Abstract
Background:
Treatment outcomes of children with cancer have improved significantly in recent decades. Despite this, cancer remains the leading cause of mortality among children aged 5–9 years in Japan. The precise trends over time and the changing locations of deaths for pediatric cancer patients in Japan are not clearly elucidated.
Objectives:
This study aimed to validate the annual change in the number and location of deaths in children with cancer in Japan.
Methods:
A retrospective study was conducted using vital statistics (2005, 2010, 2015, 2020, and 2021) and detailed Japanese death records (2018). We sampled patients aged under 15 years who died of cancer, analyzing trends based on demographic characteristics, cause of death, location of death, and place of residence.
Results:
Over the 15-year period (2005–2020), cancer-related deaths declined, with hematological malignancies falling by approximately one-third. Concurrently, the proportion of home deaths increased significantly, rising by 30 percentage points for cancer-related deaths. This was highest for brain tumors, where more than half of the associated deaths occurred at home. Regional disparity was evident; the proportion of home deaths was higher in densely populated areas but lower in sparsely populated regions. Furthermore, only 20% of children died at designated core hospitals, while 50% died at collaborating hospitals, mostly near their residence.
Conclusion:
Japan has seen a decrease in pediatric cancer deaths alongside a growing trend toward home deaths. This shift indicates increased support for patient- and family-centered end-of-life care but highlights the need to strengthen collaboration in regions with fewer medical resources.
Key Messages
This study first investigated the trends in the place of death among children with cancer in Japan. The number of deaths among children with cancer in Japan has been declining over time, and the proportion of home deaths has been increasing.
Introduction
The treatment outcomes for children with cancer have significantly improved in recent decades, and the 5-year survival rate of patients in high-income countries reaches approximately 80–90%.1,2 However, according to the 2021 vital statistics in Japan, cancer remains the leading cause of death among children aged 5–9 years and the second leading cause of death among those aged 10–14 years. In 2012, the second term Basic Plan to Promote Cancer Control Programs based on the Cancer Control Act was enforced in Japan. This act stated that pediatric cancer is one of the priority items and that the improvement of pediatric palliative care and psychosocial support is essential throughout the nation. 3 In 2013, the government assigned 15 designated core hospitals for pediatric cancer, and the consolidation of pediatric cancer care has progressed, with 41% of all pediatric patients with cancer admitted in 2019 being treated in designated core hospitals. These movements may have led to a change in the number and location of deaths among children with cancer. In particular, the treatment outcomes of patients improved, and patients have been moving away from their local residence and transferring closer to the treating hospital area. However, the actual status is still unknown. The current study aimed to validate the annual change in the number and place of death in children with cancer in Japan.
Methods
A retrospective study was performed using vital statistics and data from the Japanese death records. Vital statistics is a national survey conducted from January 1 to December 31 of the survey year. This survey aims to identify important events in Japan and to obtain basic data sources for population and policymaking on health, labor, and welfare. The Japanese death record is a death statistic based on data from death certificates. The subjects of the survey included the total number of live births, deaths, marriages, divorces, and fetal deaths notified based on the Family Registration Law and Provisions Regarding Notification of Stillbirths. The subjects of the current survey were events that occurred in Japan within the survey year that concern persons with Japanese nationality. In this study, patients aged <15 years who died of cancer were sampled according to demographic characteristics such as age, sex, cause of death, location of death, and place of residence. This study was divided into 10 regions based on place of residence: Hokkaido (1), Tohoku (1), Kanto (4), Koshinetsu (0), Hokuriku (0), Tokai (3), Kinki (4), Chugoku (1), Shikoku (0), and Kyushu (1). The information in the parentheses indicates the number of base hospitals in each region. The cause of death was examined by classifying the name of the disease as listed on the death certificate using the International Classification of Diseases, Tenth Revision. Data from 2005, 2010, 2015, 2020, and 2021 were examined to determine the annual trends. To compare different types of cancer, data on all causes of death were also examined. The Japanese death records, which are a separate statistical record requiring a different procedure to obtain than vital statistics but provide more detailed data, were reviewed using data from 2018, and the data on the location of death based on hospital function and address of the place of death were collected. This specific year was chosen because these data were readily available, having been previously obtained under an approved application for a separate research project. These data were summarized descriptively. Ethical approval was not required for this study as it utilized only publicly accessible data.
Results
Trends in the number of deaths
Table 1 shows the trends in all-cause and cancer-related deaths. The annual number of all-cause deaths decreased by 50% over the 15-year period from 2005 to 2020. The population aged <15 years also declined over this period. However, the mortality rates of this age group had decreased by nearly 50% over the last 15 years.
Number of Deaths in Patients Under 15 Years of Age in Japan (All Causes, Cancer)
Mortality rate was calculated based on the population of children aged 15 and under. The data were calculated using death certificate data from vital statistics.
The number of deaths among children with cancer had a similar downward trend, decreasing by >30% over the last 15 years. According to the disease subtypes, the number of hematological malignancy-related deaths had decreased by approximately one-third over the last 15 years. Meanwhile, the number of brain tumor-related deaths had been increasing.
The Kanto region, which has the largest population, had the highest number of deaths. However, considering cancer mortality per 1000 population, the regional differences were not significant.
Trends in the location of death
Figure 1 shows the trends in the place of all-cause and cancer-related deaths. Over a 15-year period (2005–2020), there was a clear upward trend in home deaths for all causes, with the proportion increasing from 8.9% to 12.8%—a rise of 3.9 percentage points. In particular, there was an increase of 1.9 percentage points in 2020–2021.
Place of deaths under 15 years of age. The data were calculated using death certificate data from vital statistics. “Others” include facilities other than hospitals (e.g., clinics with beds) and details unknown.
As with all-cause deaths, the location of death in patients with cancer was also increasing. However, the trend was more evident. In particular, the proportion of home deaths increased by 30 percentage points in the 15-year period from 2005 to 2020 and by 7.9 percentage points in the 1-year period from 2020 to 2021.
Figure 2 shows the location of death according to disease. The proportion of brain tumor-related deaths occurring at home was high, with more than half of children dying at home by 2021. By contrast, there were a few numbers of hematological malignancy-related deaths occurring at home. However, in the last few years, nearly 20% of children had died at home.
Place of deaths by disease among cancer patients under 15 years of age. The data were calculated using death certificate data from vital statistics. “Others” include facilities other than hospitals (e.g., clinics with beds) and details unknown.
Percentage of home deaths according to region
Figure 3 shows the percentage of cancer-related deaths occurring at home based on region. The proportion of home deaths had increased over time in the Kanto, Tokai, and Kinki regions. In contrast, the proportion of home deaths was low in the Hokuriku, Chugoku, Shikoku, and Tohoku regions. This finding indicated that most deaths occurred in hospitals.
Percentage of home deaths by region. The data were calculated using death certificate data from vital statistics.
Location of death according to hospital function
Figure 4 shows the percentage of deaths based on hospital function by disease from the 2018 Japanese death records. Approximately 20% of all cancer-related deaths occurred at the designated core hospital, and approximately 50% occurred at the collaborating hospitals. In particular, children with hematological malignancies were more likely to die at the collaborating hospitals. According to the data on the location of death, which only included in-hospital deaths, 88% of children died in hospitals in the same prefecture as their place of residence.
Place of deaths by hospital function in 2018 from certificate data.
Discussion
This study is the first to investigate the trends in the place of death among children with cancer in Japan.
One of the important findings of this study was that the number of pediatric cancer deaths had decreased over the last 15 years. The population aged less than 15 years has also declined during this period. However, the mortality rates had decreased, suggesting that the decline in the number of deaths is due to improvements in the quality of medical care. Although the number of deaths among children is also declining, the reduction in cancer mortality rates is particularly significant. This included a notable decrease in the number of deaths related to hematological malignancies. Similarly, data from other countries show a greater reduction in mortality rates for hematological tumors compared with brain tumors. 4 This is likely associated with advancements in novel immunotherapy and molecular-targeted drugs, which have increased treatment options and prolonged survival. 5
The second and most important finding of this study was that the proportion of home deaths for children with cancer is increasing. In the past, pediatric cancer treatment was often idealized as a “good fight,” with a tendency to persevere even in overwhelmingly unfavorable circumstances. 6 However, in recent years, it has become common practice to share information with parents and children and respect their wishes during the end-of-life period. 7 For adult cancer patients in Japan, data show that 11.8% die at home, and 16.5% die in palliative care units. 8 However, the situation is quite different for pediatric cancer patients. There is very limited use of hospices for children with cancer. 3 The percentage of pediatric cancer patients and their families who wish to die at home is relatively high. 9 Against this backdrop, it is thought that a system for providing end-of-life care for pediatric cancer patients at home is gradually becoming more widespread. Published data indicate that children with brain and solid tumors are more likely to die at home than those with hematological malignancies,10–13 a trend also observed in this study. In Japan, there are only a limited number of hospices and home settings where blood transfusions are available 14 due to cost issues and insufficient observation. However, in recent years, various innovations have been made in home health care, and the proportion of pediatric patients with hematological malignancies dying at home has been gradually increasing.
The third important finding of this study was that the percentage of home deaths varied per region. The number of deaths per region was unevenly distributed, but regional differences in death rates per 1000 population were not as high. Regional differences in the location of death appear to reflect variations in medical resources. For the elderly within Japan’s health care system, a high utilization of regular home visits is reported in densely populated areas. 15 The Kanto region, which has a high percentage of home deaths, has the largest population, four designated core hospitals, and a well-developed home health care system. In contrast, the Hokuriku and Shikoku regions have a low population density and no designated hospitals for childhood cancer. Therefore, establishing a home medical care system, especially for children, is challenging. Further studies should be conducted to address these regional differences.
The fourth important finding was that only 20% of children died at designated core hospitals, while about 50% died at collaborating hospitals, despite 40% of pediatric cancer patients being treated at core hospitals. Children’s cancer centers are organized into seven blocks across the country, with designated core hospitals for each block. Each core hospital designates medical institutions within its regional block that provide high-quality medical care and patient support for children with cancer as “children’s cancer collaborating hospitals” and coordinates with them. Close cooperation between core and collaborating hospitals is important, even during the end-of-life stage. Based on these data, approximately 90% of patients die at a hospital near their place of residence. This suggests that patients who receive treatment away from home die at a facility closer to their residence. Cancer-related deaths are rare among children, and the number of deaths per year per facility is limited, making it difficult to collect more cases. Improving the quality of end-of-life care at collaborating hospitals and supporting health care providers is an important issue.
The present study has several limitations. First, this study relied on national vital statistics and death records, which, while providing a comprehensive overview, lack granular data on individual patient histories, comorbidities, and the specific nature of end-of-life care, such as the involvement of palliative care teams or the details of home care services. This limitation prevents a deeper understanding of the specific factors driving the shift in the location of death. Second, the analysis of death location by hospital function was limited to a single year (2018) due to data availability. This cross-sectional view prevents us from examining the longitudinal trends in end-of-life care provision at designated core hospitals versus collaborating hospitals, which would be crucial for understanding the evolving care landscape. Third, while the data suggest that patients receiving treatment away from home often return to their local area for end-of-life care, the specific reasons for this transfer, whether due to medical necessity, patient/family preference, or the availability of local support systems, could not be determined. This lack of context limits our understanding of the decision-making process during the end-of-life stage.
Conclusion
This study confirms a significant decline in pediatric cancer-related deaths in Japan, highlighting the tangible success of improved medical care and the consolidation of specialized treatment centers. Simultaneously, the increasing proportion of home deaths underscores a crucial shift in end-of-life care, reflecting a growing emphasis on respecting patient and family wishes for a peaceful, home-based final journey. These findings indicate that despite the centralization of care, effective regional collaboration is essential. To further enhance the quality of end-of-life care, future efforts should focus on strengthening home health care systems in underserved regions and providing dedicated support to health care providers at collaborating hospitals, ensuring that high-quality, patient-centered care is accessible nationwide, regardless of location.
Authors’ Contributions
N.Y. conceptualized and designed the study, collected data, drafted the initial article, and reviewed and revised the article. Y.N. conceptualized and designed the study, collected data, and reviewed the article. D.T., K.M., Y.K., and J.H. conceptualized and designed the study, supervised the data collection, and critically reviewed the article. All authors approved the final article for submission and agreed to be accountable for all aspects of the work.
Footnotes
Author Disclosure Statement
The authors have no conflicts of interest relevant to this article to disclose.
Funding Information
This work was funded by the Health and Labor Sciences Research Grant (23EA1021). The funding source was not involved in the study design, the collection, analysis and interpretation of data, or writing of the article.
Declaration of Generative AI in Scientific Writing
During the preparation of this work, the authors used DeepL and Gemini in order to improve the translation’s accuracy. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.