Characteristics of first-time users of the nursing small-scale multifunctional home care service: a pooled cross-sectional study using Japanese long-term care insurance claims data from 2012 to 2019

Abstract

Objectives

In April 2012, the Japanese government launched a new nursing service called the nursing small-scale multifunctional home care (NSMHC) to meet the nursing care demands of individuals with moderate-to-severe activities of daily living (ADLs) dysfunction and who require medical care, thereby allowing them to continue living in the community. We aimed to preliminarily analyse the characteristics of first-time users of NSMHC service.

Design

This pooled cross-sectional study used the Japanese long-term care insurance (LTCI) claims data from the users’ first use of NSMHC (from April 2012 to December 2019).

Setting

NSMHC includes nursing home visits, home care, daycare, overnight stays and medical treatment.

Participants

The study population included LTCI beneficiaries who received their first long-term care requirement certification in Japan from April 2012 onwards, died between April 2012 and December 2019, and used any LTCI service at least once.

Results

Among the 836 563 individuals who used any LTCI service at least once, 3957 (0.47%) used NSMHC. We analysed 3634 individuals without any missing data regarding long-term care requirement certification. Most individuals were aged 80 years or older, with 64.3% requiring care level 3 or above, indicating complete assistance with ADLs. Regarding ADLs in individuals with dementia, 70.6% were at level 2 or below, indicating they can live almost independently even with dementia. A large proportion of NSMHC users availed the service approximately 6 months before death, with no prior use of any LTCI services; they continued using the service for around 4 months, although some people continued to use NSMHC until their month of death.

Conclusions

Using individual data on nationwide LTCI, we described the characteristics of first-time users of NSMHC among those who died within 7.5 years from the first certification of care needs. Further studies are needed to investigate the effect of NSMHC use on user outcomes.

STRENGTHS AND LIMITATIONS OF THIS STUDY.

• The Japanese government launched a new nursing service called the nursing small-scale multifunctional home care (NSMHC) in April 2012 to meet the nursing care demands of individuals with moderate-to-severe activities of daily living dysfunction and medical care, allowing them to continue living in the community.

• This is the first cross-sectional study to use nationwide long-term care insurance claims data to examine the characteristics of first-time users of NSMHC.

• As this analysis only includes first-time users of NSMHC, it cannot compare their characteristics to those who have used other community services.

• Our data encompassed individuals who have died, suggesting that our sample primarily comprised individuals with a more severe condition.

Introduction

Over 20 years ago, Japan introduced the long-term care insurance (LTCI) system and achieved long-term care universal health coverage.^{1 2} In 2000, the number and the percentage of people aged 65 years or older were 2.2 million (17.2%), increasing to over 3.6 million (29.1%) in 2022.³ With increase in the number of older adults with moderate-to-severe activities of daily living (ADLs) dysfunction, the concept of a ‘community-based integrated care system’ has been the Japanese LTCI system’s primary goal since 2005. Specifically, it aims at enabling older adults to live their lives in familiar environments even if they become heavily dependent on long-term care.^4–6

In this context, in April 2006, a small-scale multifunctional home care service (SMHC) was established to support individuals with moderate-to-severe ADL dysfunction to continue living in their community settings. This programme primarily provides daycare, home visit and overnight stay care services, and the decision to provide these services to users is based on individual conditions with a fixed monthly fee.^{7 8} However, the number of people requiring medical care has been increasing due to a growth in the vulnerable population that is transitioning from a long-term care facility to home care.^{9 10} This population, if managed accordingly, could avoid unnecessary hospitalisation due to inadequate medical apparatus and condition management.^{11 12} To address this issue, the government launched a new nursing service, called nursing small-scale multifunctional home care (NSMHC), in April 2012 to meet the nursing care demands of individuals with moderate-to-severe ADL dysfunction and who require medical care, thereby allowing them to continue living in the community. The NSMHC service complements the previously established SMHC.^{13 14} In addition to the services provided in SMHC, NSMHC provides medical care to the users 24 hours a day and 365 days a year, in collaboration with nurses and home doctors.

Nevertheless, few studies have investigated this new service. Studies related to NSMHC were primarily qualitative studies conducted from the perspective of care providers^15–19—two agency-level surveys investigated the characteristics of NSMHC agencies^{16 17}: one qualitative interview survey on 11 NSMHC nurses investigated the benefits and challenges of nursing practice in NSMHC¹⁸ and one review of care reports investigated the strengths and problems of NSMHC service.¹⁹ Although previous studies suggested that NSMHC helped individuals continue to live in their homes, avoid hospitalisation and improve physical activities among users in need of mid-to-severe levels of medical care, they were, as mentioned above, qualitative and highlighted the perspectives of nurses or agencies. No study has yet determined the characteristics of NSMHC users using individual data, which is crucial to understand which populations are using the service as expected or contrary to expectations.

Therefore, we aimed to preliminarily analyse the characteristics of first-time users of NSMHC using nationwide LTCI claims data. We also aimed to show the NSMHC service use trajectory, such as the number of months of continuous NSMHC use, and types of LTCI services from which users transitioned.

Background

Overview of the LTCI system in Japan

In 2000, the Japanese government introduced LTCI, a compulsory social insurance system to which citizens aged 40 and above contributed. LTCI services are available to patients aged 40 and above with intractable diseases and for those aged 65 and over who are certified as requiring long-term care.¹ LTCI services are divided into two main categories: in-home services, including home-visit services, day services, short-stay services, and institutional services.⁵ The co-payment for LTCI services ranges from 10% to 30% of the expenditure and the percentage is determined by the income of the person receiving care. Beneficiaries are classified into seven levels of need according to the degree of ADL and instrumental activities of daily living (IADL): support level 1 and 2, care need levels 1–5. The higher the level of care needs, the greater the need for care. The maximum monthly service entitlement is determined according to this classification; it ranges from approximately 50 000 yen (≒345 USD as of 14 January 2024) for those requiring support 1 to 362 000 yen (≒2500 USD as of 14 January 2024) for those requiring care need level 5. The procedure for obtaining LTCI service is as follows: individuals seeking LTCI certification apply to the municipality in which they reside. A qualified surveyor visits the applicant and conducts a home survey covering 74 items in five categories: physical function for ADLs (eg, sitting, standing and morbidity), daily function for IADLs (eg, feeding, toileting, housing and cleaning), cognitive function (eg, communication of intention and whether the applicant can state his/her date of birth and age), mental and behavioural disorders (eg, paranoia and confabulation) and adaptation to social life (eg, money management and taking medication). In addition to the aforementioned categories, 12 special medical treatments (eg, administration of intravenous infusion, catheters) received in the past 14 days and ADL dysfunction in people with dementia are investigated during the survey. Subsequently, the total standard time to provide the required care is calculated by a computer and a primary judgement of care need level is made accordingly. Based on the results of this primary determination and the opinion of the doctor in charge, the Needs Assessment Review Committee, comprising approximately five medical and welfare professionals, determines the applicant’s final level of care required. Individuals who receive a care needs certification should designate a care manager with at least 5 years of experience as a healthcare professional or long-term care staff. The care manager then designs long-term care service plans based on the preferences of the user and their family regarding the needed services.

NSMHC service

NSMHC is one of the LTCI services introduced in April 2012. NSMHC includes nursing home visits, home care, daycare, overnight stays and medical treatment. Access to these services requires the payment of a fixed monthly fee, ranging from 124 380 JPY (≒835 USD as of 14 January 2024) (level 1) to 313 860 JPY (≒2166 USD as of 14 January 2024) (level 5). The number of registered users in one NSMHC agency is 29 or less—19 or less for daycare, and 9 or less for overnight stays. Individuals with care need levels 1 to 5, whose care managers have determined that they need to introduce NSMHC services, are eligible to avail said services. A doctor must authorise medical treatment provided by nurses in NSMHC. Individuals using NSMHC services cannot concurrently use other home-based and short-term residential care services. Nonetheless, they can concurrently use home-visit rehabilitation, home medical care management, welfare equipment rental and home modification services.¹³ Since the opening of about 50 NSMHC facilities in 2012, their numbers have grown to 744 as of March 2022, with approximately 19 000 users per month.¹⁴

Methods

Design and data source

In this pooled cross-sectional study, we conducted a subgroup analysis of a large-scale 7-year cohort study aimed to investigate the trajectory between long-term care utilisation and certification of death. The inclusion criteria were LTCI beneficiaries who received their first long-term care requirement certification in Japan from April 2012 onwards, lost their certification due to death between April 2012 and December 2019, and used at least one LTCI service. Of the 836 563 individuals in the source population, we identified 3957 who used the NSMHC once or more. We used the Japanese LTCI claims data through the approval of the Ministry of Health, Labour, and Welfare (approval number 1008-3 (8 October 2020)). The Japanese LTCI claims data provide monthly claims data and long-term care needs certification information for all LTCI service beneficiaries. The long-term care needs certification information includes sex, age, date of certification, the result of all home-visit surveys (75 items, special medical treatment and ADL dysfunction in people with dementia), the total time required for care delivery, and primary and final judgement of the care need levels.

Variables

For the identified 3957 NSMHC users, LTCI service utilisation was investigated for all months, from the first care needs certification to death. The month of first NSMHC utilisation since the initial certification of care needs was defined as the time of first NSMHC use. As variables for demographic and characteristics indicating their level of physical and cognitive independence and medical needs available from the LTCI claims data, we used sex, age, care need level, ADL dysfunction in people with dementia, necessity of special medical treatments and total time required for care delivery, all of which were obtained from the care requirement certification information immediately before the month in which they started using the NSMHC. We also present the first-time usage of NSMHC obtained from monthly LTCI claims data.

Variables of NSMHC users’ characteristics at the time they started to use the service are as follows:

Care need levels

The five care need levels are as follows: level 1—individuals require daily assistance with any IADL; level 2—individuals require daily assistance with both IADL and ADL; level 3—individuals require complete assistance with both IADL and ADL, including those with difficulty walking independently; level 4—individuals require full assistance in all activities; level 5—individuals are almost bedridden.

ADL dysfunction in people with dementia

ADL dysfunction in people with dementia is a rating scale used to assess the level of independence in daily living. This scale was used to determine the extent to which older adults with dementia can live independently. ADL dysfunction in people with dementia is divided into the following six categories: without dementia; level I—individuals with dementia who can live independently in their homes; level II—individuals who can live independently but exhibit some symptoms and communication difficulties; level III—individuals who need care and have some psychiatric symptoms and communication difficulties; level IV—individuals who need 24-hour care and exhibit frequent psychiatric symptoms and communication difficulties; level V—individuals who need specialised medical care with significant psychiatric symptoms.

Necessity of special medical treatment

The following 12 services are assessed as special medical treatments in the LTCI certification process: administration of intravenous infusion, catheters, oxygen therapy, monitoring measurements (eg, blood pressure, oxygen saturation), tube feeding, treatment of bedsores, central venous hyperalimentation, pain management, dialysis, stoma treatment, tracheostomy procedures and respirator. The criteria for determining ‘use’ or ‘no use’ was determined by the preceding 14 days from the date of home-visit survey for care needs certification. Only when the need was ongoing, rather than temporary, was it classified as ‘use’.

Total time required for care delivery

In the LTCI certification process, the primary judgement was based on the total standard time for delivering the required care. This time refers to the sum of time necessary for providing care related to each of the eight ADLs (ie, feeding, toileting, morbidity, bathing, and grooming, housing, and cleaning, care for behavioural and psychological symptoms of dementia, functional training-related activities and medical treatment) added to the time required for delivering the services related to the ‘additional dementia care’ category. The time required for care is indicated in ‘minutes’ based on statistical data from the person’s ‘ability’, ‘care method’ and ‘the existence of disability’.

The first-time usage of NSMHC and trajectory of NSMHC service use

For variables that reveal the trajectory of NSMHC service use in the observation period(from the first care needs certification to death), we used the year when the individual started using the NSMHC; months from the first care needs certification to NSMHC use; months from the first NSMHC use to death; months of continuous NSMHC use; whether individuals used NSMHC until the month of death or not; and use of other care services in the previous, current and post-month of the first use of NSMHC.

Regarding the use of care services, for the month that preceded individuals’ first use of the NSMHC, we first divided the variable ‘care service use’ into the following three categories: ‘home services’, LTCI facility services and ‘multiple services’ (online supplemental table S1). Then, the individuals themselves in the previous month were also divided into four categories: (A) only home services, (B) only LTCI facility services, (C) multiple services and (D) no service use. Regarding the current and the post-month of first NSMHC use, we decided on the following four categories: (A) only NSMHC, (B) transition between NSMHC and LCTI facility services during the month, (C) transition between the NSMHC and other home services during the month and (D) transition between the NSMHC and multiple services during the month. Cases with NSMHC and home care services allowed in conjunction with NSMHC were categorised into group A.

Analysis

Our descriptive analyses of the numbers and proportions of individuals’ demographic status and medical and long-term care needs are shown in table 1. We graphically outlined the distribution of standard-time care delivery to individuals by categories through box plots and care service, using the transition from the previous month to the month after the first use of the NSMHC; this was visualised by using Sankey diagrams (using the Sankey plot ado file).

Table 1.

Individuals’ characteristics (N=3646)

Sex (n, %)
	Male	1990	54.8
	Female	1644	45.2
Age (years) (n, %)
	Under 65	140	3.9
	65–69	206	5.7
	70–74	334	9.2
	75–79	554	15.2
	80–84	836	23.0
	85–89	877	24.1
	ver 90	687	18.9
Months from the first certification to death (median, IQR)		27	10–51
Year of first use of NSMHC services (n, %)*
	2013	102	2.8
	2014	267	7.4
	2015	488	13.4
	2016	599	16.5
	2017	710	19.6
	2018	848	23.3
	2019	620	17.1
Care need level (n, %)
	1	565	15.6
	2	735	20.2
	3	683	18.8
	4	864	23.8
	5	787	21.7
Activities of daily living dysfunction of people with dementia (n, %)
	Without dementia	456	12.6
	I	671	18.5
	II	1438	39.6
	III	872	24.0
	IV	184	5.1
	Not available	13	0.4
Number of special medical treatments required (n, %)
	0	2300	63.3
	1	835	23.0
	2	315	8.7
	3	121	3.3
	4 or more	53	1.8
Months from the first certification to NSMHC use (median, IQR)		12	3–34
Months from the first NSMHC use to death (median, IQR)		6	2–17
Months of continuous NSMHC use (median, IQR)		4	2–10
NSMHC use until the month of death (n, %)		488	13.4

*This study included those who received their first long-term care requirement certification in Japan from April 2012 onwards, died between April 2012 and December 2019, and used any long-term care insurance service at least once.

NSMHC, nursing small-scale multifunctional home care.

To compare the proportions of transition of service use from the previous month to the following month across care need levels, Cochran–Armitage test for trend was performed. A two-sided p value <0.05 was interpreted as statistically significant. All statistical and graphical analyses were performed using Stata (V.16).

Patient and public involvement

No patients were involved.

Results

We assessed the data from 3957 individuals who used the NSMHC at least once. Among them, we analysed 3634 individuals who did not have missing information on long-term care requirement certification. The number of men and women was approximately equal, with the majority aged 80 years or older. Around 64.3% of the individuals required care level 3 or above. Regarding ADLs for people with dementia, 70.6% of the individuals were at level II or below (table 1).

Need for special medical treatment

The use of special medical treatment was right-skewed; 2300 (63.3%) NSMHC users did not require any of the special medical treatments, 835 (23.0%) required only one special medical treatment, and 489 (17.5%) required multiple special medical treatments (table 1). Among the 1353 who required treatment, intravenous infusion (41.1%) was the most frequent (table 2).

Table 2.

Need for each special medical treatment among those who need at least one special medical treatment (N=1353)

Special medical treatment (multiple answer)	n	%
Administration of intravenous infusion	548	41.1
Catheters	368	27.6
Oxygen therapy	224	16.8
Health measurements monitoring	207	15.5
Tube feeding	206	15.4
Treatment of bedsores	173	13.0
Central venous hyperalimentation	91	6.8
Pain management	83	6.2
Dialysis	78	5.8
Stoma (artificial anus) treatment	73	5.5
Tracheostomy procedures	37	2.8
Respirator	21	1.6

The sum of % exceeded 100% because the NSMHC users used multiple service.

NSMHC, nursing small-scale multifunctional home care.

Total time required for care delivery by care need levels

The median (IQRs) of the total time required for delivery of required care for each care need level can be found in figure 1 and online supplemental table S2. The longest required care time categories by care need levels were bathing and grooming in level 1, feeding in level 2 and toileting in levels 3–5, respectively. Among individuals in levels 3–5, the time required for medical care delivery varied widely, with the values being right-skewed.

Figure 1.

Standard time for required care delivery by care need level.BPSD: Behavioral and Psychological Symptoms of Dementia

The first-time usage of NSMHC and trajectory of service use

The median timing of the first use of NSMHC was 1 year after the certification of care need level and 6 months before the individual’s death. Individuals continued to use NSMHC for 4 months as the median and 14.3% used NSMHC services until their month of death (table 1).

Regarding the transition from the previous month to the month after the first use of the NSMHC, overall, 2642 (72.7%) individuals fully transitioned to only NSMHC in the month following their first NSMHC use, and others used NSMHC and other LTCI services, which included starting or stopping to use NSMHC during the month (figure 2). Among 2642 who fully transitioned to NSMHC, 59.4% had not used NSMHC services before, 28.8% had used only home care services, 7.0% had used multiple services and 4.8% had used only long-term care facility services (figure 2 A and online supplemental table S3). In care need levels 1–5 (figure 2 B-F), the proportion of individuals who had not used NSMHC services previously was 61.6%, 54.7%, 49.8%, 64.2% and 64.8%, respectively. The proportion of transition from the previous month to the following month differed significantly across care need levels (p<0.001) (online supplemental table S3).

Figure 2.

Transition from the previous month to the month after the first use of nursing small-scale multifunctional home care (NSMHC).

Discussion

This is the first study that described the characteristics of first-time users of NSMHC among those who died within 7.5 years from the first certification of care need level by using nationwide LTCI claims data and certification information data. Our two main findings were as follows: NSMHC services were used mainly by those aged 80 years and older with mid-to-severe care need levels associated with medical treatment that did not present severe dementia; a large part of the NSMHC users began NSMHC approximately 6 months before death, with no use of any LTCI services in the previous month, and then continued NSMHC for around 4 months. Even though some people continued to avail NSMHC until their month of death, this trend was observed irrespective of the care need levels.

Characteristics of users as they begin using NSMHC services

We found that people aged over 80 years and those with severe care need levels tended to begin NSMHC use. The most common care categories among individuals in need of nursing care were toileting assistance, housing/cleaning, mobility assistance and dietary assistance. Individuals with higher levels of care needs required a wider range of time for medical care; commonly, administration of intravenous infusion, catheters and oxygen therapy, with the top 25% needing a significant amount of medical care time.

Although making direct comparisons is difficult because the population in this study was limited to only those who had already died, NSMHC users may have required a higher level of care and medical care than users of home-visit nursing services or SMHC.¹⁴ This finding is consistent with those of previous studies. A literature review identified that the NSMHC service’s strength lies in ‘caring for users with high medical needs, including end-of-life care’.¹⁹ Moreover, nurses working at NSMHC facilities provide medical care services for users with higher medical needs.¹⁵ Thus, our results suggest that the NSMHC supports community residents with severe ADL dysfunction and medical needs. In FY2012, before the NSMHC was established, the delivery of these services was considered an issue. A deeper understanding of the characteristics of NSMHC users is necessary, considering disease information from national health insurance data by matching medical insurance claims data and the LCTI claims data.

The first-time usage of NSMHC and trajectory of service use

Regarding the first-time NSMHC usage, about 60% of the cases that fully transitioned to NSMHC in the month of its first use were individuals who had not used NSMHC services in the previous month. The remaining 40% included individuals who used NSMHC and other in-home or facility services in the month of its first use. This included those who stopped using the NSMHC midway through the month or those who started using NSMHC in the middle of the month, making it impossible to identify them because the LTCI claims data is aggregated on a monthly level and the dates of service usage are not identifiable. The proportions of individuals with care need levels 1, 4 or 5 were higher than those with care need level 2 or level 3. According to the database we used, individuals could be classified as not having used the services in the previous month for two main reasons. First, being hospitalised or using only home medical and nursing care provided through the National Health Insurance in the previous month. Second, they were certified as needing long-term care but did not avail the services. Since it is implausible that people with low levels of physical independence requiring care need level 4 or level 5 are not using any LTCI services, the higher percentage of ‘no use’ among those in care need level 4 or level 5 may be attributed to the former reason, especially hospitalisation. Although it cannot be determined whether those with no use were hospitalised, this study’s results suggest that NSMHC may support the transition from hospital to home. Previous studies suggested that NSMHC supports individuals immediately after discharge. More than 80% of NSMHC facilities’ administrators believed that NSMHC service take a role of accommodate users with moderate-to-severe medical needs, coordinate the transition to home care after discharge.¹⁶ Using the stay function of the NSMHC, individuals can receive nursing support (eg, instructions regarding medical techniques to users/families/caregivers) to ensure a smooth hospital–home transition.¹⁹

Additionally, we found that individuals who started using the NSMHC service did not use it temporarily, such as only for 1 month, but continued to do so for an average of about 4 months. Furthermore, 14.3% of NSMHC users continued to use the service until their death. A previous nationwide cohort study using LTCI claims data in Japan indicated that the percentage of home deaths in individuals who used in-home LTCI service was 14.3%. That is, use of in-home LTCI services contributed to deaths at home,⁵ suggesting that there is a proportion of individuals who live in the community until the end of their lives with special medical support for their ADLs, using the NSMHC service, which combines nursing home visits, home care, daycare, overnight stays and medical treatment. Before the NSMHC was introduced, administrators of SMHC agencies experienced difficulties in delivering care for individuals with severe ADL dysfunction and those in need of medical treatments¹⁶; this is because these individuals tend to be readmitted to hospitals when they experience acute symptom exacerbation. After the introduction of the NSMHC, nurses working under NSMHC reported that ‘when a patient’s condition deteriorates, he/she can stay in the NSMHC instead of being hospitalised’.¹⁸ As hospitalisation may shorten older adults’ life expectancy,²⁰ providing continuous and coordinated care to prevent avoidable hospitalisations is necessary.²¹ Future research can analyse the effectiveness of NSMHC with a more robust design to further investigate its efficacy. This could involve comparing outcomes, such as avoidance of hospitalisation or home death, between NSMCH users and non-users, considering disease information and home care provision, using national health insurance data by matching medical insurance claims data with LCTI claims data.

Limitations

Our study had several limitations. First, our data encompassed people who died; therefore, our sample comprised individuals with a relatively more severe condition. The average time from the first certification to death in this study population was 32.7 months (2 years 8 months), which was relatively shorter than in a population-based cohort study (4.37 years).²² However, according to a cohort study of all persons certified with care need at base line in a municipality, 70% of persons died within 7.5 years.²³ Since all subjects in this study died within 7.5 years, our findings regarding the severity of care needs among NSMHC users and the percentage of those requiring special medical treatment might be overestimated; there is a threat of generalisation, although it is not deemed substantial.

Second, the LTCI claims data did not include detailed services of the NSMHC (eg, overnight stay, home visiting nursing care, home visiting care or day care). We were unable to analyse which function and combination of functions were the most effective for users’ outcomes. Our data also did not include details pertaining to individuals’ medical condition and prescription.

Third, we focused solely on the initial use of NSMHC. The lack of national health insurance information in the LTCI claims data complicates determining whether interruptions in NSMHC use or use of NSMHC every few months is due to hospitalisation or utilisation of medical services.²⁴ Including all instances of NSMHC use would be complicated. Therefore, this study focused on the time of first use, as a first step in capturing the characteristics of NSMHC users. Further studies are necessary to provide an overview of NSMHC usage.

Fourth, our preliminary analysis does not allow conclusions to be drawn on NSMHC’s effects. To confirm whether the NSMHC is reaching its expected goals, future research can investigate its effectiveness by assessing demographic differences across other services’ user types, whether it contributes to death at home, and whether it avoids hospitalisation as mentioned above. Despite these limitations, this study is the first to use individual data to demonstrate that NSMHC services are used by individuals with high levels of care needs, requiring assistance in toileting, mobility, feeding and medical care, specifically among those who died within 7.5 years from the first certification of care need level. The study indicates that NSMHC may be helping these individuals live at home.

Conclusion

Using nationwide LTCI comprehensive data, we described the characteristics of the first-time users of NSMHC who died within 7.5 years following the first certification of care needs. These users were mostly aged 80 years or older, with mid-to-severe ADL dysfunction, and required medical support in terms of their demographics and status of ADL. A large proportion of the NSMHC users began using NSMHC approximately 6 months before death, with no use of any LTCI services in the previous month and then continued using it for around 4 months. Further studies are needed to investigate the effect of NSMHC on user outcomes.

Data availability statement

Data may be obtained from a third party and are not publicly available. All datasets in this study have ethical or legal restrictions by the Japanese National Government for public deposition due to inclusion of sensitive information from the human subjects.

Ethics statements

Patient consent for publication

Not applicable.

Ethics approval

This study was approved by the institutional review board of the Tokyo Medical and Dental University (C2021-005). It was conducted in accordance with the ethical guidelines for medical and health research involving human participants issued by the Japanese National Government. These guidelines stipulate the protection of patient anonymity; hence, the requirement of informed consent was waived for our study.