The estimated burden of 15 vaccine-preventable diseases from 2008 to 2020 in Japan: A transition by the COVID-19 pandemic

Abstract

Background

Evaluating the national burdens across multiple vaccine-preventable diseases (VPDs) can be informative to identify the areas for improvements in the national immunization program.

Methods

The annual burden of diseases from 2008 to 2020 in Japan were calculated with the incidence- and pathogen-based approach for the 15 VPDs (hepatitis B virus infection, human papillomavirus (HPV), influenza, invasive pneumococcal disease, invasive Haemophilus influenzae type b (Hib) disease, invasive meningococcal disease, Japanese encephalitis, measles, mumps, pertussis, rotavirus, rubella, tetanus, tuberculosis and varicella), using disability-adjusted life year (DALY).

Results

The average annual burden between 2008 and 2020 is the highest in influenza (114,129 DALY/year), followed by HPV infection, hepatitis B virus infection, tuberculosis and mumps (109,782, 69,883, 23,855 and 5693 DALY/year). In the pre-COVID-19 period (2008–2019), the decreasing trend of burden was observed in hepatitis B virus infection, invasive pneumococcal disease, invasive Hib disease, tuberculosis and varicella. HPV infection is the only VPD which had more than 100,000 DALY/year for all years during the study period. In 2020, the estimated annual burdens are decreased in influenza (71%), invasive pneumococcal disease (51%), invasive Hib diseases (54%), invasive meningococcal disease (64%), measles (98%), mumps (47%) pertussis (83%), rotavirus infection (95%), rubella (94%) and varicella (35%) compared with those in 2019.

Conclusions

The study demonstrated decreasing trends of burdens for some VPDs, while a persistently high burden has been observed for other VPDs, including HPV infection. The COVID-19 pandemic has caused dramatic reductions in the burdens of many VPDs in 2020.

1. Introduction

Although routine vaccination programs have expanded worldwide over the past few decades, there have been outbreaks of vaccine-preventable diseases (VPDs) [1,2]. Japan has introduced the routine vaccination program for diphtheria, hepatitis B virus, human papillomavirus (HPV), influenza (for elderly), Streptococcus pneumoniae, Heamophilus influenzae type b (Hib), Japanese encephalitis, measles, pertussis, polio, rotavirus, rubella, tetanus, tuberculosis and varicella. While the routine vaccination program has significantly contributed to the control of VPDs, the national epidemiological surveillance for infectious disease (NESID) system has reported recent outbreaks for some VPDs, including rubella in 2013 and 2018/2019 and mumps in 2016/2017, partly because of suboptimal vaccine coverage rates [[3], [4], [5]].

In addition, the ongoing pandemic of coronavirus disease 2019 (COVID-19) by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has dramatically changed the epidemiology of infectious diseases, including VPDs. The mitigation measures to prevent the transmission of SARS-CoV-2 has reduced the transmission rate of other infectious diseases [[6], [7], [8]]. While recent studies evaluated the impact of COVID-19 on burdens of some specific VPDs [[9], [10], [11]], the data regarding an evaluation or a comparison of disease burdens across multiple VPDs before and during the COVID-19 outbreak have not been reported.

The Global Burden of Diseases, Injuries, and Risk Factors (GBD) Study has estimated the burdens of a variety of diseases from a global perspective using disability-adjusted life years (DALYs), which allows a comparison of disease burdens across different diseases [12]. However, this study did not include some important VPDs such as rubella and mumps, and their estimates were not specific for organisms or subtypes of organisms covered by vaccines, including HPV, S. pneumoniae, and Hib. Therefore, evaluating and comparing the national burdens of each VPD by year, including 2020, can be informative for policy makers to identify the areas of improvements in the national immunization program during the COVID-19 pandemic. The primary objective of the study was to evaluate the changes in the burdens of 15 VPDs in Japan from 2008 to 2020, a period when some vaccines were introduced into the national immunization program. The secondary objective was to evaluate the impact of COVID-19 on the burdens of each VPD by comparing the burden in 2020 with those in previous years.

2. Methods

2.1. Study design and settings

In this study, the annual burdens from 2008 to 2020 in Japan were calculated with the incidence- and pathogen-based approach for the 15 VPDs, including hepatitis B virus infection, HPV, influenza, invasive pneumococcal disease, invasive Hib disease, invasive meningococcal disease, Japanese encephalitis, measles, mumps, pertussis, rotavirus, rubella, tetanus, tuberculosis and varicella, according to the national vaccination documents and the recommended vaccination schedule from a Japanese organization [[13], [14], [15]]. Since no diphtheria or wild-type polio case has been reported during the study period, the burdens of these two VPDs were not estimated in this study. For the following VPDs, disease burdens caused by serotypes or genotypes covered by the vaccines were estimated; genotypes covered by the 9-valent HPV vaccine (6, 11, 16, 18, 31, 33, 45, 52, and 58) for HPV-related diseases, serotypes covered by the 13-valent pneumococcal conjugate vaccine (1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19 A, 19F, and 23F) or 23-valent pneumococcal polysaccharide vaccine (1, 2, 3, 4, 5, 6B, 7F, 8, 9 N, 9V, 10 A, 11 A, 12F, 14, 15B, 17F, 18C, 19 A, 19F, 20, 22F, 23F, and 33F) for invasive pneumococcal disease, serotype B for invasive Haemophilus influenzae disease, and serotypes A, C, W, and Y for invasive meningococcal disease. The data regarding the annual number of new cases were obtained from the NESID system unless specified otherwise [3]. Given the lack of data regarding the disability weight for COVID-19 related conditions, the burden of COVID-19 was not estimated in this study. The year of routinization and the national coverage rate for each vaccine preventable disease are presented in Table 1 [[12], [13], [14], [15], [16], [17], [18], [19]]. The burden of herpes zoster was not calculated because there is no nationwide surveillance to timely monitor its incidence at the time the study was conducted.

Table 1.

Vaccine preventable diseases included in this study.

	Vaccine mainly used in Japan [[13], [14], [15], [16], [17]]	Year of routinization [[16], [17], [18]]	Coverage rage in 2019a [19]
Hepatitis B virus infection	Monovalent vaccine	2016	94.4%
Human papillomavirus infection	4-valent vaccine (9-valent vaccine was approved in 2020)	2013b	1.3%
Influenza	4-valent inactivated vaccine	2001 (for elderly)	47.9%
Invasive pneumococcal disease	13-valent conjugate vaccinec	2013	95.2%
Invasive Haemophilus influenzae type b disease	Monovalent vaccine	2013	95.0%
Invasive meningococcal disease	4-valent conjugate vaccine for serogroups A, C, W and Y	Not routine
Japanese encephalitis	Monovalent vaccine	1995d	125.7%
Measles	MR vaccine	2006 for two doses	98.5%
Mumps	Monovalent vaccine	Not routine
Pertussis	DTaP-IPV	1950	95.0%
Rotavirus	5-valent or monovalent	2020
Rubella	MR vaccine	2006 for two doses	98.5%
Tetanus	DTaP-IPV	1968	95.0%
Tuberculosis	BCG vaccine	2005e	95.4%
Varicella	Monovalent vaccine	2014	96.0%

Abbreviations: BCG, Bacille Calmette-Guérin; DPaT-IPV, diphtheria, acellular pertussis, tetanus and inactivated polio vaccine; MR measles rubella.

^aFor the first dose. Because the coverage rate is calculated by the number of total people at any age who received a vaccine dose divided by the population in a recommended age for the vaccine, the coverage rage may be more than 100% [19].

^bWithdrawal of governmental recommendation since 2013.

^c23-valent polysaccharide vaccine mainly for adult.

^dWithdrawal of governmental recommendation between 2005 and 2009.

^eTransition from vaccination for children with negative tuberculin skin test to routine vaccination.

For all indicators, data using Japanese cohorts were selected wherever possible. In case of the lack of Japanese data, data from outside of Japan were used. The outcome tree models and the details of indicators selected in each VPD are presented in the supplemental file 1 and supplemental file 2. All analyses were conducted using Microsoft Excel 2016 (Redmond, WA, USA). Because this study only used publicly available and non-identifiable data, the ethical review was not required.

2.2. Disability-adjusted life years calculation

The annual DALYs were calculated to evaluate the burden of each VPD and its related outcomes. For the type-specific vaccines, including Hib, 9-valent HPV, meningococcal and pneumococcal vaccines, DALYs caused by the vaccine-covered types were calculated. The DALYs are the sum of burdens of morbidity and mortality. The morbidity and mortality were expressed as the years of life lived with disability (YLDs) and the years of life lost (YLLs).

YLDs are calculated as the annual number of new cases (n) with each disease or condition stratified by age and sex, multiplied by the disability weight (w) and the duration of the condition (d). The disability weight is presented as a value between 0 and 1 depending on the severity of each health outcome (0 is perfect health and 1 is death).

YLD for each health outcome = n × w × d.

The disability weight data were obtained from the recent GBD study unless specified otherwise (Supplemental file 2) [2].

YLLs were calculated as the number of deaths (de) due to each VPD at each age and sex grouping, multiplied by the remaining life expectancy (RLE) obtained from the Japanese life table in 2019 [20].

YLL for each health outcome = de × RLE.

The mortality data were obtained from the Japanese vital statistics or other national reports whichever more comprehensive in each VPD [21]. While the disease burden of asymptomatic infection was assumed to be zero, the burdens of subsequent complications from asymptomatic infections were estimated [12,22]. The burdens of complications which present at a later stage (i.e. cancer and liver cirrhosis) were counted at the year of the reported diagnoses or deaths for each complication. The burdens of acute complications secondary to each VPD (i.e. meningitis due to pneumococcal infection) were counted at the same year when each VPD case was reported unless specified otherwise (Supplemental file). Because recent studies about DALY estimations have not included age weighting and discounting according to the guidelines [12,[22], [23], [24], [25]], our study also did not consider them.

2.3. The number of new cases and mortality

The reported annual number of cases stratified by age and sex for each VPD were obtained from the NESID system [3].

The NESID system has the notifiable surveillance (a passive surveillance in which all confirmed cases should be reported) and the sentinel surveillance (only designated health care facilities report the confirmed cases). The notifiable surveillance includes diphtheria, acute hepatitis B infection, invasive Haemophilus diseases (since 2013), invasive meningococcal diseases (since 2013), invasive pneumococcal diseases (since 2013), Japanese encephalitis, measles, polio, rubella, tetanus and tuberculosis. The sentinel surveillance includes genital warts, influenza, mumps, pertussis (since 2018), rotavirus infection (since 2013) and varicella (hospitalized varicella is in the notifiable surveillance since 2013).

The underestimation was also considered for the reported incidence by the NESID system [22,26]. For notifiable diseases, the underestimation ratio (a multiplication factor to estimate the total number of cases from the reported number of cases) for each VPD was calculated. For sentinel diseases, the data regarding the estimated total number of cases were obtained [[27], [28], [29]]. The details about evaluating underestimation in each VPD are described in supplemental file.

For the death data obtained from Japanese vital statistics, because of the vital statistics data were only available until Oct 2020 at the time the study was conducted, the annual number of deaths in 2020 were estimated as.

The estimated number in 2020 = The annual number in 2019 × The number of deaths between January 2020 and October 2020/The number of deaths between January 2019 and October 2019.

Because the age of cases or deaths in 2020 has not been reported in the NESID system or in the Japanese vital statistics when the study was conducted, the age distribution of cases in 2020 was assumed to be the same to that of 2019, except for invasive meningococcal disease, Japanese encephalitis and tetanus for which the average age distribution from 2008 to 2019 was used to assume the age distribution in 2020 given the rarity of these diseases.

A proportion of cases with hepatocellular carcinoma are related with hepatitis B virus infection. HPV related cancers include cervical, head and neck, and anogenital cancers. For these cancers, the annual numbers of newly diagnosed cases and deaths stratified by age and sex were obtained from the national cancer registry, which were multiplied by the fraction attributable to the type(s) of infection covered by the vaccine (hepatitis B virus or 9 genotypes of HPV covered by the vaccine) to calculate YLDs and YLLs, respectively [30].

3. Results

The estimated annual DALYs of the 15 VPDs between 2008 and 2020 are presented in Supplemental file 3, and those of four key VPDs (invasive pneumococcal disease by the 13 valent pneumococcal conjugate vaccine, invasive Hib disease, influenza and HPV infection) are presented in Fig. 1 . The average annual burden between 2008 and 2020 is the highest in influenza (114,129 DALY/year), followed by HPV infection, hepatitis B virus infection, tuberculosis and mumps (109,782, 69,883, 23,855 and 5693 DALY/year, respectively). The average annual burden is the lowest in Japanese encephalitis (73 DALY/year). While the estimated annual burdens of hepatitis B virus infection, HPV infection, invasive pneumococcal disease, invasive Hib disease, invasive meningococcal disease, Japanese encephalitis, measles, and tuberculosis were mainly driven by YLL, those of influenza, mumps, pertussis, rotavirus infection, tetanus and varicella were primarily derived from YLD.

Fig. 1.

The estimated disability-adjusted life years by year for four vaccine-preventable diseases in Japan.

Abbreviations: DALY, disability-adjusted life year; Hib, Haemophilus influenzae type b; PCV, pneumococcal conjugate vaccine; PPSV, pneumococcal polysaccharide vaccine; YLD, years of life lived with disability; YLL years of life lost.

Regarding the trend of disease burdens in the pre-COVID-19 period (2008–2019), the decreasing trend of the annual burden was observed in hepatitis B virus infection, invasive pneumococcal disease, invasive Hib disease, tuberculosis and varicella. For example, the annual burdens of invasive pneumococcal disease by serotypes covered by the 13-valent pneumococcal conjugate vaccine and invasive Hib disease were 8319 and 1241 DALY/year in 2008 and 3053 and 12 DALY/year in 2019, respectively (Fig. 1A and B). However, the reduction in the burden of invasive pneumococcal disease by serotypes covered by the 13-valent pneumococcal conjugate vaccine or 23-valent pneumococcal polysaccharide vaccine was more stagnant (9540 DALY/year in 2008 and 6974 DALY/year in 2019) compared with that by the 13-valent pneumococcal conjugate vaccine alone. Influenza had a wide variation in the annual burden. The annual burden of influenza surged in 2009 when there was a pandemic of influenza A (H1N1) pdm09, followed by the drop of the burden in 2010 (Fig. 1C). Measles had a peak of annual burden in 2008 (4247 DALY/year). The surges of burdens in rubella were observed in 2013 (2461 DALY/year) and 2019 (530 DALY/year). Mumps showed variable annual burdens in each year with epidemics in 2010 (12,929 DALY/year) and 2016 (11,541 DALY/year). The annual burden of rotavirus infection also had a variability with the highest burden in 2011 (5476 DALY/year) and the lowest burden in 2014 (1791 DALY/year) during the pre-COVID-19 period. On the other hand, some VPDs, including HPV infection and tetanus, showed consistent annual burdens. The estimated annual burdens of HPV ranged from 104,795 in 2009 to 115,677 DALY/year in 2014 (Fig. 1D), and the estimated annual burdens of tetanus ranged from 529 DALY/year in 2015 to 823 DALY/year in 2010. HPV infection is the only VPD which had more than 100,000 DALY/year for all years during the study period.

In 2020, the estimated annual burdens are decreased in influenza (71%), invasive pneumococcal disease (51%), invasive Hib diseases (54%), invasive meningococcal disease (64%), measles (98%), mumps (47%) pertussis (83%), rotavirus infection (95%), rubella (94%) and varicella (35%) compared with those in 2019. However, the estimated reduction in annual burden in 2020 (vs 2019) is none or relatively small in hepatitis B virus infection (3%), HPV infection (0%), Japanese encephalitis (7%), tetanus (11%), and tuberculosis (10%).

4. Discussion

The study highlights the burdens of 15 VPDs for the recent 13 years in Japan. Between 2008 and 2019, the reducing trends of burdens were demonstrated in some VPDs, including hepatitis B virus infection, invasive pneumococcal disease by the 13-valent pneumococcal conjugate vaccine, invasive Hib disease, and varicella. While the majority of VPDs investigated in this study showed the dramatic reductions of burdens in 2020 compared with those in the previous years, the degrees of reductions were highly variable in each VPD.

During the pre-COVID-19 study period (2008–2019), Japan newly introduced routine vaccinations for hepatitis B virus, HPV, pneumococcal infection, Hib infection, and varicella. Although the decreased trend in the burden of hepatitis B infection before the introduction of routine vaccination is likely because of other measures introduced well before the initiation of the routine vaccination [[31], [32], [33]], the reductions in the burdens of invasive pneumococcal disease by the 13-valent pneumococcal conjugate vaccine, invasive Hib disease and varicella support the effectiveness of these routine vaccinations with achieving high national coverage rates for these vaccines [19]. On the other hand, the reduction in the burdens of invasive pneumococcal disease by the 13-valent pneumococcal conjugate vaccine or the 23-valent pneumococcal polysaccharide vaccine was less remarkable compared with that by the 13-valent pneumococcal conjugate vaccine. Since the routine 13-valent pneumococcal conjugate vaccination was introduced for children, the serotype replacement has been reported in both children and adult [34,35]. Although the majority of invasive pneumococcal diseases have recently been caused by serotypes not covered by the 13-valent pneumococcal conjugate vaccine, maintaining the high coverage rate of the 13-valent pneumococcal conjugate vaccine for children is still important because a reduction of vaccine coverage may lead to increased burden of diseases by the vaccine-covered serotypes [36].

The burden of tuberculosis also showed a consistent reduction. Of note, the number of reported cases with active tuberculosis and the number of deaths due to tuberculosis have been relatively stable during the study period [3]. The reason of reduction in the burden may be because of the shift of the age distribution. For example, while the elderly population aged 85 years or older accounted for 38% of deaths due to tuberculosis in 2008, this proportion increased to 64% in 2019 [3].

The annual burden of influenza may look increased in recent years, although this may be within a range of annual variations. Similar trends have been reported from other countries [37,38]. Vaccine coverage rates have not had a reducing trend in recent years [39]. Of note, a new method to estimate the number of cases with influenza has been applied since 2018 [40]. While this study followed the suggestion by Ministry of Health, Labour and Welfare to adjust the annual number in 2017 or before so the number is estimated with the same method throughout the study period [40], a continuous monitoring the reported number and the estimated number is crucial to find the best way to estimated the number of cases with influenza.

While the reduction in the disease burden in 2020 was observed for the majority of VPDs investigated in this study compared with those in previous years, the magnitude of the reduction was more prominent in VPDs with droplet and contact transmissions. Among respiratory infections, tuberculosis showed a smaller reduction of the burden compared to other respiratory infections. This may be explained by the fact that the majority of active tuberculosis cases are from the reactivations of latent infections, instead of primary infections [41]. On the other hand, VPDs transmitted by other routes (i.e. soil, mosquitos, and sexual intercourses) showed relatively stable trends of burdens in 2020 compared with those in the previous years. The shift in the burdens of VPDs in 2020 would reflect the change of people's behaviors due to the COVID-19 mitigation measures. Of note, Japan experienced a first wave of COVID-19 since March 2020 with their first declaration of state of emergency in April 2020 [42]. Therefore, the reduction of the number of cases with some VPDs have been more prominent after March or April in 2020, compared with January and February in 2020 [3]. For example, the annual number of influenza cases with medical visits in a calendar year of 2020 was estimated as more than 4 million, while that of 2020/2021 season was estimated as about 14,000 [27]. This means that the relative reduction in the annual burden of some VPDs in 2020/2021 season (vs pre-COVID-19 seasons) would be much more significant than that in the calendar year of 2020.

Although the vaccine coverage rate data during the COVID-19 period in Japan are limited, a report suggested the reduction of childhood vaccination rates in early 2020 [43]. Although the negative impact of the reduced vaccination rates may be masked when the transmissions of infectious diseases are reduced due to the mitigation measures related with the COVID-19 pandemic (i.e. physical distance), the continuous reduction of vaccination rates may cause the increased burdens of VPDs once the mitigation measures are released [10,11].

Some articles discussed the potential under-detection of other infectious diseases due to fewer medical visits, less opportunities for testing or misdiagnosis during the COVID-19 pandemic [44,45]. Although this can be potentially applied to Japan, the impact may not be large given that the reduction in the disease burdens has been specific for infectious diseases transmitted through droplet or contact. In addition, the relatively stable trends in the numbers of some other infectious diseases, including zoonosis, arthropod-borne diseases and sexually transmitted diseases, have been reported in 2020 compared with those in the previous years [3].

The dramatic shift of VPD burdens in 2020 has some implications for the development of the vaccine program during the COVID-19 pandemic. First, the relative importance of VPDs with consistently high burdens even during the COVID-19 period should be more emphasized. For example, because the coverage rate of HPV vaccine has been very low in Japan, the Japanese vaccine program would need to focus on the improvement of HPV vaccination as one of top priorities irrespective of the duration or the magnitude of the impact of COVID-19. In addition, even if the burdens of some VPDs have reduced during the COVID-19 pandemic, it would be important to maintain the high vaccination coverages for these VPDs because some studies have shown that the incidence of these VPDs would significantly increase without optimal vaccine coverages once the COVID-19 mitigation measures are released [10,11]. It should be noted that the study does not negate the importance of maintaining vaccination programs for diseases with minimal burdens.

There are several limitations of this study. First, there are some uncertainties about some indicators, including the underestimation ratios. The sensitivity analysis, such as Markov chain Monte Carlo methods, was not conducted given the lack of the data regarding the probability distribution of these indicators. In addition, some indicators, including disability weights, were the data reported from outside of Japan given the lack of Japanese data. Despite these limitations, the study result is still helpful to evaluate which vaccine program needs to be prioritized to decrease the overall disease burden of VPDs in the future. Although the study is specifically for Japanese health care system, the methodology presented in this study can be informative for international vaccine policy makers to evaluate and prioritize under-controlled VPDs in each country during the COVID-19 outbreak. Finally, the data regarding the magnitude of potential under-detection of VPDs specifically during the COVID-19 period are limited, although the possibility of underreporting and the impacts of changes of the reporting criteria during the study period were adjusted in the study.

In conclusion, the study showed decreasing trends of burdens for some VPDs over the past decade, while a persistently high burden has been observed for other VPDs, including HPV infection. The COVID-19 pandemic has caused a dramatic shift in the burdens for many VPDs in 2020. The vaccination program for VPDs with high burdens even during the COVID-19 pandemic should be more promoted. Maintaining the optimal vaccine coverage is still important for VPDs with declined burdens during the COVID-19 period.

4.1. ICMJE statement

All authors meet the ICMJE authorship criteria.TK designed the study, collected and analyzed data, drafted the initial manuscript, reviewed the manuscript and approved to submit the final version.

Funding sources

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interest

No potential conflict of interest relevant to this article was reported.

Appendix A. Supplementary data

The following are the Supplementary data to this article: