To the best of our knowledge, this is the first nationwide study in Japan on unplanned readmission and out-of-hour ED visits among children who underwent tracheostomy. This study adds to previous studies by analyzing risk factors for readmissions and out-of-hour ED visits, considering factors related to patient characteristics and those that may affect medical resource use, such as differences between hospital bed size, distance to hospitals, and type of clinical course.

We detected 2308 children who underwent tracheostomy in our database. According to all Japan open data on receipt information known as the “National Database of Health Insurance Claims and Specific Health Checkups of Japan Open Data [24]” categorized by surgical procedure, 2564 cases occurred in the 0–19-year-old age group (474 cases in the 15–19-year-old age group) between April 2016 and March 2019. Although the age classifications differ slightly, our database is estimated to cover over 90% of pediatric tracheostomy in Japan.

Additionally, 20% of patients were readmitted within 90 days after tracheostomy, and > 40% were readmitted within 180 days after index hospitalization. While most previous studies were conducted over a short period, our results highlight the importance of long-term follow-up for children who undergo tracheostomy. The characteristics of our study cohort were similar to those in previous reports, with half of the cases being < year of age, approximately 40% on mechanical ventilators and tube feeding, and about 70% having NIs. Surprisingly, over 60% of the tracheostomies were unplanned, which has not been reported previously. Therefore, appropriate assessment of tracheostomy and daily care in cases of deteriorating health conditions is important, and guidelines on the indications for pediatric tracheostomy in chronic conditions are required.

Respiratory-related diseases were the most common cause of readmission, consistent with previous studies where respiratory-related diseases, such as respiratory failure and airway infection, were the most common causes of readmissions [4, 8, 9]. This finding indicates the importance of expectoration, education of care, and prevention of respiratory infection among CMCs.

Comparing our study with previous research, the readmission rates in Japan were lower than those in other developed countries: 30 days (17% vs. 18–45%, respectively) [1, 8, 24], 90 days (25% vs. 44%, respectively) [4], and 180 days (34% vs. 63–66%, respectively) [9, 25] after tracheostomy. This is partly because this study was limited to readmissions to the same hospital and those for treatment, which may have underestimated hospitalization rates. Differences in the indications for tracheostomy, discharge criteria, and healthcare systems may have also influenced this disparity. Most studies did not consider avoidable or unavoidable readmissions. Therefore, further studies on unavoidable readmissions and long-term outcomes are required due to the internationally high readmission rates from long-term assessments.

In this study, < 1 year of age, NI, tube feeding, and ventilation support were identified as risk factors for readmission within 180 days after tracheostomy. Infants and mechanical ventilators have been identified as risk factors for readmission and frequent medical resource use in previous reports [9, 21, 26]. Younger patients, especially those < 1 year of age, experienced increased medical complexity with a higher risk of mortality and adverse events [10]. Additionally, infants are known to be immunologically vulnerable, and their caregivers, who just started raising their child, are usually unfamiliar with care; even if they have no medical technologies, they are potentially at risk of using hospital resources. Complex cases, such as those with NIs and children depending on medical technology, are known to use a large proportion of medical resources [6, 8]. Therefore, preparing sufficient care training and dense care plans, such as home-visiting nurses or doctor plans, before initial discharge for these children is advisable [27, 28].

In our cohort, 220 (20%) patients visited the ED within 180 days of tracheostomy. Frequent ED visits have been associated with readmission [20]. The risk factors tended to be similar to those for readmission, although children using only HOT had a higher risk of ED visits than those on mechanical ventilation, and patients living farther from the hospital had a lower risk of ED visits. Patients living long distances from hospitals usually visit multiple hospitals [19]. Since we only tracked visits to the same hospital, a potential bias exists in this study, and the frequency of visits may have been underestimated, particularly among the group with a hospital distance of > 20.7 km. Additionally, home ventilators may not be a risk factor for ED visits since severely complex cases, such as children depending on home ventilators, are typically supported by home-visiting doctors and nurses in Japan.

We found that unplanned tracheostomy was associated with a higher risk of frequent out-of-hour ED visits than planned cases. While guidelines for managing pediatric patients undergoing tracheostomy in the acute care setting exist [29], no evidence currently links the clinical course before tracheostomy to hospital resource use. Therefore, further research is needed to identify why unplanned tracheostomies are performed and why ED visits are frequent, which may reveal important factors in addressing this issue. Tracheostomy after appropriate evaluation during non-emergency conditions reduces the risk of frequent out-of-hour ED visits in children with unstable airways.

Furthermore, < 1 year of age was a common risk factor, and children depending on medical technologies have some risk of medical resource use. Emergency visits and subsequent hospital admissions have been reported to increase the length of hospital stay and lead to increased costs [30]. Therefore, providing close care planning and patient education for children with these risk factors to minimize the use of medical resources is important, not only to improve patients’ health outcomes and quality of life but also to address the issue of social costs. The care of children with medical technology, such as tracheostomy independence, places a heavy burden on families [31]. Consequently, further research into the needs of the family is required, as it is necessary to secure personnel and improve the uneven distribution of home medical care in order to expand the welfare system and services to ensure that the burden of care does not fall solely on the family.

This study had several limitations. First, the characteristics of this database imposed certain constraints on the level of detail and accuracy of the data. We could only track readmissions and visits to the same hospital. However, CMCs are generally less likely to be transferred to other hospitals; therefore, we focused on patients who planned on attending the outpatient department of the hospital in which they were admitted for their index hospitalization. This is related to the reason for the follow-up period, 180 days; more patients tend to visit other hospitals as the observation period lengthens. Although we selected unplanned admissions among readmissions within 30 days after tracheostomy, we adopted admissions for treatments in the absence of the same information between 31 and 90 days after the index discharge, which may have included a small number of hospitalizations for respite. Our database system did not capture daytime ED visits, and we could not investigate the reasons for these visits. In addition, we could not obtain detailed information on the characteristics of the home ventilator and family background environment (structure, income, and educational standards of their parents). Second, disease classification of comorbidities was based on previous reports since no validated classification of diseases exists. Third, there may be differences in ICD-10 coding between hospitals. Differences in billing reimbursement might also affect the choice of codes for a patient during hospitalization. Finally, regional differences in the density of support by home-visiting doctors and nurses may have been associated with readmissions; however, we could not account for this due to a lack of detailed data on home care medicine.

In conclusion, we found that 43% of children required unplanned readmission after tracheostomy, and 20% experienced frequent out-of-hour ED visits within 180 days after tracheostomy. We identified age < 1 year, tube feeding, NI, and ventilation support as risk factors for readmission. Age < 1 year, HOT, a hospital distance of > 20.7 km, and unplanned tracheostomy as risk factors for out-of-hour ED visits within 180 days of tracheostomy. Therefore, estimating these risk factors before the index discharge would be helpful in coordinating appropriate home care plans and reducing preventable medical resource use. This study may help improve health outcomes, healthcare plans, and evidence-based policymaking. However, further research taking into account for additional factors may be required to validate our findings.