Abstract
OBJECTIVE:
Neonatal tele-homecare implies that parents of clinically stable preterm infants can manage tube feeding and establishment of oral feeding in the home. Support is provided from the neonatal intensive care unit (NICU) through a telehealth service. The aim of this study was to compare growth and breastfeeding rates amongst infants being managed in the NICU (conventional care) and by neonatal tele-homecare.
METHODS:
A total of 96 preterm infants with tube feeding requirements participated in the observational study of neonatal tele-homecare. Retrospective data in 278 preterm infants receiving standard care in the same neonatal intensive care unit prior to implementation of neonatal tele-homecare were used for comparison. Rates of breastfeeding and growth were monitored during neonatal tele-homecare. Infant weights were converted to standard deviation weight-for-age z-scores based on a reference.
RESULTS:
There was no significant difference in rates of exclusive breastfeeding between the neonatal tele-homecare infants and the controls. Among the very preterm singleton infants more neonatal tele-homecare infants were exclusively breastfed at discharge compared to the controls (p = 0.04). There was no significant difference in median weight for age z-scores at discharge.
CONCLUSION:
This study demonstrates that neonatal tele-homecare may be an appropriate model of care for the management of preterm infants outside of the hospital environment; with the added benefit of higher rates of breastfeeding at time of discharge for very preterm infants.
Background
Optimising nutritional status of preterm infants during admission in the neonatal intensive care unit (NICU) is a very important requirement to ensure optimal growth and development. Infant growth can be evaluated individually using standard deviation (SD) scores (z-scores) according to a growth reference [1]. The European Society of Pediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) recommends that the goal for postnatal growth for preterm infants is a change in weight z-score change of no more than -1 SD from birth to discharge [2]. Breast milk is recognised as the most optimal nutrition for both term and preterm infants while it provides health advantages for the preterm infant by reducing the risk of morbidity [3] and enhancing long-term neurodevelopmental outcomes [4]. For the majority of preterm infants, the final weeks of admission is primarily focussed on achieving optimal growth parallel to the establishment of full oral feeding with the transition from tube feeding to breast or bottle feeding [5, 6]. To optimise the conditions for the families during the weeks of transition from tube feeding to breast-or bottle feeding, neonatal homecare (NH) has ‘gained ground’ in clinical practice [7, 8]. NH implies that parents manage tube feeding and breast-or bottle routines at home with the support from a qualified nurse visiting the family instead of staying in the NICU until the infant has obtained full oral feedings. Studies have shown high parental satisfaction with NH [6, 8–11] and that NH may also increase breastfeeding rates [12, 13]. Further, it has been shown that infants receiving NH experience similar growth to infants staying in the NICU [12]. Providing home visits can be challenging due to the time used for travel. Few studies have investigated whether the use of telehealth (where services can be provided at a distance using communication technology) is useful for supporting parents and infants during NH. One study found that access to videoconferencing during NH could reduce the need for home visits [14]. Another study found that access to videoconferencing during NH reduced the need for hospital visits and that parents perceived that they had more scheduled visits in the NICU than needed [15]. While telehealth seems feasible during NH, none of these studies have compared actual clinical outcomes in terms of breastfeeding rates or infant growth when using telehealth. Neonatal tele-homecare (NTH), which is the focus of this study, was designed using a participatory design approach, involving clinicians and parents [16, 17]. NTH gave families with preterm infants the opportunity to return home with their tube fed preterm infant, just like the concepts of NH, except for not providing home visits but instead videoconferences.
Objective
As part of testing NTH in a clinical setting we conducted an observational study to describe growth and breastfeeding rates in preterm infants receiving NTH. Further, we investigated total length of admission and readmission rates and the utilisation rates of the telehealth service.
Methods and subjects
Setting
The study was conducted in the NICU at the Hans Christian Andersen Hospital for Children and Adolescents (HCA), Odense University Hospital in Denmark. Preterm infant’s nutritional needs were met according to national [18] and a local [19] nutritional guideline. For infants born≤32 + 0 weeks of gestational age (GA) human milk fortifier (HMF) was initiated when the infant was 7–10 days old. If the infants required HMF they were given PreNan FM85 powder (Nestlé, Vevey, Switzerland). Full enteral nutrition was reached at 160 ml/kg/day (increased to 180 ml/kg/day if no ductus arteriosus persistent or bronchopulmonary dysplasia), regardless of birth weight. When receiving HMF; calcium and phosphate levels were checked on a weekly basis. Mothers were requested to start breast milk pumping within six hours after birth and at least six times per day, until the infant no longer required tube feeding. Electric breast pumps were available in the unit. Parents were encouraged to stay in the NICU as much as possible. Mothers could spend the nights in the unit mostly besides their infant, fathers too, when beds were available. Parents were encouraged to provide skin-to-skin contact whenever possible and to take part in infant care and tube feeding. Infant weighing scales used in the NICU were Seca© 376.
Neonatal tele-homecare
Five experienced nurses from the NICU were responsible for NTH, three of them certified International Board Certified Lactation Consultants. Parents were informed about the study by the first author or the NTH nurses. When parents had provided written consent for participation in the study, the NTH nurses started the training of the parents (Table 1). Parents were provided with an iPad air 2 pre-installed with our developed app for NTH [17]. The app was integrated into the infant’s electronic medical record in the NICU. The app provided options for videoconferences and for parents to enter infant growth data for the NICU to monitor infant growth. Further, the parents could write chat messages and upload pictures and videos from the app to the NICU. The app and videoconference system met national security and privacy requirements. Further functionalities and specifications of the NTH service have been previously described [17]. Besides the iPad, iPad holder, and utensils for infant care and disposable bottles for expressed breast milk, the parents were also loaned an electronic weighing scale (Seca© 376) and were recommended to rent an electronic breast pump for home use during NTH. Parents weighed the infant at home twice a week and measured length and head circumference once a week and entered the data to the app for the nurses to monitor infant growth. Two weekly videoconferences were arranged between a NTH nurse and the parents to ensure the well-being of the infant and the parents and progress of oral feeding. Check-ups in the NICU occurred if infants were given HMF and phosphate, while weekly blood samples were required according to clinical guidelines. If infants required eye examinations for retinopathy of prematurity (ROP), ultrasound check up of previous intraventricular haemorrhage (IVH) or periventricular leukomalacia (PVL) the examination was carried out the same day as blood samples to minimise the transportation of the infants. Participation in the NTH study did not require that the mother planned to breastfeed, as NTH was for all preterm infants, regardless of maternal breastfeeding preferences. Infants were discharged when they no longer required tube feeding and presented weight gain on solely oral feeding.
Training of parents for neonatal tele-homecare
Training of parents for neonatal tele-homecare
*Tube replacements were a voluntary task for parents.
The intervention group comprised preterm infants and their parents who met the inclusion criteria for NTH (Table 2). All infants had started weight gain and were fully enteral fed when starting NTH. NTH services were delivered from November 2015 to January 2017. Since this was an observational study no sample size was calculated.
Inclusion criteria for starting neonatal tele-homecare
Inclusion criteria for starting neonatal tele-homecare
For comparison of growth and breastfeeding rates at discharge (end of tube feeding) two historical control groups of preterm infants previously treated in the NICU at HCA were used: 1) Preterm infants participating in a national study of breastfeeding rates in the Danish NICUs conducted in 2009-2010 [20, 21] and 2) very preterm infants participating in a randomised controlled trail investigating nutrition, growth, and allergic diseases among very preterm infants after hospital discharge conducted in 2004–2008 [22]. The second control group was included because there were only one very preterm infant included in the breastfeeding study. Full enteral nutrition for the control groups were reached similar to the intervention group. If the control infants required HMF they were given Enfamil HMF powder (Mead Johnson, Chicago, USA). Indication for HMF was the same in all groups (intervention and controls). Further, indication for blood samples, eye examinations for ROP or ultrasound check up of previous IVH or PVL were the same in all groups. There was no available homecare program in 2004–2008 or 2009-2010, so all control infants remained in the NICU until end of tube feeding.
Baseline data in infants and parents were collected by the NTH nurses. From commencement of NTH; weight (gram), length (cm), head circumference (cm) and nutritional status was measured. Nutritional status referred to the infants being categorised as exclusive, partial or no breast milk. Further, we recorded whether the infant received HMF or not. On the day of discharge; weight, length and head circumference was measured again and nutritional status was registered as: exclusive breastfed, partly breastfed, or not breastfed. Data were manually entered into a RedCap database, to ensure integrity of information. In this study exclusive breastfeeding was defined as being fed solely with breast milk, according to the definition from the World Health Organisation (WHO) [23].
The number of videoconferences and check-ups in the NICU were documented by the NTH nurses. The use of weight registrations, documentation sharing (nutritional information sheets) and chat messages were monitored by the IT-company that developed the telehealth service, using a code for tracking usage.
Ethics
The study was approved by the Danish Data Protection Agency (2008-58-0035) and approved by the management at HCA. The study was presented to the local ethical committee. But the study did not require ethical approval, according to Danish legislation. According to the Helsinki Declaration [24] parents received written and oral information about the study and were included after signing informed consent. Parents were informed that participation was voluntary and that they at any time could withdraw from the study and return to the NICU for ongoing care.
Statistical analysis
Data were analysed using STATA (version 14.0, Stata Corporation, College Station, Texas, USA). Continuous variables presenting normal distribution are presented with mean and standard deviation and variables with a non-normal distribution are presented with median, minimum and maximum values. Comparison of groups was made using t-test for continuous variables showing normal distribution. For continuous variable presenting non-normal distribution comparison were performed using Mann-Whitney test. Categorical data were analysed using χ2/Fishers exact test. All infant weights were converted to standard deviation weight-for-age z-scores based on a growth reference [1]. P-values≤0.05 were considered statistical significant.
The analysis was performed and presented in two groups due to nutritional regimens in the NICU; 1) infants born with a GA≤32 + 0 weeks and 2) infants born with a GA > 32 + 0 weeks.
Results
Infant outcomes
Characteristics of parents to neonatal tele-homecare infants
Characteristics of parents to neonatal tele-homecare infants
During the intervention period, a total of 187 preterm infants (GA < 37 + 0 weeks) received treatment and care in the NICU. A number of 103 preterm infants met the inclusion criteria for the NTH study, but parents of seven eligible infants (two sets of twins) elected not to participate in the NTH study, due to certain circumstances; i.e. cultural differences, single mother and father not able to take paternity leave. A final number of 96 preterm infants participated in the NTH study. The remaining 84 preterm infants did not meet inclusion criteria with most of these infants being discharged from the NICU fully oral fed, not needing NTH. In two cases of twins in the NTH group, only one of the twins participated in the study, due to the other twin were discharged as it had obtained full oral feeding. A number of 27 of the included NTH infants were born with a GA≤32 + 0 weeks. The remaining 69 infants were born with GA between 32 + 1 and 36 + 6 weeks. Characteristics of the parents of the NTH infants are shown in Table 6. The control group from 2009-2010 comprised 128 preterm infants, with 21 of them born≤32 + 0 weeks of gestation. A total of four infants with a GA > 32 weeks had missing data on weight, which led to exclusion, resulting in 124 infant with complete data for analysis. The control group from 2004–2008 comprised 154 very preterm infants, all born with a GA≤32 + 0 weeks.
For infants born with a GA > 32 + 0 weeks, no significant differences at baseline between the two groups were detected, besides more multiple births among controls (p = 0.03) (Table 4). However, the infants in the NTH group were younger at birth and presented slightly lower birth weight for age z-scores compared to the controls. The NTH infants stayed in the NICU five days less than the controls. The NTH infants started NTH at a PMA of 36.1 weeks with discharge at PMA 38.4 weeks, which was significantly later than the controls who were discharged at PMA 36.9 weeks (p < 0.00). At discharge there was no significant difference in median weight for age z-scores and rates of exclusively breastfeeding.
Characteristics of infants born with a GA > 32 + 0 weeks*
*Variables presented in median (min-max) for non-normal distribution and mean (SD) when normal distribution.
For infants born with a GA≤32 + 0 weeks there was no significant difference between the NTH infants and the controls in baseline characteristics at birth, besides a higher frequency of multiple birth amongst the controls (p = 0.01) (Table 5). The NTH infants presented slightly higher birth weight for age z-scores compared to the controls. The NTH infants started NTH at a PMA of 35.4 weeks, with discharge at PMA 39.9 weeks, which was significantly later than the controls. However, they left the NICU 10 days earlier than the controls. At discharge there was no significant difference in median weight for age z-scores between the NTH infants and the controls (p = 0.06). Slightly more infants in the NTH group were exclusively breastfed at discharge compared to the controls (78% vs. 62%) (non-significant). For singleton infants significant more of the NTH infants were exclusively breastfed at discharge compared to the controls (84% vs. 60%, p = 0.04) (Table 5).
Characteristics of infants born with GA≤32 + 0 weeks*
*Variables presented in median (min-max).
Six infant required treatment in the NICU during NTH. (Table 6). The parents of two infants experienced breathing difficulties which required assistance from the parents and further support in the NICU. One un-planned videoconference was requested by the parents for breastfeeding guidance. Unplanned check-ups in the NICU occurred seven times, including the six leading to treatment of the infant in the NICU. All six infants requiring treatment in the NICU during NTH resumed to NTH due to parental request.
Infants requiring treatment in the NICU during neonatal tele-homecare
A total of 381 videoconferences were carried out between nurses and parents during the study period. Parents entered 907 weight registrations and wrote 526 chat-messages. Nurses sent 631 nutritional sheets to the parents describing information such as quantities of milk, fortifier, phosphate and vitamins to give the infant. Nurses wrote 1119 chat-messages, including the standard status message after each videoconference or a planned check-ups in the NICU. Besides the use of the telehealth service, there were 245 planned check-ups of the infants in the NICU for blood tests, eye examinations for ROP or ultrasound check up of previous IVH or PVL.
Discussion
This is the first study evaluating both growth and breastfeeding rates among preterm infants receiving NTH. Our study demonstrates positive results on breastfeeding rates. The analysis identified that slightly more NTH infants born with a GA > 32 + 0 weeks, were exclusively breastfeed at discharge compared to the historic controls. For infants born with a GA≤32 + 0 weeks we also identified an increase in exclusive breastfeeding rates at discharge, and the increase was significant for singletons. Studies have shown that low GA at birth is associated with less breastfeeding a discharge [20, 25] and a Swedish study has showed decreasing breastfeeding rates at discharge among preterm infants during the last ten years [26]. A study investigating facilitators and barriers to maternal milk supply identified a stressful NICU to be a barrier for maternal milk supply [27]. Facilitators for maternal milk supply are that parents contribute to infant feeding and well-being and have the opportunity to hold and connect with their baby [27] which NTH highly allows. While other studies of NH also have identified increased breastfeeding rates [12, 13], it might indicate that prolonged NICU admission is not a facilitator to breastfeeding establishment, but rather that sufficient support is provided to the parents at home. Our findings show that increased breastfeeding rates can be achieved through a telehealth support for families at home instead of in the NICU.
It was also identified, that there was no significant difference in z-score at discharge between the groups. However, the weight z-score was lower at discharge than at birth for both the NTH infants and the controls. One study of NH has presented infant z-scores [12] similar to our study, showing lower z-scores at discharge compared to birth z-scores. Overall, the NTH infants fulfilled the recommendations from ESPGHAN on postnatal growth by not decreasing more than 1 SD in weight z-score from birth to discharge. The NTH infants were discharged significantly later than the controls. An explanation to the later discharge could be that when being at home, there is no hurry for discontinuation the feeding-tube and the families have more patience in establishing breastfeeding. However, the PMA at NTH discharge was similar to other studies of NH [5, 15]. Unfortunately, we do not have data on growth, breastfeeding or PMA at discharge from infants not participating in our study, which is why we have compared with historical controls. Our finding of readmission rate on 6% are similar to other studies of NH, which have been presented to range from 5% [7] to 10% [12]. The NTH study showed that there was a need for 245 planned check-ups in the NICU. If the study had been conducted in a NICU not having guidelines requiring weekly blood tests when receiving HMF and/or phosphate, there might have been fewer visits in the NICU and more videoconferences. The American Academy of Pediatrics recommends observing for cardio-respiratory compromise in preterm infants while placed in a car seat prior to discharge for 90–120 minutes [28] however, this is not standard practice in NICUs in Denmark. During NTH there was one episode of desaturation/apnoea during transportation to the NICU, but this infant suffered from respiratory syncytial virus, which is a well-known risk-factor for apnoea in preterm infants.
Our work in this field to date has demonstrated that NTH is an effective way of supporting family-centered care [11] with infant outcomes comparable or better than conventional models of care for clinically stable preterm infants. Clearly NTH has potential for preterm infants and their families, by providing expert care in the comfort of the family at home. When studying new initiatives in clinical practice economic factors must also be considered. Resource-and utility data of NTH has been collected during the NTH study and these data will undergo a cost-analysis and be published.
Limitations
This study was a small scale study to describe infant outcomes after discharge comparing historical controls with care provided by NTH. We acknowledge the risk of bias in the historical data due to potential changes within the organisation and changes in clinical treatment guidelines. The aim of our study was to describe breastfeeding rates and growth of infants managed by NTH. A larger study over a longer period of time would be useful to detect even more rare events such as sudden infant death syndrome (SIDS), as episodes of SIDS have reported to occur during NH [7, 12].
Conclusion
Our observational study demonstrated that a higher proportion of infants born with a GA > 32 weeks and a significant higher proportion of infants born with a GA≤32 + 0 receiving NTH were exclusively breastfed at discharge compared to historical controls. Growth rates of NTH infants did not differ from infants receiving standard care in the NICU. This study demonstrates that it is possible to establish breastfeeding in the home with NTH support without risk of growth failure in preterm infants and NTH may be an alternative model of care for the management of preterm infants outside of the hospital environment. This model of care could also be considered in cases where traditional homecare services are not available–or restricted because of logistical and economic factors.
Disclosure statements
None.
Footnotes
Acknowledgments
The authors would like to thank the infants and parents participating in the study and Medware for developing the telehealth service used for Neonatal tele-homecare. Further, the authors would like to thank the Neonatal tele-homecare nurses Inge Svarer, Johanne Fabricius, Charlotte Niebuhr, Birte Møller and Sonja Jensen.