Introduction

An outbreak of novel coronavirus infection began in Wuhan, China, in December 2019, subsequently spreading globally. Since then, a series of mutant strains have emerged. The Omicron BA.1 strain, which first emerged in South Africa at the end of November 2021, [1], swiftly disseminated worldwide, displacing previous strains such as Delta [1, 2]. To counter the emergence of a succession of mutant strains, countries have been promoting infection control measures, but there has been no convergence toward sharing best practices. However, vaccination stands as one of the most crucial preventive measures.

In Japan, the administration of third vaccine dose initiated in December 2021, coinciding with the replacement of Omicron strain by Delta [3], with priority given to healthcare workers and the elderly people. Subsequently, a fourth dose was began in May 2022 for individuals aged 60 or older, those aged 18 to 60 with underlying medical conditions, and those at a recognized high risk of serious illness upon coronavirus infection [4]. In response to the spread of infection with Omicron strain, the Japanese government expanded vaccine eligibility to healthcare workers and staff in elderly care facilities from July 2022. In Japan, most of vaccine types were administered: Pfizer Biontec BNT162b2 and Moderna mRNA-1273 [5].

Regarding vaccine effectiveness (VE), countries with more advanced vaccination programs than Japan have reported the VE against Omicron strain in preventing infection [6, 7], severity [8–10], symptomatic diseases [9, 11, 12], hospitalization [8, 13] and mortality [10]. Preventing infections, alongside averting severity, hospitalization, and mortality, becomes crucial for alleviating the burden on healthcare facilities and resources, and play crucial role in protecting the health and safety of their communities [14].

At our knowledge, few studies were conducted to study the VE, particularly effectiveness of booster vaccines, against Omicron strains in preventing COVID-19 infections in the general population [15–17]. However, repetitive evaluations of VE offer merits in identifying any gaps or challenges in VE across different populations or specific geographic areas [18]. The study’s context and findings contribute to the global understanding of VE. In addition, such repeated analyses contribute to build public trust in vaccination efforts. This fosters greater vaccine acceptance, bolstering infectious disease control [19]. International health organizations might need to adjust their guidance based on emerging data to ensure that vaccination strategies remain effective across different regions. Therefore, in our study, we aim to elucidate the VE against the Omicron strain. Our analysis relies on data collected at PCR centers in Hiroshima Prefecture, Japan from 1 February and 17 March 2022, a period dominated by the Omicron variant in Hiroshima Prefecture [20].

Materials and methods

Results

Table 1 shows the test positivity rates categorized by vaccination status and further grouped by age, sex and additional factors including occupation, pre-existing disease, symptoms, and exposure to close contacts. During the study period, a total of 117,335 individuals underwent PCR testing, yielding an average test positivity rate of 5.0 percent. The test positivity rate was higher among individuals who were not vaccinated (7.9%), in comparison to those who received two doses (4.5%) and three doses (2.8%). The group of individuals who had received three doses exhibited considerably lower test positivity rates across all age groups, both males and females, as well as within occupation groups—excluding medical professionals. Among medical professionals, those who had received two vaccine doses demonstrated a lower test positivity rate (3.0% with 54 cases) than those who had received three doses (3.3% with 130 cases).

Download:

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

Table 1. Summary of vaccination status and PCR test positivity rate by demographic variables and some potential risk factors.

https://doi.org/10.1371/journal.pgph.0003071.t001

Fig 1 illustrates a 7-day moving average of the test positivity rate among individuals categorized by their vaccination status: non-vaccinated, vaccinated with two doses, and vaccinated with three doses. Throughout the entire study duration, the lowest test positivity rate was consistently observed among individuals who received three doses of the vaccine. In contrast, the non-vaccinated group exhibited a comparatively higher test positivity rate.

Download:

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

Fig 1. PCR test positivity rates by vaccination status.

The x axis represents the study period; y axis represents the test positivity rate. Figure displays PCR positive rates for three distinct groups: non-vaccinated, vaccinated with two doses and vaccinated with third doses (booster).

https://doi.org/10.1371/journal.pgph.0003071.g001

Table 2 depicts the calculated vaccine effectiveness. In adjusted models, VE was 34.7% (95% confidence interval [CI]: 28.4%–40.4%) for individuals who received two doses of vaccine compared to the non-vaccinated group. Similarly, VE was 26.4% (95%CI: 16.4%–35.2%) for individuals who were vaccinated with three doses compared to those who were vaccinated with two doses.

Download:

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

Table 2. mRNA vaccine effectiveness administered in Hiroshima Prefecture, Japan.

https://doi.org/10.1371/journal.pgph.0003071.t002

Discussion

In general, the study finding supports the effectiveness of mRNA vaccines in preventing infections caused by Omicron strains. The observed trends in VE reveal an effectiveness between individuals who received two doses in comparison to those with less than two doses. Similarly, the effectiveness was evident when comparing individuals who have had three doses against those with two doses.

While the study findings demonstrate a higher effectiveness of the third dose of the vaccine compared to the second dose, it’s crucial to interpret these results while taking certain situational factors into account. For example, the effectiveness of the second dose of the vaccine was already considered to have declined during the period of this study. The two-dose immunization coverage reached its peak (about 70%) in Hiroshima in November 2021 [29]; thus at least 4 months passed after the second vaccination. According to a study conducted in the U.S. on the efficacy of infection prevention against the Omicron strain in persons aged 18 years and older, two doses of the Moderna vaccine have been shown to be 44.0% effective in preventing infection after 14–90 days, with decreasing effect over time thereafter [15]. Similarly, another U.S. study that collected information on Pfizer vaccine recipients has reported a significant decline in the effectiveness of the second dose in preventing infection in persons aged 12 years and older, from 88% after one month to 47% after five to six months [30]. Furthermore, our result showed that the test positivity rate of elderly people over 80 years old who may have received the vaccine earlier than other generations, was highest among those who received two doses of vaccine. Thus, it is considered that during the study period, the infection prevention effectiveness of the second vaccination had already weakened and coincided with the time when the highly infectious Omicron strain began to spread. Therefore, it underscores the importance of monitoring the timing and intervals between vaccine doses. This highlights the need for regular assessment and potential adjustments to vaccination schedules.

In addition, the infection-preventive effect of the third dose vaccine was considered to be maintained during the period of this study. Previous study reported that the effectiveness of the third dose vaccination was 71.6% after 14–60 days and 47.4% after 61 days, indicating that the third vaccination temporarily restores the effectiveness of the vaccination in preventing infection [11]. Another nationwide representative study in Spain estimated 51.3% effectiveness of booster vaccines against infection up to 34 days after its administration [31]. UKHSA also estimated 45% (35% to 55%) effectiveness in preventing infection up to three months, but insisted there is currently little evidence, and the findings are not conclusive [11]. In this study, the third vaccination was administered within two months before the analysis, so that the infection-preventive effect was still maintained, potentially contributing to a reduced susceptibility to infection. However, a paradox emerged as healthcare workers who received the third vaccine dose exhibited a higher test-positive rate compared to healthcare workers who received the two doses. This disparity could be attributed to the heightened likelihood of breakthrough infections among medical professionals, owing to their increased exposure to concentrated contacts and patients, in contrast to other occupations. This observation suggest that certain professional groups may require tailored vaccination and preventive strategies. Such considerations could include additional protective measures for individuals with heightened exposure to the virus.

Our study’s notable strength lies in the comprehensive data collection conducted across all PCR centers in collaboration with the Hiroshima prefectural office. This extensive and unique epidemiological survey stands as a significant contribution, with no comparable large-scale study conducted in other prefectures of Japan. Moreover, the inclusion of detailed information encompassing occupation and various risk factors, such as the likelihood of close contact and pre-existing conditions, further enhances the robustness of our analysis. These factors were adjusted within the model to account for potential confounding influences.

There are certain limitations and challenges that warrant considerations. Firstly, our study did not permit the specific estimation of VE for individual vaccine types, booster types (heterologous or homologous), time-sensitive effectiveness (effectiveness at different intervals post-vaccination), or the interval between vaccine doses. This limitation stemmed from the absence of pertinent information in the questionnaire, including details regarding vaccine types and vaccination dates. Secondly, the generalizability of our results may be affected due to the nature of our study population. PCR test recipients who visited a PCR center may have a higher level of health consciousness compared to those who didn’t visit a PCR center. Additionally, it’s worth noting that a considerable proportion of the participants who underwent PCR testing at PCR centers were asymptomatic. This scenario potentially skews the representation of individuals with more pronounced symptoms, who may be more inclined to visit clinics rather than PCR centers. Therefore, the applicability of our findings should be interpreted with caution and consideration of these inherent limitations.

Conclusion

These findings collectively underscore the protective impact of mRNA vaccination against Omicron strain infections and highlight the importance of completing the recommended vaccine dosing regimen. We believe that the repeated VE evaluations in this study will contribute to global understanding of VE and increasing public confidence in vaccination efforts. This would promote vaccine acceptance and strengthen new infectious disease control efforts in the future.

Acknowledgments

The authors deeply appreciate Hideki Nishikawa, Jun Hiranaka, Shigeru Watabe, Kenta Ueda, Toshimasa Asahara, Yasuhiro Ishimura for their operational contribution to set up novel data collection system in Hiroshima prefecture, Japan.