Abstract
Introduction
Medical care for ventilator-dependent children must avoid hospital confinement, which is detrimental to the patient, their family and Paediatric Intensive Care Unit. Our objective was to assess the role of telemedicine in facilitating early and permanent discharge of such patients to home care.
Methods
This was a prospective clinical study (2007–2017) in tracheotomised ventilator-dependent children. We used a Big Data Telemedicine home system (Medlinecare 2.1) from the Paediatric Intensive Care Unit. Specialised home-nursing services were available. Clinical events were analysed using the Chi-square test (significance p < 0.05). Families subsequently completed a satisfaction survey. The Paediatric Intensive Care Unit management indicators were analysed.
Results
All of our ventilator-dependent children were included (n=12). At time of discharge from the Paediatric Intensive Care Unit, they all required continuous mechanical ventilation and met the criteria of groups I–III of the OTA classification. In the first two years there were 141 events; the main cause was respiratory (69.5%, p < 0.001) and telemedicine was the main care approach (86.5%, p < 0.001). Eleven events required hospitalisation (7.8%) but 38 (27.0%) hospitalisations were avoided. The emergency readmission time accounted for 0.99% of the total time. Six patients were decannulated, and one patient died due to primary cardiac arrest. All the families considered that the telemedicine had helped to avoid hospital visits, was not an intrusion into their privacy, and improved the child’s safety and quality of life. An improvement in Paediatric Intensive Care Unit indicators was achieved.
Discussion
Telemedicine facilitated early and permanent discharge of our ventilator-dependent children to home care without affecting their quality of care.
Keywords
Introduction
Advances in paediatrics have led to increased survival rates for many children with serious illnesses. However, some of them remain temporarily or permanently dependent on technology. The group requiring the most complex care is that of ventilator-dependent children (VDCs), and notably, the incidence of such cases is increasing.1–4 The complexity of this type of medical care explains why, even after reaching medical stability, many VDCs are confined to hospital for non-medical reasons. 3 This causes major healthcare and financial problems and is seriously detrimental to both the patients and their families. 4 Therefore, home-care programmes for VDCs have been developed and their current prevalence is estimated at 6.3–6.6/100,000 children aged under 16 years.3,5 However, although home care improves quality of life,6–10 it causes high morbidity and mortality rates.8,11–14
Telemedicine, is defined as the use of telecommunications for remote patient monitoring, diagnosis, or treatment. 14 Two previous studies have used videoconferencing for the care of VDCs with proven efficacy and safety.15,16 Telemedicine has also been used in many other medical fields in both adult17–21 and paediatric16,22–27 patients. The objective of this study was to assess the usefulness of telemedicine in the reduction of hospital confinement among VDCs by achieving early and, ideally, permanent discharge to home care, with the familial, social and financial benefits that this implies, but without affecting quality of care. For this purpose, we designed, organised and put into operation a Paediatric Intensive Home Assistance Unit (PIHAU) equipped with a more complete telemedical system than previously used, as part of our Paediatric Intensive Care Unit (PICU) services.
Methods
This was a prospective clinical study: we included all tracheostomised children who required continuous mechanical ventilation (MV) at the time of discharge from our PICU, between October 2007–May 2017. The unit is a six-bed multivalent PICU in a tertiary university hospital, which has had a four-bed Intermediate Care Unit since December 2012. This study was conducted in accordance with the amended Declaration of Helsinki, and was approved by the Hospital Clínico Universitario de Valencia's (HCUV’s) Institutional Review Board (2015/199). Written informed consent was obtained from all families before admission to the PIHAU programme.
The telemedicine programme is a health service included in the public health system in our region. We used a high-density data remote monitoring system, Medlinecare 2.1 (Medical Online Technology SL, Valencia, Spain), which allows us to monitor patients in real time from the PICU; it has a visual and acoustic medical alert for each patient and parameter. In addition, it can store and process information (a mean of 3500 records per hour) for subsequent further evaluation. Finally it includes a tele-auscultation and videoconference device via an Internet protocol (IP) camera with pan-tilt-zoom (PTZ) which can be controlled from the PICU (Figure 1). Patients were monitored during rest periods (day and night) and whenever necessary, and care was provided using equipment from several manufacturers (Table 1).

Scheme of the home telemedical system. IP: Internet protocol; SMS: short message service.
Medical equipment used in the home to administer care in the Paediatric Intensive Home Assistance Unit (PIHAU).
Analysis of the Paediatric Intensive Home Assistance Unit (PIHAU) events. Variables analysed.
MV: mechanical ventilation.
This was decided after closing each event, following independent analysis of the event by two medical team members.
For statistical analysis, these were grouped by organs or systems, on a case-by-case basis.
The PIHAU care team comprises PICU medical staff, specific nursing staff, and the patient’s usual caregivers (family, teachers, etc.). The medical staff carry out work in the hospital, including programmed care (daily rounds with telemedical data assessment, videoconferencing, tele-auscultation and interviews with caregivers when necessary) and also provide emergency care in response to an event.
The PIHAU provides three specialised nurses who attend these and other patients, undertaking conventional home hospitalisation tasks. Their daily routine begins with a telemedicine visit at the hospital and continues with school and/or home visits. This work may include respite care, and varies according to the basic needs and contingencies of each patient. Non-programmed care interventions were provided when required in order to minimise transfers to the hospital. Parents and caregivers received health training in the PICU and, afterwards, undertook a supervised transfer of care responsibility in a conventional hospital ward prior to the patient’s discharge to their home.
The results were analysed based on the three principles of the quality-of-care triangle: technical quality, perceived quality and resource management.
Technical quality
The following aspects were studied:
Aetiological diagnosis of chronic respiratory failure or other diagnoses. Home address. Readmissions to hospital and duration of hospitalisation. Prospective analysis of the ‘events’ corresponding to each patient during the first 24 months of admission. Events were defined as incidents with a clinical or technical aetiology that could be detrimental to the patient or negatively affect their basal state, which required at least one non-programmed care intervention. We used the SPSS v15.0 statistical package for statistical analysis and the Chi-square goodness-of-fit test to analyse the distributions of all the samples. A significant result (p < 0.05) indicated that the events were more clustered towards one variable more than the others (Table 2). Patient evolution. Patient’s current situation.
Perceived quality
Families were asked to complete a satisfaction survey after completing between 3–6 months of home-based care; for each question, we requested both semi-quantitative (based on 0–3 and 0–2 Likert scales) and free-text evaluations.
In preparation for school enrolment, PIHAU staff trained teachers and physiotherapists in optimised resuscitation, tracheostomy, gastrostomy and emergency response, and a nurse from our team remained in the school for as long as was considered necessary; telephone support was subsequently maintained.
Resource management
We calculated the percentage of total hospital readmission time, differentiating between programmed causes (clinical tests, programmed surgery, etc.) and emergencies. Furthermore, we studied the main PICU care indicators for the year before the PIHAU was set up and in the following five years (2007–2012). In December 2012, a High Dependency Unit was put into operation, thus making comparison with subsequent years difficult. Next, we collected data on the programmed nursing hours each child received per week and conducted a comparative study with the home care programme in our region for these patients. Finally, we recorded the evolution of parents' employment status.
Results
We included all the tracheostomised VDCs who required continuous MV at the time of discharge from the PICU (n=12); the patient’s diagnosis at admission and evolution are shown in Table 3. All the patients met the criteria for groups I, II, and III of the 1987 Office of Technology Assessment (OTA) classification for technology-dependent children; 1 75% were colonised by multi-resistant bacteria, mainly Pseudomona aeruginosa and Staphylococcus aureus. The patients’ homes were in six different healthcare catchment areas, with an average distance to the hospital of 23.3 km (range: 1.5–112 km).
Evolution of patients up to the Paediatric Intensive Home Assistance Unit (PIHAU) discharge.
ARDS: acute respiratory distress syndrome; BPD: bronchopulmonary dysplasia; CPD: chronic pulmonary disease; CRF: chronic respiratory failure; NICU: Neonatal Intensive Care Unit; PICU: Paediatric Intensive Care Unit; VPS: ventriculoperitoneal shunt.
Patients 1, 3, 4, 6 and 9 were admitted after birth in the NICU and later transferred to the PICU. Only patients 9 and 11 are female.
Age at death; btracheostomy at another hospital; cpatient required PICU admission twice: initially from birth to 25 months old, and then she was discharged for conventional home care. Again at 11.6 years old when she was admitted to the PIHAU; dn=10; patients 9 and 11 were excluded because of their late admission to the PIHAU; en=11; patient 11 was excluded because of a delayed tracheostomy; fn=7; only for patients discharged from the PIHAU.
A total of 141 events were recorded. No correlation was found with age or with the season of the year, although events tended to occur less frequently in summer (18.4%). However, they were more frequent in patients ventilated for more than 12 h/day (70.9%, p < 0.001). The main causes were respiratory diseases (69.5%, p < 0.001), followed by otorhinolaryngologic diseases (14.2%) and these were more frequently detected by parents (60.3%, p < 0.01) than by telemedicine (27.7%) or the home nurse (12.1%). Telemedicine was the principal care mode (86.5%, p < 0.001) and the main follow-up mode, both exclusively (66.7%, p < 0.001) and combined with health centre (6.4%) or hospital (17%) visits. Patients had to be hospitalised in the case of 11 events (7.8%).
We found that in 38 events (27%) the patient would have required hospital admission if the telemedicine tool had not been available. In older patients (Patients 9 and 11) we observed that in the two years prior to their inclusion in the PIHAU they required five urgent admissions, accounting for a total of 158 days. In the two subsequent years they required 11 days’ admission and only two were of an urgent nature.
During the full study period, there were 43 readmissions (0.84 readmissions/child/year, 292 stays); 69.8% occurred in the first two years and 41.9% in the first year, and 21 admissions were programmed (19 for surgical interventions). There were 22 emergency admissions (0.43 admissions/child/year for a total of 186 days). The reasons for emergency admission were respiratory decompensation in 11 (50%), seizure in five (22.7%), febrile syndrome in three, and post-surgery complications in three cases. The average hospital stay was 6.8 days (range: 1–35 days).
Only Patient 2 remains on continuous MV; after enrolling at school and more than 12 months of clinical stability, he was discharged 33.2 months after admission to the PIHAU. Patients 10 and 11 tolerate short disconnections (30 min). Patient 10 started intrathecal treatment with nusinersen at 4.6 years and after the first four doses we observed a positive evolution, both respiratory (tolerates disconnections and respiratory assistance could be reduced) and motor (recovery of some mobility: he is able to move his forearms against gravity, activates the communication switch, and has improved facial expression). Patient 9 progressed to MV > 12 h/day (rest periods). Patient 5 progressed to MV < 12 h/day and was enrolled at school at three years; although his cardiology tests were normal, at 5.3 years he died as a result of primary cardiac arrest at school, which was probably related to his underlying disease.
Patient 6 required MV only during flare-ups, and the remaining patients (n = 6) completed weaning, were decannulated and discharged without the need for any type of respiratory support or oxygen therapy (Table 3). Upon admission to the PIHAU, eight patients received nutrition exclusively by gastrostomy; currently, five patients use it only for fluid intake.
All families participated in the satisfaction survey (Table 4). The worst memories of their stay in the PICU were clinical relapses (58%) and separation from their child (25%). Although 100% felt that they had been correctly trained in the PICU, 50% felt frightened and/or anxious upon leaving it.
Satisfaction survey using semi-quantitative evaluation.
At the time of conducting the survey, only five families had had to go to hospital because of a relapse.
All the children were enrolled or re-enrolled at school. Ten children were enrolled at 3.4 ± 0.64 years, five in special education schools and five in ordinary schools. Patients 9 and 11 were re-enrolled at school and maintained a normal attendance level. Only two patients had a nurse made available to them at school.
Readmission to hospital accounted for 1.56% of the total time following discharge to home; 0.57% were planned and 0.99% were emergency admissions.
We observed a significant and sustained improvement in PICU care indicators (Table 5(a)). In comparison with the Regional Ministry of Health plan, our programme facilitated an early, significant, and sustained reduction in home nursing (Table 5(b)). In addition, parents’ employment status improved following admission to the PIHAU (Table 5(c)).
Resource management.
PICU: Paediatric Intensive Care Unit; PIHAU: Paediatric Intensive Home Assistance Unit.
(a) Evolution of PICU care indicators 2007–2012; the PIHAU was started in October 2007. We calculated the occupancy index using five beds, while saving the sixth bed for short stays. (b) Evolution of home nursing hours over time since admission to the PIHAU. Comparison with the regional home care programme for ventilator-assisted children. (c) Evolution of parents’ employment status. After discharge from the PIHAU (n=7), the employment status of the families did not alter.
Discussion
Our study only includes 12 patients. This is because of the low prevalence of children with such complex clinical characteristics (continuous invasive MV). However, our sample is similar to those published by other authors15,16,28,29 and we conducted an extended and comprehensive prospective follow-up.
The use of telemedicine allowed us to initiate home care earlier (during the continuous MV phase) and definitively, in all our VDCs. The complexity of caring for these patients is indicated by their average hospital stays of more than 10 months and the fact that they meet the OTA groups I, II and III criteria. 1 Maintaining these patients in hospital increases their morbidity, has a negative impact on families, and occupies highly technical, scarce and expensive healthcare resources.4,8 In addition, these patients are frequently colonised by multi-resistant hospital bacteria and so act as reservoirs, which is detrimental to other patients.30,31 As in other studies,3,32 in some patients we observed excessive delays in discharge for different non-medical reasons. This suggests that discharges need to prepared in advance and administrative staff informed about the benefits of specialised home hospitalisation for VDCs. In our region, this health benefit is included in the plan to improve the humanisation of critical care units, in compliance with the 1986 European Charter of Rights of Children in Hospital. 33 Additionally, telemedicine has allowed us to treat patients located far away from our hospital and avoid unnecessary transfers.
Our analysis of the events indicates that the patient age is not a barrier to discharge to home care. On the other hand, there was a correlation with respiratory symptom severity; events were more frequent in patients who needed MV for more than 12 h/day. This is logical given that 69.5% of the events were caused by respiratory decompensation, as other authors have also observed.15,28,29
Sixty per cent of the events were detected by the caregivers, probably because they are in continuous contact with the patient. However, it is noteworthy that almost 30% were detected early in the daily rounds as the result of changes in the basal values of the monitored vital signs, before their caregivers had noticed a change in the child’s condition. Therefore, the clinical information collected during sleep hours and provided by the Big Data telemedicine home system facilitated the early detection or confirmation of changes in the patient's evolution and early prescription of the necessary treatment and follow-up. Follow-up was also carried out via telemedicine because only 7.8% of the cases required admission. In addition, we estimate that 27% more cases (representing up to 10.3 days hospital stay/patient/year) would have required admission if telemedicine had not been available.
The total number of readmissions was lower than observed by Bertrand et al. in a series similar to ours (1.6 vs 0.43 episodes/child/year). 28 In addition, hospitalisations were short, with children staying at home 98.4% of the time. We have not found this information in other studies. It is impossible to know if events would have been more frequent without the telemedical information, nonetheless, the evolution of Patients 9 and 11 suggests that their severity would have been greater.
The evolution of patients’ respiratory failure was also very positive. Our results are comparable to the best published outcomes.3,28,29,34 Daily analysis of collected data enabled us to adapt the ventilator parameters to the real needs of the child, safely weaning them without the need for assessment or repeated hospital tests. Only one patient with severe myopathy still remains on continuous MV, but with an excellent pulmonary status which allowed him to be discharged from the programme because of his clinical stabilisation (Patient 2). The rest of the patients have also progressed in terms of the weaning process. Two patients tolerate disconnections, which facilitate their care and improve their quality of life, one of them (Patient 10) after being included in a clinical trial with nusinersen (unpublished results). Of interest is that all infants with severe lung disease evolved to improvement/cure, indicating that with adequate follow-up, lung growth and maturation facilitates functional recovery. 35 We also observed a simultaneous improvement in oral feeding capacity; the evolution of Patients 9 and 11 suggests that older children also benefit from this care, becoming more clinically stabilised with an improved quality of life.
In our series, one patient died from an unpreventable cause, thus giving a 8.3% mortality rate which is lower than the 17–32% reported by other authors in larger series.7,14,28,29,35–37 It has been reported that up to 19% of deaths are preventable; 36 we concur with other authors that professionalising home care in combination with adequate carer training and proper responses and surveillance by clinicians, contributes to a reduction in mortality. 38 In our experience, telemedical monitoring, along with training and motivating caregivers, was the key to avoiding problems and reducing morbidity.
Discharging children to home care improves their quality of life.6,7,9,10 However, it did create anxiety and exhaustion among non-professional caregivers 13 and so we initially included periods of family rest when necessary. In addition, as also reported by Miyasaka et al., 16 telemedicine reduced family stress levels without intruding on privacy and with excellent acceptance (Table 4). Furthermore, the school enrolment reported elsewhere varies between 45–70%,2,7,39 but reached 100% in our patients. This facilitates socialisation, children’s personal development and normalisation of family dynamics including the birth of new siblings, return to work, etc.28,40,41
Although we have not yet performed a full cost-analysis study, as in other studies, 42 we estimated that substantial cost savings could be made with our programme by avoiding hospitalisations as well as in comparison to the current regional home-care programme (Table 5). This programme provided 56 nursing hours per week with no time limit or specific medical supervision. In addition, telemedicine contributed to a significant improvement in our PICU care indicators, increasing the number of patients attended and the availability of beds.
In conclusion telemedicine has allowed us to initiate home care in all our VDCs efficiently, safely and permanently advancing in their care and improving the quality of life both for patients and their families.
Footnotes
Acknowledgements
The authors would like to express our gratitude to all the staff at the PICU for their undivided dedication to the care of children hospitalised there and their families. They especially thank Patricia Roselló, Julia Sánchez, Joaquín Donat and Cecilia Martínez for their continuous support. They also thank Maria Ledran for linguistic services and Pedro M Medina for his help with the data management and statistical advice.
Declaration of conflicting interests
The author(s) declared the following potential conflicts of interest with respect to the research, authorship and/or publication of this article: Juan I Muñoz Bonet declares, as a researcher at the University of Valencia and the Biomedical Research Institute INCLIVA, that he collaborates in the development of telemedicine solutions at the company, Medical Online Technology SL. None of the other authors have any conflicts of interest to declare.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Supported, in part, by Fundación para la Investigación del Hospital Clínico Universitario de Valencia (INCLIVA). Grant/Research Fellowship GV05/285 from Conselleria de Empresa, Universidad y Ciencia. The Comunidad Autónoma Valenciana. Spain: ‘Sistema de telemedicina para la implantación de un programa de hospitalización domiciliaria para pacientes pediátricos dependientes de tecnología’. To request any further information about this please contact the Biomedical Research Institute INCLIVA (
).
