In Japan, the human papillomavirus (HPV) nonavalent vaccine was approved by the Pharmaceuticals and Medical Devices Agency (PMDA) for girls aged 9 years or older in July 2020 [1]. The corresponding quadrivalent inoculation was ratified for boys in the same age group in December 2020 [1]. Nevertheless, the level of awareness regarding HPV, the resultant cancer (cervical cancer [CC]), and the vaccine needs to be higher in the Japanese population [2,3]. Although vaccination is the exclusive means of preventing HPV infection, the immunization rate has fallen below 1% owing to subsequent adverse events [4], which are regarded as functional disorders. These cases were reported repeatedly in Japanese media in sensational ways [2,4]. As a result of the dissemination of misinformation and the misunderstanding of the HPV vaccine, most Japanese people have distrust toward the HPV vaccine [2]. Thus, the immunization rate of the bivalent or quadrivalent HPV vaccine for the target population from the 6th grade of elementary school students to the 1st grade of high school students was 0.8% in 2018 [5]. This has persisted from 2013 for almost 8 years [4-6].

According to the strategy of the World Health Organization devised in 2019 for controlling CC [7], by 2030, 90% of girls worldwide would be vaccinated with the HPV vaccine by the age of 15 years. Malaysia, Mexico, Bhutan, Brunei, and Rwanda have achieved an immunization rate of 90% or higher in the target population [7]. Furthermore, in Australia, the success of two national programs, the National Cervical Screening Program (NCSP) and the Australian National HPV Vaccination Program (NHVP), resulted in achievement of the threshold for rare cancer in 2020 [8]. By 2028, the estimated number of cases would be less than 4 per 100,000 [9], which is the threshold for elimination [8].

In addition to HPV-related cancers, vulvar, anal, and throat malignancies may be prevented by the vaccine [9,10]. A Japanese research group provided evidence of a significant reduction in the incidence of cervical intraepithelial neoplasia grade 3 or higher in vaccinated women [11].

Specific information regarding the effects and adverse events of HPV vaccines needs to be fully disseminated to the Japanese population. Considering this, approximately 40% of the population is willing to be vaccinated [2]. Therefore, it is critical to disseminate adequate scientific knowledge regarding the beneficial effects of the vaccine so that people in the target age group actually take the vaccine.

Vaccine awareness programs are necessary, and campaigns through the media and social network services can play significant roles [2,4]. In an information-overloaded society, people frequently make decisions related to health issues based on a bunch of information [12]. Therefore, it is critical to consider the influence of behavioral insights to promote change [13]. This broadly refers to concrete approaches based on the knowledge of behavioral science and economics. The Easy, Attractive, Social, and Timely (EAST) principles are a simple way of applying behavioral insights to interventions and have been used to change human awareness and behavior [13]. This study aimed to assess the effects of these behavioral insights utilizing brief scientific information on vaccine benefits, along with statistics on CC.

We recruited a total of 1660 participants in March 2018 via a specially designed webpage for this study. These were registered members of the research panel owned by Macromill Inc (Tokyo, Japan). The participants were 20 years old or above as on March 12-13, 2018. They were recruited until the target sample size was fulfilled. We randomly assigned each participant to respond to an identical questionnaire after (intervention group) or prior to (control group) providing behavioral insights material (BI-material) featuring brief scientific information presented in an easy-to-read format (as displayed in Figures 1 and 2).

Informational materials are designed primarily to increase effectiveness; they are based on a specific purpose rather than a template. In this study, first, we provided one-page BI-material containing three sentences, which were deemed to be in line with scientific evidence and objective facts associated with the theory of behavioral change and economics. This theory was suggested and represented by Richard Thaler [14]. The first sentence of our material reflected the Japanese epidemiological data on CC (Figure 2), which was intended for the framing effect [15,16]. For example, a statement like “Cervical cancer is the 11th most frequent malignancy among women in Japan” or “2.8% of women diagnosed with cervical cancer die” does not emphasize the deterioration in health. Therefore, this would not contribute to general behavior change. Additionally, we applied the prospect theory by drawing similarities between the number of deaths due to CC in Japan and fatalities due to traffic accidents. Second, to promote a change in the consciousness of participants, the presentation of social norms was included as part of the EAST framework (S-social) in the second sentence of the BI-material (Figure 2). This was employed for devising behavioral insights of greater effectiveness [17]. Third, heuristics (awareness of prejudice and intuition) [18] hypothesized a reduced interest in male participants in case the HPV immunization focused exclusively on CC. Therefore, it was conveyed that non-CC HPV malignancies are preventable through HPV vaccination [9]. The first author (YS) is one of the specialists of the behavioral design team in Yokohama, which was established as the first nongovernmental nudge unit in Japan [19]. Thus, this BI-material was created given the behavioral insights methodology.

Participants were randomly allocated (1:1) to each group. The intervention was performed using an automatic web-based allocation system stratified by the sex (female/women and male/men) and age (20s, 30s, 40s, 50s, and 60s) of the participants. Randomization was performed by the web-research system of Macromill Inc. Participants and investigators were blinded to the distribution (double-blinded). Once the upper limit of each stratum was reached, new participants could not be added to the web system. This ensured uniform distribution of the stratification factors. In the intervention group, we provided the BI-material prior to answering questions related to preventive awareness, following consent for the online study. The control group was provided with the same material as that provided to the intervention group after all responses were completed.

The participant demographics included age, sex, marital status, number of children, sexual experience, household income, medical background, education, and tobacco use. For female participants, we also collected information regarding HPV vaccination history and previous pap screening. Medical background was defined as follows: a respondent who had a history of working as a medical professional (medical doctor, dentist, nurse, or allied health care professional) or who attended medical school.

The respondents completed a 10-item questionnaire. The first half determined HPV awareness as background information, and the second half addressed willingness to receive HPV vaccines and undergo screening tests. The respondents were instructed to answer “Yes” or “No” for each question.

The awareness questions were as follows:

Q1. It is possible to detect both cancer and precancerous lesions through CC screening.

Q2. Sexual experience is associated with HPV infection.

Q3. CC screening is necessary for women even after vaccination.

Q4. I have heard of the benefits of the HPV vaccine.

Q5. I have heard of the adverse events associated with the HPV vaccine.

The willingness questions were as follows:

Q6. If you have/had a daughter, do/would you consider getting her vaccinated against HPV?

Q7. If you have/had a son, do/would you consider getting him vaccinated against HPV?

Q8. Would you consider undergoing a pap smear? If male, will you want your family or partner to undergo a smear test?

Q9. Would you undergo the cancer screening tests recommended by the government?

Q10. Do you plan informing family members, friends, or others about cancer prevention and screening (through Facebook, LINE, Twitter, etc)?

The study protocol was approved by the Institutional Research Ethics Committee of Yokohama City University School of Medicine (A180200004). The trial registration number is UMIN000049745. We received research funding from the Japan Agency for Medical Research and Development (grant number 15ck0106103h0102). The survey was outsourced to Macromill Inc.

From March 12 to 13, 2018, 1660 participants were recruited. Stratifying factors, such as sex and age, were evenly allocated. The following variables displayed no significant differences between the intervention and control groups: marital status (P=.96), children (P=.84), household income (P=.58), sexual experience (P=.26), education (P=.44), medical background (P=.50), and tobacco use (P=.64) (Table 1).

For questions Q1 to Q5 regarding HPV awareness, the recognition rate for Q2 was significantly higher (by 7.8%) in the control group than in the intervention group. For questions Q4 and Q5, the values in the intervention group were significantly higher (by 6.5% and 6.6%, respectively) than those in the control group (Table 1).

Only 484 (29.2%) of the respondents were aware of the benefits of HPV immunization (Q4), whereas 543 (32.7%) were aware of the adverse effects (Q5).

Only 352 (21.2%) of the respondents displayed a favorable attitude toward HPV immunization for their daughters (Q6). However, an additional 40 (4.8%) participants responded affirmatively in the intervention group (aOR 1.32, 95% CI 1.04-1.69) compared to those in the control group (Table 2). For Q7, there were additional 33 (3.9%) satisfied respondents willing to consider vaccination for their sons in the intervention group (aOR 1.38, 95% CI 1.05-1.80) compared to those in the control group (Table 2).

The number of respondents expressing a desire to receive a screening test (Q8) (P=.37) or to communicate HPV-related issues (Q10) did not increase (P=.17) (Table 2).

Table 3 presents the subanalysis results according to sex.

Differences were identified in Q6 (men: aOR 1.46, 95% CI 1.05-2.02 vs women: aOR 1.20, 95% CI 0.83-1.73) and Q7 (men: aOR 1.53, 95% CI 1.08-2.18 vs women: aOR 1.21, 95% CI 0.80-1.83). The willingness to consider vaccination for daughters in men was significantly higher in the intervention group (by 8.2%, P=.02; Multimedia Appendix 1); however, such a difference was not observed in women (P=.22; Table 3, Multimedia Appendix 2).

In an overall comparison between men and women irrespective of intervention, the willingness to consider vaccination for daughters in men was significantly higher than that in women (25.1% vs 17.3%, P<.001), and the willingness to consider vaccination for sons was also higher in men than in women (20.1% vs 12.8%, P<.001).

In the intervention group, higher rates were identified in men than in women for Q6 (29.2% vs 18.1%, P<.001) and Q7 (23.9% vs 13.0%, P<.001). While in the control group, differences were not identified between men and women for Q6 (21.0% vs 16.6%, P=.11) and Q7 (16.4% vs 12.5%, P=.11).

We conducted a web-based randomized controlled trial (RCT) to assess the benefits of BI-material employing brief scientific information and its ability to motivate individuals to consider the HPV vaccine for their children. Our results showed that providing brief scientific information could increase the willingness to consider HPV vaccination for daughters and sons. This effect was observed typically among male participants. Similar minor interventions may potentially modify mindsets favorably. However, such brief digital information failed to affect the mindset in women. A possible reason why the intervention was more effective among men than women is that women had a more negative image toward HPV vaccination. There was a significant difference between men and women in the awareness level. Overall, 45.7% of women responded that they know about the adverse events of HPV vaccination, while this rate was 19.8% in men (data not shown). Such awareness might have influenced the difference in the intervention effect.

In terms of the COVID-19 pandemic, a difference in awareness of prevention strategies was observed between men and women [20]; therefore, it is essential to build a method appropriate for sex subgroups to transform general behavior via the internet and social networking services.

According to a systematic review regarding the effect of the pedagogical approach on social awareness or action, more reliable and validated studies are required to change the perception or mindset of the target population [21]. This paves the way for mitigating hesitancy toward vaccination [21]. Therefore, this RCT may be valuable to ensure a change in public attitudes toward vaccination.

A systematic review revealed that general communication about childhood vaccination resulted in a positive change by 20%; however, this excluded the HPV vaccine [21]. In contrast, the brief educational intervention was observed to improve the willingness to consider the HPV vaccine for daughters and sons, with aORs of 1.32 and 1.38, respectively (Table 2). Therefore, when compared to the review data of the in-person approach [21], the aORs were not small. Furthermore, the BI-material was designed by a specialist of behavioral insight, resulting in a moderate cost; therefore, this intervention is more reliable and cost-effective than those used in previous studies.

A US RCT reported that education through social media is effective for improving general awareness regarding vaccination. However, the study was limited to a sample of 58 participants [22]. A phase 2 Japanese trial evaluated how the extent of intervention affected HPV vaccination acceptance and reported that providing appropriate medical information resulted in beneficial effects [23]. Thus, this study proved the utility of effective and adequate guidance in improving vaccine acceptance. There is a paucity of literature on the efficacy of pragmatic educational materials for HPV vaccination promotion [22]. Therefore, more extensive trials focusing on the mode of endorsing HPV vaccination like our trial are required for evidence-based promotion.

We recognize several limitations in this study. First, the sustainability of effective change was not evaluated. Typically, with respect to health issues, taking action requires time. Therefore, a study should assess not only a change in mindset, but also the appropriate course of action. We have already performed an RCT (UMIN000039273) assessing the sustainability of general acceptance and concrete behavior for HPV vaccination. Second, the impact of the study on the behavioral outcome was unclear; specific vaccination functioning needs to be tracked. Third, selection bias was present as the respondents were Japanese individuals enrolled by an internet survey company.

Based on the sex-wise subanalysis, improved information in male participants may be the key to improve the rate of HPV vaccination in Japan. Additionally, video-based content with patient feedback is expected to result in a broader impact. Thus, a varied approach for men and women may be required.