Long-term trends of pediatric type 1 diabetes incidence in Japan: before and during the COVID-19 pandemic years (1999–2021)

Background: The incidence of type 1 diabetes has increased worldwide whereas a long-term trend for pediatric type 1 diabetes has remain elusive in Japan. Objective: This study aimed to investigate the incidence and secular trend of pediatric type 1 diabetes before and during the COVID-19 pandemic from 1999 to 2021 in Oita Prefecture, Japan. Subjects: We investigated the increase in numbers of newly diagnosed patients with type 1 diabetes aged <15 years from1999 to 2021. Method: We surveyed the medical information from hospitals in Oita and Oita Prefecture database in Japan. Result: The incidence of type 1 diabetes in those aged <15 years increased annually by 5.3% in all children, especially by 8.1% in boys aged 10–14 years during the past 23 years. The average incidence rate of 3.9/100,000/year was almost consistent with the previous report from Asian countries. However, no signi�cant change was found in the increasing incidence trend of type 1 diabetes before and during the COVID-19 pandemic. Conclusion: The incidence of pediatric type 1 diabetes in Japan has signi�cantly increased over the past 23 years, which is consistent with a worldwide trend. No signi�cant increasing trend was observed during the past 2 years during the COVID-19 outbreak.


Introduction
Type 1 diabetes (T1D) is recognized as an autoimmune disease characterized by insulin de ciency and resultant hyperglycemia 1 .Intensive research for T1D has rapidly increased over the past several decades, resulting in a broad understanding of many features of the disease, including its genetics, epidemiology, autoimmunity, β-cell phenotypes, and the burden of disease.The cause of T1D is known to involve both environmental and genetic factors 2 .In addition, many reports of local or regional epidemics of T1D occurrence are best explained by environmental exposure 3 .One of the most common triggering factors includes enteroviral infection, such as Coxsackievirus B 4,5 .
T1D is classi ed as a polygenic disease with identical twin concordance of 30-70% 6 , sibling risk of 6-7%, and a risk of 1-9% for children who have a parent with T1D 7 .Ethnicity deeply contributes to the incidence of T1D; for example, Caucasians are more susceptible to T1D than African-Americans and Hispanic-Americans, whereas Asians, including Chinese, Korean or Japanese, and the people in South America were the lowest in the United States (US) 8 .
The gradual increase of incidence T1D in children <15 years has been reported worldwide.In the US, children and adolescents aged <20 years indicated a 1.8% annual increase during 2002-2012 9 , and a similar increase rate of 1.3% has been reported for the Canadian province of British Columbia in the period 2002-2013 10 .In Australia, a signi cant increase of 1.2% (95% CI 0.4%, 2.1% was observed in the 10-to 14-year-old age group during the period 2000-2011 11 .Within Europe, no increase was found in Sweden during the period 2005-2007 12 , and similar attening incidence rates were subsequently reported in two other high-incidence Scandinavian countries (Finland 13 and Norway 14 ).Asian populations with a very low incidence of T1D (0.4-1.1 cases/year/100,000 individuals), such as Uzbekistan, Korea or China, also have increasing trends (3-12% increase) [15][16][17] .However, no remarkable increase has been reported in Japan; for example, Kawasaki et al. reported T1D incidence in patients aged 0-14 years averaging 2.37 cases/100,000/year from 1993-2001 18 .Onda et al. also reported that T1D incidence in Japan did not increase from 2. 24 (1998-2001) to 2.27 (2005-2010)   cases per 100,000 persons/year 19 .However, the results seemed elusive, partly because numbers of newly diagnosed patients might have been omitted in the registration system of "Speci c Chronic Pediatric Diseases," also called the Medical Aid Program for Chronic Pediatric Diseases of Speci ed Categories (MAPChD) in Japan 20 .Since the nationwide whole-covering registration system for children with T1D has not been established in Japan, it would be di cult to estimate the national epidemiological study simply from the data of MAPChD in Japan.Alternatively, the local registry systems in each prefecture or small-scale adjacent areas would be better for extracting the epidemiological data of T1D.
For instance, a recent study in Yamanashi Prefecture in Japan reported that a modest and non-signi cant increase was observed from 1986 to 2018 with an increasing annual rate of 1.16%; among them, the subpopulation aged 5-9 years had an annual signi cantly increased rate of 5.38% (CI: 2.34-8.35%) 21.
Therefore, this present study aimed to explore the incidence, prevalence, and annual trend of childhoodonset T1D in Oita Prefecture in Japan.The chronological trends in the T1D incidence in children <15 years were carefully examined based on the medical records of three core hospitals in charge of pediatric T1D treatment in Oita Prefecture.The data from MAPChD were carefully applied to con rm the accuracy of the epidemiological data.Based on the annual trend of childhood-onset T1D, we focused on a possible change in the incidence of pediatric T1D during the COVID-19 pandemic in Japan.
This study also determined the viral infection occurrence in Oita Prefecture during the evaluation period.
The surveyance data from Oita Prefecture demonstrated the prevalence of Coxsackievirus (Groups A and B) infection in Oita Prefecture seemed to be roughly synchronized with the T1D occurrence (Figure S2).In addition, we explored whether COVID-19 pandemic expansion alters the occurrence of new-onset T1D.We found that the average incidence rate (2007-2018) was 4.6/100,000/year, whereas during the pandemic years of 2019, 2020, and 2021, annual incidence rates were 6.1, 5.5, and 6.2/100,000/year, respectively, suggesting no signi cant change in the T1D prevalence in Oita Prefecture, Japan, before and during COVID-19 pandemic (Table 3).

Discussion
Our study demonstrated that the incidence of T1D in Oita Prefecture, Japan, has been signi cantly increasing every year by 5.3% during the past 23 years (1999-2021).The increasing trend of T1D incidence in Oita Prefecture was almost consistent with the previous report from Asian countries 16 .The average incidence rate of 3.9/100,000/year in Oita Prefecture was higher than that previous report conducted in Yamanashi Prefecture (2.0/100,000/year; 1988-2016) 21 located in the middle of Honshu Island and surrounded by mountains with a population of about 800,000 residents.The Japanese population is classi ed as a single race country consisted with Yayoi race of > 95% of the population, the same population size of Germany or Finland.However, T1D is recognized as a multifactorial genetic disease and therefore local and speci c environmental factors play important roles in the T1D development.Our data from Oita Prefecture was slightly different from those of Yamanashi, Japan; however, the differences observed remained unknown.The environmental difference might exist between Yamanashi and Oita Prefectures might have caused the T1D development.Therefore, further studies conducted in other prefectures in Japan would be bene cial to verify our result.
In the Asian population, a report from Zhejiang province in the low-incidence region of China described a rapid annual incidence increase rate of 12.0% among those aged < 20 years from 2007 to 2013 17 .The incidence and prevalence of childhood-onset T1D in Korea from a nation-based registry demonstrated an increase of 3-4% every year from 2007 to 2017.In Korea, the overall incidence of T1D in childhood-onset increased from 3.70% in 2008 to 4.77% in 2016, according to the Health Insurance Review and Assessment Service 16 .The causes for this ethnicity-independent T1D increase remains unelucidated.The increase in T1D incidence rates in Korea are considerably high compared to those in other countries.Therefore, changing the populations' genetic pool was unlikely to affect the increasing trend.However, local or regional environmental factors are may have impacted the increasing trend of pediatric T1D.The Oita Prefecture data demonstrating the Coxsackievirus infection prevalence seemed to be synchronized with the T1D occurrence.However, it remained elusive whether epidemic Coxsackievirus infection impacted the annual increase of T1D occurrence in Oita Prefecture.Therefore, a further nationwide study is needed to evaluate the association between the occurrence of T1D and types of viral infection.
We did not nd a signi cant change in the T1D occurrence before and during the COVID-19 pandemic.The T1D incidence in children during the COVID-19 pandemic has been reported mainly in Europe.In Poland, the incidence seemed to have decreased after the COVID-19 pandemic compared to that before the pandemic 22 .Comparatively, there was no signi cant change in Italy and Germany between before and after the COVID-19 pandemic 23,24 .The COVID-19 pandemic drastically changed hygiene practices, such as washing hands, wearing masks, maintaining social distance, and reducing other viral infections, especially for children; therefore, these may have caused the gradual changes in the T1D incidence.Therefore, long-term studies should be conducted to investigate the direct and indirect effects of COVID-19 infections to the development of T1D would be helpful to understand the pathogenesis and make prevention of T1D.
Our study had some limitations.First, the analyzed population size was relatively smaller than those in the previous studies in other countries.Although the numbers of pediatric T1D patients were small, the trends observed in the present study were similar to those recognized in previous studies conducted in Japan and Korea.Second, 2 years of survey during COVID-19 pandemic might be too short to conclude the results.Especially, COVID-19 infection for children has been drastically increasing since the beginning of 2022, therefore further study after 2022 will be essential to clarify the direct impact of COVID-19 infection on the development of T1D in Japanese children.Despite these limitations, we think that the present study might be relatively accurate incidences and prevalence of pediatric T1D in Japan.
In conclusion, the incidence of pediatric T1D in Oita Prefecture in Japan has signi cantly increased over the past 23 years, which is consistent with a worldwide trend.No signi cant increasing trend was observed during the past 2 years during the COVID-19 outbreak.

Methods
Geographic features of Oita Prefecture.Oita Prefecture is on the northeast side of the coastal area of Kyushu Island.Oita Prefecture has a population of 1,124,983 (October 1, 2020) and a geographic area of 6,340 km 2 .Oita faces the seaside and is surrounded by mountains.It consists of 18 municipalities, including Oita City.Oita City, the capital of Oita Prefecture, is located in the east center of Oita Prefecture of the coastal area, with a population of approximately 470,000 in 2020, accounting for 40% of the prefecture's population.The population of Oita Prefecture had been steadily decreasing year by year.Children aged < 15 years and older individuals aged > 65 years account for 12.1% and 33.3% of the population, respectively.The corresponding numbers of the national average in Japan are 11.9% and 28.6%, respectively.Therefore, Oita Prefecture appears a roughly 1/100 scaled-down version of Japanese society.
Study population.The patients newly diagnosed with T1D, who were < 15 years of age and living in Oita Prefecture, were enrolled in this study from January 1999 to December 2021.Those who moved out from Oita Prefecture were excluded from this study.Three strategies of data collection methods were prepared for this study.First, clinical information of newly diagnosed patients with T1D was searched from the medical records in three core hospitals: Oita University Hospital, Oita Prefectural Hospital, and National Hospital Organization Nishi-Beppu National Hospital.Second, the pediatricians in all domestic hospitals were directly asked regarding inpatient facilities for children in Oita Prefecture as follows: National Hospital Organization Nishi-Beppu Hospital, Nakatsu Municipal Hospital, Kunisaki Municipal Hospital, Bungo-Ono Municipal Hospital, Saiseikai Hita Hospital, and Tsurumi Hospital.Third, the patients were con rmed by the T1D registered in the MAPChD database.The MAPChD records in Oita Prefecture are separately stored under the management of two government o ces; the Oita City Government for Oita City citizens and the Oita Prefecture for citizens of all cities and towns, except for Oita City.Since the Oita City data before 1999 and those of Oita Prefecture before 2009 were not stored, we used the Oita City data from 2000 to 2021 and those of Oita Prefecture from 2010 to 2021.The population statistics in Oita Prefecture were referred from the database of vital statistics in Oita Prefecture (https://www.pref.oita.jp/site/toukei/index-cpe.html).
Longitudinal epidemiological data of viral infections in Oita Prefecture.The annual occurrence 27 types of infectious diseases, including rhinovirus, parechovirus, enterovirus, parain uenza virus, mumps virus, adenovirus, and herpes family viruses has been investigated > 20 years in Oita Prefecture.The samples of nasopharyngeal swab uid, feces, cerebrospinal uid, or blood were submitted to the Oita Prefectural Research Center for Sanitation and Environment at approximately 10 x-point medical institutions in Oita Prefecture.The epidemiological data report of viral infections in Oita Prefecture has been annually published by the Oita Prefectural Research Center for Sanitation and Environment and is available on the website (https://www.pref.oita.jp/soshiki/13002/nenpo-list.html).
Statistical analysis.Incidence rates were calculated by dividing the numbers of registered children by annual population estimates in Oita Prefecture, Japan.Similarly, the T1D incidence by municipality was measured by dividing the number of registered children in the municipality.Estimates of rates increase were obtained using a mixed effects Poisson regression model with age and sex as xed effects.Time trends of age-standardized rates and annual percent change and P-values were estimated by Joinpoint analysis (Joinpoint Regression Program, Version 4.9.; Statistical Research and Applications Branch, National Cancer Institute, US).To calculate rates, the denominator values (i.e., number of boys and girls aged < 15 years) were obtained from the Japanese Model Population.Subgroup analyses were performed according to sex and age group (0-4, 5-9, and 10-14 years) in patients with T1D.The 95% CI were also calculated for proportions.All other calculations were performed using the R version 3.5.2(R Foundation for Statistical Computing, Vienna, Austria [https://www.R-project.org/]).Rates are given per 100,000 person/year.The signi cance level was set at 5% for two-sided tests.Tables Table 1-3 is available in the Supplemental Files section.