| 1463 AYALA
et
AL
.
evaluated every 15 minutes for 75 minutes. During the intervention,
the fathers were free to spontaneously interact and care for their
child without limitations, except for the fathers in the cot group,
who were asked not to pick up the newborn infant. The staff in the
neonatal unit continuously assessed the infant's physical well-being.
After 2 hours, the newborn infants and fathers went to the ob-
stetric unit, to be reunited with the mothers, and the mothers and
their infant were then transferred to the maternity ward for conven-
tional care.
2.3 |Data collection The Neonatal Behavioral Assessment Scale (NBAS) was adminis-
tered to measure the infants' state of wakefulness 45 minutes after
delivery (Table 1). The first author (AA) or a trained research assis-
tant conducted the naturalistic observations and performed all the
measurements to ensure the reliability of the study. All the varia-
bles were then measured six times, during the first minute of every
15-minute interval. This means that the observation period spanned
45-120 minutes after birth.
The infant's ear temperature was measured using the Braun
ThermoScan thermometer (Braun GmbH, Kronberg, Germany).
Heart rate and peripheral oxygen saturation were measured on the
infant's index finger with a Saturator Masimo, which is a compact
pulse oximetry device (Masimo Corp, California, USA). The physi-
ological parameters were performed by the first author (AA) and a
trained research assistant.
Demographic data on the health of the mothers and infants
were obtained from the birth records and registered in the research
protocols.
2.4 |Data analysis Descriptive statistics are used to present the participants' demo-
graphic data; continuous variables were described by means, standard
deviation and quantiles. To compare differences in the state of wake-
fulness (NBAS) between the groups, we evaluated the trend of the re-
sponses over time analysis of covariance for repeated measurement
estimated through mixed models was performed. The evolution of
the physiological parameters has been described by a mean value and
standard deviations for each time and measurement. The comparison
between the three groups was tested by analysis of variance for re-
peated measurements. Statistical significance was set at 0.05. All data
were processed in Stata version 14.0 (Stata Corp LLC, Texas, USA).
14
This study was approved by the Ethics Committee of the
Scientific Assessment.
Metropolitan Health Service South East (D-nr16-05-2008).