Health Service

Children with medical complexity receiving home healthcare devices in Japan: a retrospective cohort study

Abstract

Background Limited evidence exists regarding children receiving home healthcare devices (HHDs). This study aimed to describe the range and type of HHD use by children with chronic medical conditions in Japan and explore factors leading to increased use of these devices.

Methods This retrospective cohort study was conducted using data from the National Database of Health Insurance Claims and Specific Health Checkups of Japan. Children receiving HHD aged ≤18 years between April 2011 and March 2019 were included. Children newly administered HHD between 2011 and 2013 were followed up for 5 years, and logistic regression analysis was performed to assess the relationship between increased HHD use and each selected risk factor (comorbidity or types of HHD). The models were adjusted for age category at home device introduction, sex and region.

Results Overall, 52 375 children receiving HHD were identified. The number (proportion) of children receiving HHD increased during the study period (11 556 [0.05%] in 2010 and 25 593 [0.13%] in 2018). The most commonly administered HHD was oxygen (51.0% in 2018). Among the 12 205 children receiving HHD followed up for 5 years, 70.4% and 68.3% who used oxygen or continuous positive airway pressure, respectively, were released from the devices, while only 25.8% who used mechanical ventilation were released from the device. The following diagnosis/comorbidities were associated with increased HHD use: other neurological diseases (OR): 2.85, 95% CI): 2.54–3.19), cerebral palsy (OR: 2.16, 95% CI: 1.87 to 2.49), congenital malformations of the nervous system (OR: 1.70, 95% CI: 1.34 to 2.13) and low birth weight (OR: 1.68, 95% CI: 1.41 to 2.00).

Conclusions This study provides nationwide population-based empirical data to clarify the detailed information regarding children receiving HHD in Japan. This information could assist healthcare professionals in improving the quality of life of these children and their families and help health policymakers consider measures.

What is already known on this topic

  • Children with medical complexity, characterised by underlying chronic health conditions and significant functional limitations, account for a large proportion of healthcare use and expenditure for all children.

  • Less is known about when and how many children receiving home healthcare devices (HHDs) are released from their devices and what risk factors are associated with increased HHD.

What this study adds

  • This study provided fundamental information about children receiving HHD in Japan, revealing its trend in the total number and proportion of children receiving HHD, a wide variation in the proportion of children receiving HHD released by the types of devices, and risk factors associated with the increased HHD use during the 5 year follow-up.

How this study might affect research, practice or policy

  • This study would help guide healthcare professionals, improve the quality of life of children receiving HHD and their families and facilitate evidence-based policymaking.

Introduction

Enormous advances in medical technology have saved many children’s lives and contributed to improving their survival rates and outcomes. The prevalence of children with medical complexity (CMC), characterised by underlying chronic health conditions and significant functional limitations, is increasing.1 2 Despite the small proportion of children, they account for a large proportion of healthcare use and expenditure for all children.2 3 CMC were followed up by healthcare professionals (including home visiting doctors/nurses) and periodical hospital/clinic visits for maintenance, treatment and counselling.

Recent efforts have attempted to reveal important information regarding CMC from various perspectives, including home healthcare use,3 evaluation of care coordination/programmes4 5, medication safety6 and their caregivers.7 8 However, evidence remains inadequate for these children, especially children receiving home healthcare devices (HHDs). For example, (1) data regarding how many and when these children are released from their devices or are still using them are not well reported, and (2) epidemiological studies on children receiving HHD are also scarce; factors associated with increased HHD use in these children have not been investigated. Although previous studies have reported the benefits of home medical care for CMC, such as fewer hospitalisations and ED visits, decreased morbidity, lower inpatient costs and a lower likelihood of readmission,9–11 this basic information regarding children receiving HHD is essential for these children, their families, health service providers and health policymakers.

This study aimed to describe the range and type of HHD use by children with chronic medical conditions in Japan and explore factors leading to increased use of these devices.

Method

Data source

This retrospective observational study used data from the National Database of Health Insurance Claims and Specific Health Checkups of Japan (NDB), a nationwide claims database in Japan. The NDB covers >95% of all medical care claims for the entire Japanese population of >127 million people because Japan provides universal health coverage with a very small selection bias.12 The only exceptions constitute less than 2% of the population, who receive public income support and are covered by automobile liability and workers’ compensation insurance. The NDB data included information on anonymised personal identifiers, age group, sex, administration dates, description of the procedures performed, International Classification of Diseases (ICD-10) diagnostic codes, medical care received, medical examinations conducted without the results and prescribed drugs with the local claim codes of Japan. The data stored in the NDB are anonymised to protect personal information, but a unique identifier is applied to each patient, enabling longitudinal follow-up. The NDB has also been used in clinical epidemiological studies.13–16

After linking all the claims of children aged 18 years and younger, we identified children receiving HHD aged 18 years and younger between 1 April 2010, and 31 March 2019, from the NDB. We defined children receiving HHD using Japanese Procedure codes regarding HHD: respiratory support (C107: mechanical ventilation including noninvasive positive pressure ventilation (NPPV), C107-2: continuous positive airway pressure (CPAP) for chronic heart failure and sleep apnea syndrome, and C103: oxygen), tracheostomy care (C112), nutritional support (C105, C105-2, and C105-3: tube feed and C104: total parenteral nutrition) and renal urological support (C106: urinary catheterisation and C102 and C102-2: dialysis). We estimated the use and duration of HHD according to the rules of the medical service fee. Details of this methodology are provided in the online supplemental file. Data about individual-level characteristics were extracted: age when the home device was introduced, sex, major diagnosis and comorbidities and region. We only used the first duration of HHD. The duration was not aggregated when HHD was re-introduced during the study period. For longitudinal analysis, we included children newly administered HHD between 2011 and 2013 with a 5 year follow-up but excluded children who died or were aged 19 and above during the 5 year follow-up period.

Age was categorised into six groups: <1, 1–2, 3–5, 6–11, 12–14 and >15 years. Major diagnosis and comorbidities were categorised as congenital malformations of the nervous system (ICD-10: Q00-Q07), chromosomal abnormalities (ICD-10: Q90-Q99), other congenital malformations (ICD-10: Q08-Q89), cerebral palsy (ICD-10: G80), other neurological diseases (ICD-10: G00-G99, excluding G40-G45, and G47), neoplasms (ICD-10: C000-D48), low birth weight (1500–2499 g, ICD-10: P071b) and very low birth weight (including extreme low birth weight, <1500 g, ICD-10: P071a, P070). The duration of HHD use was categorised as<1, 1–2, 2–4, 4–6 or 6+ years. The region was acquired based on 47 prefecture-level data, used for visualisation and classified into eight areas for statistical analysis. The number of children was obtained from Population Estimates, Statistics Bureau of Japan, as of October first of each year.

Statistical analysis

Continuous variables are summarised using descriptive statistics (mean±SD for values with normal distribution and median with IQR for values with non-normal distribution), and categorical variables are summarised as frequencies and proportions. A Mann–Whitney U test was performed to assess between-group differences for continuous variables, whereas a Fisher’s exact or chi-squared test was performed for categorical variables. Univariate and multivariate logistic regression analyses were performed to assess the relationship between increased HHD use and selected risk factors (comorbidities or types of HHD). Multivariate logistic regression analyses were adjusted for age category at home device introduction, sex and region. All statistical analyses were performed using R software, version 3.6.3 (R Foundation for Statistical Computing, Vienna, Austria). Data visualisation was performed using Tableau Software version 2022.2 (Tableau Software LCC, Seattle, USA). The analyses were two-tailed, and p-values of <0.05 were considered statistically significant.

This study was approved by the Institutional Review Board and Ethics Committee of the National Centre for Child Health and Development, Tokyo, Japan (no. 2261).

Results

A total of 52 375 children receiving HHD were qualified based on the inclusion criteria, and 12 205 children newly administered HHD between 2011 and 2013 with a 5 year follow-up were identified (figure 1). The total number and proportion of children receiving HHD increased during the study period (0.05% in 2010 and 0.13% in 2018) (figure 2). Among children receiving HHD in 2018, the most administered home device was oxygen (home oxygen therapy [51.0%], followed by tube feeding [42.6%] and mechanical ventilation, including NPPV [22.4%]; table 1). The number of children receiving HHD, according to the type of device, doubled or even quadrupled, while the proportion of children receiving HHD using tracheostomy remained stable or even decreased. The number of children who were newly administered HHD by fiscal year and the type of HHD are shown in online supplemental eTable 1. While the number of newly administered tracheostomy home devices decreased (31.6% in 2018 compared with 2011), the number of children receiving HHD using multiple devices increased (264.2%).

Figure 1
Figure 1

Study population and the selection criteria.

Figure 2
Figure 2

The number and proportion of children receiving home healthcare devices. *The end of each fiscal year; **Number of childern aged ≤18 (denominator) were obtained from Population Estimates, Statistics Bureau of Japan, as of October 1st in each year.

Table 1
|
Number of children with home healthcare devices in each fiscal year

The characteristics of the children receiving HHD at the end of the 2018 fiscal year are shown in online supplemental eTable 2. The male-to-female ratio was 53.7:46.3. Almost half were introduced to home devices at less than 1 year of age (45.9%). Approximately one-third of them (30.8%) used HHD for more than 6 years, leading to an increase in the population. Regarding major diagnoses and comorbidities, other congenital malformations showed the highest proportion (22.3%), followed by other nervous diseases (21.3%) and low birth weight (11.6%).

The duration of HHD use and when patients were released from the device (health conditions improved) with selected settings during the 5 year follow-up are shown in figure 3 and online supplemental eTable 3, respectively. A wide variation existed in the proportion of children receiving HHD who were released based on the type of HHD. The results indicated that approximately 70% of children receiving HHD using oxygen or CPAP were released from the devices at the end of follow-up. Meanwhile, the results also indicated that only one-fourth of children receiving HHD using mechanical ventilation were released from their devices at the end of follow-up (25.8%).

Figure 3
Figure 3

Timing of release from home healthcare device. *Mechanical ventilation including NPPV; **CPAP is for CHF and/or SAS. CHF, chronic heart failure; CPAP, continuous positive airway pressure; FY, fiscal year; NPPV, noninvasive positive pressure ventilation; SAS, sleep apnea syndrome; TPPV, tracheostomised positive pressure ventilation.

Multivariate logistic regression analysis revealed the impact of each comorbidity and type of HDD on the increase in HHD use (table 2). Among comorbidities, the following factors were associated with a higher risk of increased HHD use: other neurological diseases (OR): 2.85, 95% CI (CI): 2.54–3.19, p<0.001), cerebral palsy (OR: 2.16, 95% CI: 1.87 to 2.49, p<0.001), congenital malformations of the nervous system (OR: 1.70, 95% CI: 1.34 to 2.13, p<0.001) and low birth weight (OR: 1.68, 95% CI: 1.41 to 2.00, p<0.001). However, neoplasms (OR: 0.55, 95% CI: 0.38 to 0.79, p=0.002), other congenital malformations (OR: 0.62, 95% CI: 0.62 to 0.70, p<0.001) and very low birth weight (OR: 0.66, 95% CI: 0.54 to 0.81, p<0.001) were associated with a lower risk of increased HHD use. Compared with no respiratory HHD, mechanical ventilation was associated with a higher risk of increased HHD use (OR: 3.39, 95% CI: 2.93 to 3.92, p<0.001), while CPAP or O2 was associated with a lower risk of increased HHD use (OR: 0.67, 95% CI: 0.59 to 0.77, p<0.001). Urinary catheterisation was also associated with a lower risk of increased HHD use (OR: 0.35, 95% CI: 0.22 to 0.51, p<0.001).

Table 2
|
Impact of diagnosis/comorbidities and types of devices on increased home device usage

Discussion

Using a nationwide claims database, this is one of the first detailed epidemiological studies on children receiving HHD, providing fundamental information. The study revealed a wide variation in when and how many children could be released from their HHD and identified risk factors for increased HHD use during the 5 year follow-up period. The provided information could guide healthcare professionals, improving the quality of life of children receiving HHD and their families and facilitating health policymakers to consider measures.

The number of children receiving HHD increased each year (figure 2), while the number of newborn infants decreased,17 although several factors may have affected the number of children receiving HHD, such as the revision of the fee table, practice in coding/claiming and changes in home healthcare delivery system. This is partly because of the increase in advanced maternal age following women’s empowerment, fertility treatments and advancements in home medical care (deinstitutionalisation). Additionally, the cohort gradually became severe/complex, owing to the increase in the proportion of devices, which required more resources such as mechanical ventilation and multiple devices. These results imply that the burden on children receiving HHD and healthcare providers will continue to increase. Family caregivers highlighted the financial burden associated with the care of a child requiring medical technology,18 while financial difficulties could be universal for all families of children with chronic illness.19 20 This study reconfirms the need for evidence-based, systematic, comprehensive and immediate action to address this situation.

From the perspective of the types of children receiving HHD, the results indicated a stable decrease in the number of children receiving HHD using tracheostomy care, partly owing to the dissemination of noninvasive respiratory home care, such as NPPV. Some families may have rejected tracheostomy because of its invasiveness and incongruence with their care goals. An increase in noninvasive respiratory home care has also been reported internationally.21

In terms of international comparison, the male-to-female ratio (53.7:46.3) was comparable with previous similar studies from the USA, where children used home healthcare in the Medicaid cohort (54.5:45.5)3 and Canada, where children in Ontario with medical complexity had a ratio of 55.4:44.6.2 From the available data, we found that (1) the crude incidence rate of children receiving home mechanical ventilation in Japan (4.1 per 100 000 children; online supplemental eTable 1) was higher than that in Canada (2.9 per 100 000 children) and (2) there was an increasing trend of these children in both Canada and Japan.22 Interpreting this comparison is challenging, as differences in morbidity, applicable criteria of HHD, intensity of care (withdrawal or not), family involvement in medical decision-making, healthcare system and culture may contribute to these variations. While direct comparison is difficult owing to the differences in ways to identify children receiving HHD, its definition, the terminology used to describe and classify CMC23 and the structure of the database, further international comparison is needed to facilitate a better understanding of the epidemiology of children receiving HHD and improve their quality of life.

Although the data were not risk-adjusted, there was wide variation in when patients could (not) be released from their HHD (figure 3). To our knowledge, this is one of the first reports on the timing of HHD worldwide. While guidelines for discontinuing home oxygen therapy in children are available,24 few recommendations exist for other HHD. Therefore, the weaning process of mechanical ventilation is based on clinical practice,25 and there is no guidance on the timing of tracheostomy closure in children.26 Further empirical data, evidence and recommendations would help understand and improve their health conditions and quality of life, including that of their families. Especially, further research on differences in morbidity, applicable criteria of HHD, intensity of care (withdrawal or not), family involvement in medical decision-making and healthcare system, which may contribute to variations in the timing of HHD release, is strongly required.

We found that neurological diseases were risk factors for increased home device use, confirming expectations but providing a first-time assessment of their impact. These neurological diseases, including intractable ones, generally have a poor prognosis and seem to contribute to this association. Additionally, we observed that low birth weight was associated with a higher risk of increased use of home devices, whereas very low birth weight was not. We suspect that this is partly because low-birth-weight infants were likely to have other comorbidities related to neurological diseases or several more severe comorbidities than very-low-birth-weight infants, and very low birth weight infants tended to be excluded from the cohort (owing to death during the 5 year follow-up and longer hospitalisation as they could not be discharged home). Further investigation into detailed major diagnoses is worthwhile to understand the characteristics of children receiving HHD for each disease.

Regarding regional variations in following up children receiving HHD, there was wide variation in the types of institutions (online supplemental eFigure 1). A similar regional variation was reported in the USA regarding the types of care and EDs.27 28 We suspect that this septal variation was related to available healthcare/community resources by region (which was not included in our dataset), clinical practice for severely ill children, including withholding/withdrawal of intensive treatment and recognition of best primary care (family-centred medical home or not).29 Indeed, a previous study indicated numerous gaps in the current system for transitioning CMC from hospital to home healthcare.30

Paediatricians and healthcare providers must understand the complexities of a child’s medical needs, family needs, prognosis, end-of-life care and available community resources.31 However, Foster et al reported that home healthcare for CMC is a ‘patchwork of policies and programmes’ that does not align with the needs of many children; there seems to be a lack of policy programmes for these children.32 Further research is required to improve policymaking regarding children receiving HHD.

Strengths and limitations

This study used a nationwide administrative database that included most children receiving HHD and had the strengths of large sample size and representativeness. Furthermore, the database enabled nationwide follow-up of the study cohort. However, this study has certain limitations. First, confounding factors could not be controlled for. For example, there is a lack of clinical information regarding the severity of each disease, despite its wide variation. Second, our study followed the children for only 5 years, and we did not follow up children receiving HHD aged ≥19 years. Third, there is the possibility of underestimation and overestimation. The number of children receiving HHD for each type of device was sometimes underestimated and/or overestimated, partly owing to the revision of the fee table every 2 years. Major changes are as follows: (1) we showed two types of tube feeding owing to the changes in the fee table in 2012 and (2) we revised the definition of home oxygen (deregulation of applying the fee in 2016). In general, the revision enables more precise capture of children receiving HHD. The number of children receiving HHD could also be overestimated because we could not detect all deaths. Deaths not resulting in an insurance claim at the time of occurrence, such as accidental deaths and those discovered later, could not be identified. The duration of HHD might be underestimated or overestimated as no precise algorithm has been established to detect the continuous use of HHD. Fourth, we did not include the socioeconomic context of children receiving HHD or families, which is currently reported to be associated with adherence to patient-centred medical home visits.33 Finally, there may be differences in the criteria for applying HHD (discharge to home or hospitalisation) among institutions/regions. We did not investigate the supply and demand of home devices for needy children. We also did not consider therapy compliance due to data unavailability, which may affect health resource use and outcome.

Conclusion

This study provides nationwide population-based evidence to clarify the detailed epidemiology of children receiving HHD in Japan as well as fundamental information. The information provided could assist healthcare professionals in making decisions, enhance the quality of life of these children and their families through more appropriate care and help health policymakers consider measures. Further research pertaining to children receiving HHD from various aspects is required.