Study design and population
This was a cross-sectional study of infants discharged from eight NICUs in April 2018. The protocol and standard operating procedures were designed by the academic study group along with international experts, with contributions from each hospital’s NICU medical and nursing staff.
The eight NICUs were self-selected from a network established under China’s Premature Birth and Preterm Infants Intervention Program, these hospitals were promoting KMC and had incorporated it into their routine practice. The hospitals from seven different provinces are located in major urban cities, two are teaching hospitals and six are specialized maternal and child-health care hospitals. All are tertiary level hospitals, the number of beds in eight NICUs rang from 22 to 60 (on average 33).
Over a two-year period, staff from the eight hospitals participated in training sessions on KMC practice and management, these trainings were delivered by international and domestic experts, in addition twice yearly workshops were held to enable staffs from participating hospitals to share their experiences of KMC implementation.
Our study population was all preterm infants (GA<37 weeks at birth) cared for in the eight NICUs. All preterm infants were discharged in April 2018, but their date of birth varied with the earliest born in January 2018. GA was determined by both early antenatal ultrasound records and clinical examination on admission to the NICU. Hospital neonatologists decided which newborns were eligible for KMC (Figure 1). This was based on the neonatologist’s perception of the balance of benefit and risk KMC posed for each newborn, dependent on their clinical condition. Provided a newborn’s parents were willing, intermittent KMC could then be provided. Due to the limited availability of space and staff in all eight units, it was not possible for all eligible newborns to receive KMC at the same time, therefore an appointment system was implemented, with parents booking time slots for KMC provision. Newborns receiving one or more sessions of intermittent KMC during the entire hospital stay were defined as the “KMC group”, others were defined as the “non KMC group”. Babies not considered eligible for KMC were those deemed by physicians to have very unstable vital signs, catheters in place preventing prone KMC position, and those whose parents refused to sign informed consent; NICU preterm babies considered by clinicians to be less benefited by KMC (ie, “low risk” infants) were not offered KMC given the limited available NICU space for KMC beds and the aim to provide babies who would experience greater KMC benefit if safely provided. Babies who never received KMC during the entire hospital stay for mainly these reasons were defined as the non-KMC group. In brief, the non-KMC group was largely comprised of two different groups – those considered too unstable for safe KMC; and those considered by medical staffs as “low risk” and less likely to benefit from KMC than smaller and/or sicker babies.
KMC was practiced using a standard protocol and flow chart (Figure 2) developed for this KMC project, mostly adapted from international KMC practice guidance . Each pilot hospital took part in the development and training of the protocol and KMC flow chart.
Intermittent KMC was conducted in the NICUs on a lounge chair next to the baby’s incubator or bed. The baby was placed in the kangaroo position, skin to skin on the chest of its mother or father. Each KMC session last from 30 minutes to two hours, there’s usually 30 time slots available per day and NICU (usually from 8:00 to 20:00). The duration of each KMC session was determined by the availability of staff and space on the NICU, the infants’ condition and the parent’s availability. Likewise, the number of lounge chair next to the baby’s incubator or bed usually determine the number of slots available. NICUs and families were encouraged to practice KMC for longer duration and more sessions (>5) during the entire hospital stay as the project progressed, based on increasing confidence of clinical staff and adaptions made in NICU environment and practice routines. During KMC, nursing staff were available to help and support the parents, they monitored the infants heart rate, respiratory rate and oxygen saturation levels. If the newborns condition deteriorated or emergency intervention was required, KMC was discontinued immediately.
Measures and variables
General information was collected from the hospital records of all discharged preterm infants whether they received KMC or not. This included GA at birth, weight on admission and discharge, and length of hospital stay. For newborns who received KMC, each KMC session was documented by a nurse using a daily recording sheet that included information on the length of the KMC session, the infant’s age, gestational age (CGA), daily weight, vital signs and details of any respiratory support needed. The sheet also included the nurses name and the name of the parent providing KMC.
Parents providing KMC were surveyed after two or three KMC sessions. The survey questionnaire was administered by NICU nurses with the intention of collecting information on parents’ knowledge and experience of KMC provision (See supplementary appendix). Questionnaire included closed-ended questions on KMC information (source of information), KMC experience (any assistance from medical staff, perceived benefits of KMC, intention to continue KMC after discharge) and KMC preference (best time period, duration and frequency of KMC). We also included the following questions in the questionnaire: “Did you receive family member accept?”, “How did medical staff help you during KMC?”, “What are the benefits of KMC for you?”.
All data was entered onto a specially designed Epidata database. A nurse from each participating NICU was identified and assumed responsibility for data entry after receiving training on data collection. Epidata data from each hospital was compiled in a central location (Peking University First Hospital) for review, reconfirmation and analysis by the authors.
Birth weight was measured using the scales available in each NICU. Respiratory support was defined as persistent oxygen via nasal cannula, invasive or non-invasive ventilation therapy. KMC exposure hours was the sum of each child's total exposure time over the course of his/her hospital stay.
Data analysis was performed using SPSS for Windows version 20 (IBM Inc., Chicago, IL). The test level was set at α=0.05, and p < 0.05 was considered statistically significant. Outcomes were presented for the full study population and separately for the KMC and non KMC groups. Percentage and chi-square were used to analyze subject questionnaires. Statistical significance was determined by chi-square, t-test or by Wilcoxon rank sum test based on the outcome of interest.