In course of the SafeBoosC III trial, we would like to analyse routinely performed GMA as secondary outcome parameter in an ancillary observational study.

First aim is to analyse in surviving neonates included into the SafeboosC III trial any differences between the experimental group group and control group in Global-GMA at term age.

Second aim is to investigate GMOS at term age and if available GMA at 9-16 weeks corrected age.

OBJECTIVE In course of the SafeBoosC III trial, we would like to analyse routinely performed GMA as as secondary outcome parameter in an ancillary observational study.

First aim is to analyse in surviving neonates included into the SafeboosC III trial any differences between the experimental group group and control group in Global-GMA at term age.

Second aim is to investigate GMOS at term age and if available GMA at 9-16 weeks corrected age.

HYPOTHESES We hypothesise that the surviving preterm neonates in the experimental group of the SafeBoosC III trial show better results in Global-GMA at term age compared to surviving neonates in the control group.

Furthermore, we hypothesise that the surviving preterm neonates in the experimental group of the SafeBoosC III trial show better results in GMOS at term age and in GMA if available at 9-16 weeks corrected age.

METHODS Trial design It will be a pilot observational ancillary study to the SafeBoosC III trial.

Eligibility Preterm neonates included in SafeBoosC III trial are eligible to be included in this ancillary study. Neonates will be included, in whom GM assessment is performed routinely at corrected term age or before discharge and optionally at a corrected age between 9 to 16 weeks. The infants are recorded according to routine after feeding, during periods of active wakefulness and lying in a supine position for 10-20 minutes. To evaluate the GMA visual Prechtl Gestalt perception are used. Because of the routine use in GMA, handling should reach the quality parameters, necessary for an optimal interpretation of these results.

Outcome measures Primary outcome measure for this ancillary observational study will be Global-GMA at term age Secondary outcome measure for this ancillary observational study will be GMOS at term age and if available GMA at a corrected age between 9 to 16 weeks.

Blinding Interpretation of Global-GMA and GMOS will be performed by analysing pseudo-anonymised video-recordings by persons trained in GMA and not involved in the care of the neonates.

GMA Data Global GMA will be classified as normal and abnormal (poor repertoire, chaotic, cramped synchronised at term age/absent fidgety movements at corrected age of 9-16 weeks). GMOS will be scored from zero to 42.

Data managing plan Demographic data will be handled according the data managing plan of the SafeboosC III protocol in its latest version.

GMA and GMOS data will be anonymised at each centre participating in this ancillary study with the local patient ID used in the SafeboosC III trial and handled according to local guidelines. Local statistical analyses will be performed in each participating centre in this ancillary study and just the results of these statistical analyses will be transferred to Medical University of Graz, Graz, Austria for overall/meta-analyses.

Statistical analyses Baseline characteristics of neonates will be given as mean and standard deviation or median and interquartile range for continuous data and as numbers and percentages for categorical data. Comparison of baseline characteristics will be done by using t-Test or Mann Whitney U-test for continuous data and Chi-square test or Fisher's exact test for categorical data. To answer the primary hypothesis whether global GMA differ between neonates of the control group and neonates of the intervention group OR will be calculated within each center. To analyze the secondary outcome measure GOMS a modified version of Agresti's (1980) generalized odds ratio (genOR), which accounts for ties, will be calculated within each center.

genOR=(Prob(Y_1<Y_2))/(Prob(Y_1>Y_2)) Since data are collected in the same study and therefore the study designs are the same, we assume the treatment effect to be the same and differences in the results between centers are due to a random error. As a consequence data will be combined using fixed effect models. Overall OR for the global GMA and the GOMS will be calculated and Forest plots drawn.

Inclusion Criteria:

• Inclusion in the SafeBoosC III Trial
• Signed informed consent

Exclusion Criteria:

• Missing written parental informed consent
• Decision not to conduct full life support

Neonates will be included, in whom GM assessment is performed routinely at corrected term age or before discharge and optionally at a corrected age between 9 to 16 weeks.