Study selection and data collection
We included published or unpublished randomized controlled trials (RCT)
or quasi-RCT that included women at risk of preterm birth (before 37
weeks), and comparing any corticosteroids against each other or against
placebo regardless the dose or schedule.
The primary outcomes for the mother (however defined by study authors)
were chorioamnionitis and endometritis/puerperal sepsis, for the
foetus/neonate were neonatal death and RDS and for the child
neurodevelopmental disability at two years follow-up (blindness,
deafness, moderate/severe cerebral palsy, or development
delay/intellectual
impairment16.
The secondary outcomes were: maternal death; perinatal death; foetal
death; chronic lung disease; intraventricular haemorrhage; mean
birthweight; and low birth weight.
Three authors (KK, DC, AC) independently screened titles and abstracts
and reviewed the full-texts of the potentially eligible studies by using
EROS17 and
COVIDENCE18 software,
and independently extracted data into a pre-piloted data extraction form
including the RoB of using the Cochrane tool6. We
classified a summary RoB for each study as high risk if at least one
domain is classified as high risk and the others as low/moderate risk.