Hospital, Fuyuko Watanabe at Niigata City Hospital, Yoko Oda at Takeyama Hospital, and Chikako Shimizu at Saiseikai-Niigata-Daini Hospital for their valuable assistance. Finally, we thank David Kipler for editing the manuscript.