I Introduction

Vaccination is the key infection control measure against Coronavirus Disease 2019 (COVID-19) [1]. However, not all targeted individuals have undergone vaccination, with some refraining due to apprehensions about side effects, while others are deterred by the intricacies in procedural complexities, even though they might be willing to receive the vaccine [2, 3]. A systematic review on COVID-19 vaccine promotion reported several methods, but very few measured real behavioral outcomes [4]. Among the interventions tested in real-world settings, one of the most effective was changing the default nudge, defined as the pre-selected option automatically applied when the chooser takes no active decision [5–9]. Indeed, individuals with an opt-out default (initially scheduled for vaccination, requiring contact only if inconvenient) exhibited a 32% relative increase in vaccination rates (13.1% vs. 9.9%) compared to those in the opt-in system (where participants adjust and schedule their vaccination) [5]. However, this study was conducted in Italy, a country more tolerant of default changes and without significant vaccine hesitancy in international comparison and focusing on a target age group of 50–59 years [5, 10, 11]. Thanks to the favorable conditions, the results may have been predisposed to positive results. For the opt-out approach in COVID-19 vaccination to gain wider acceptance, data from countries with lower vaccine acceptance and resistance to default changes, encompassing a broader age range, would be preferable, as is typical in countries like Japan [11, 12]. Additionally, the study did not address potential inhibiting factors associated with the implementation of the opt-out system, such as increased cancellation rates on the day, discomfort among individuals regarding the change in defaults, or the possibility that the opt-out effect may be a one-time occurrence [5, 13–15].

In this study, under the hypothesis that differences in the default options for COVID-19 vaccination systems influence vaccination attitudes even in countries conservative about vaccination and default changes, we aimed to examine the differences in first and second dose vaccination rates, cancellation rates, and the number of complaints between the opt-in and opt-out systems for COVID-19 vaccination.

II Methods

III Results

All ten cities responded, and one city reported not knowing the vaccination coverage, so the remaining nine cities were included in the analysis. Table 1 shows the vaccination coverage in each city.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Vaccination coverage rate.

https://doi.org/10.1371/journal.pone.0298983.t001

The opt-out group had a 95% confidence interval of 88.2%-89.2% for the first and 84.9%-86.0% for the second dose, compared to 74.1%–80.1% for the first and 55.7%–67.5% for the second dose for the opt-in group, both of which were higher in the opt-out group. Specifically, the data showed that in the opt-in group, there was a decrease of 11.1–23.6 percentage points in the vaccination rate for the second dose compared to the first, whereas the opt-out group experienced a much smaller reduction, limited to just 3.2 percentage points.

Table 2 shows the no-show and cancellation rate on the day in City B, as other cities did not record such information. The data shows that for the first dose, the vaccination rates in the opt-out group, the rates varied from 7.3% to 17.7% averaging 11.3% overall. For the second dose, the opt-out group showed an increase, ranging from 9.4% to 22.6%, with an overall average of 14.6%.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. No-show and cancellation rate on the day of the vaccination.

https://doi.org/10.1371/journal.pone.0298983.t002

Also, in the opt-out group, no complaints were received from residents about the implementation of the opt-out system.

IV Discussion

Our study revealed that in all comparable age groups between 20 and 64 years, COVID-19 vaccination rates were higher in the opt-out group than in the opt-in group, as indicated within the 95% confidence interval. This finding supports our hypothesis that "even in countries conservative about vaccination and default changes, differences in the default options for COVID-19 vaccination can influence vaccination attitudes." This aligns with meta-analyses suggesting that the default nudge is reported to have the most substantial influence on altering people’s behavior [18, 19]. Our study implies that the opt-out approach for COVID-19 vaccination might be effective across different cultural contexts and age groups, suggesting its potential applicability in countries like Pakistan, where vaccine hesitancy is prevalent [12, 20], or in other nudge-cautious countries like Hungary [11]. Thus, in considering its practical application, it is essential to contemplate the inhibiting factors identified.

Cancellation rates were unavailable except for City B, so comparisons between groups could not be conducted. However, the cancellation rate in the opt-out group for COVID-19 was not as high as that for the influenza vaccine of reported 52% in opt-out settings, indicating potentially positive data for implementing opt-out strategies [21]. In opt-in systems, the burden of operation is concentrated on vaccination personnel due to simultaneous handling of appointment receptions, progress management, and cancellation management. In contrast, opt-out systems allow securing vaccination dates and locations in advance, enabling focused management of progress and cancellations thereafter. The insights gained from the cancellation rates in the opt-out group in this study suggest the feasibility of creating waiting lists for cancellations.

In addition, City B did not receive any complaints about implementing an opt-out system, suggesting that even among the Japanese, who are generally cautious about changes in default settings, the opt-out system was accepted without significant dissatisfaction. Opt-out approaches appeal to addressing behavioral inertia, especially when individuals face psychological barriers such as limited capacity to evaluate and compare choice options [19]. In the opt-out group of this study, it is inferred that the discomfort was reduced as individuals were relieved from the process of enrollment. On the other hand, in opt-in systems, there have been instances, such as servers crashing due to a surge in COVID-19 vaccine applications, leading to resident dissatisfaction, suggesting that the opt-in group might have experienced more complaints [22].

Furthermore, the opt-out group experienced a more limited decrease in vaccination rates from the first to the second dose compared to the opt-in group. Systematic reviews indicate that nudging may be effective in producing immediate behavioral changes; however, there is little evidence that nudging interventions result in lasting behavioral changes [23]. Nevertheless, in the context of COVID-19 vaccination, it appears that the influence may last for at least two doses. This can be explained by the fact that opt-out strategies powerfully address behavioral inertia [19]. Conversely, in the opt-in group, dissatisfaction experienced during the application for the first dose might have contributed to a lower uptake of the second dose.

Consequently, we can infer that the potential inhibiting factors related to the opt-out approach have been largely cleared. This study not only demonstrated increased vaccination rates but also indicated that the inhibiting factors were minimal, suggesting benefits for both the vaccine recipients and administrators. Although the effectiveness of default nudges in increasing vaccination rates has been recognized internationally, Japan has shown conservative responses in handling the COVID-19 pandemic, with adherence to initial methods, and thus, such approaches were not readily implemented [24]. Based on the findings of this study, it is considered advisable to include the option of implementing an opt-out system in the strategies for future pandemics beyond COVID-19.

This study has several limitations. First, the scale of infection rates varies from city to city; thus, it cannot be ruled out that this may impact vaccination coverage. As the number of outbreaks or infected cases by the city was not available, their use in the analysis was not feasible. Secondly, since this study was conducted in only one region and only one city adopted the opt-out system, there are still points to be clarified before these results can be immediately generalized. Specifically, particular factors within City B (such as the strength of the mayor’s leadership) may have influenced the vaccination rate. However, this study was unable to examine these factors. Further research, including qualitative research, is needed to overcome these limitations. Despite these limitations, this study is a substantive inquiry suggesting that the opt-out approach to COVID-19 vaccination could lead to higher vaccination rates, without being significantly hindered by factors like high cancellation rates and unpopularity.

In conclusion, our study implies that the opt-out approach to COVID-19 vaccination in Japan, characterized by strong vaccine hesitancy and conservative attitudes towards default changes, increased vaccination rates across 20–64 age groups, overcoming potential inhibiting factors such as higher cancellation rates, unpopularity, and limited effectiveness of the intervention. These results underline the potential of behavioral interventions like the opt-out system in enhancing public health responses, especially in the context of potential pandemics. However, given the study’s unique factors within the single city that adopted the opt-out system, further research is warranted to validate these findings and explore their broader applicability.

Supporting information

Acknowledgments

We would like to thank the cities in A prefecture for their cooperation and Rikuta Hamaya, MD, PhD, from the Division of Preventive Medicine, Brigham and Women’s Hospital for providing advice.