Welcome to the 9th EAPS Congress Programme Scheduling

Abstract Body

Air pollution, ambient temperature and perinatal and children health

Air pollution is a complex mixture of compounds present in gaseous (e.g. nitrogen oxides) or particulate (e.g. fine particles) form, directly emitted by the different sources of pollution or formed in the atmosphere as a result of chemical reactions. Numerous epidemiological studies have highlighted the effects of exposure to air pollutants on cardiovascular and respiratory mortality and morbidity. In the DOHaD (Developmental Origins of Health and Diseases) context, the last 20 years have seen an increasing number of publications reporting adverse effects of prenatal and early postnatal exposure to air pollution. Maternal exposure to air pollution and post-natal exposure of the child can have short-term effects on birth weight or prematurity, but also long-term effects on the development of the child. In the context of climate change, there is also an increasing number of studies conducted on heat stress during pregnancy and its impact on the fetus. The mechanisms that could explain the effects of fetal exposure to environmental pollutants on subsequent health are poorly documented, but several avenues are being explored. One hypothesis is that environmental pollutants may influence epigenetic programming and the expression of certain genes (via epigenetic alterations), which would then modify the risk of disease occurrence in the short or long term. In addition to being potential biomarkers of exposure and effect, epigenetic marks, and more generally epigenetic epidemiology, represent an opportunity to identify the mechanisms of action of environmental exposures on health.

Abstract Body

Target trial emulation imagines an ideal randomized trial and then tries to emulate it as closely as possible with observational data. We demonstrate that for many perinatal research questions there is no single ideal trial of universal public health interest and discuss how considerations of eligibility criteria, protocol specifications and follow-up time for a hypothetical trial of interest inform the best practices for data collection, design, and analysis in an observational study. For example, the eligibility criteria in an ideal trial of the effect of quitting smoking during pregnancy on childhood outcomes would only enroll women who were smokers prior to pregnancy; thus, emulating this trial with observational data should likewise restrict to women who report smoking pre-pregnancy. A hypothetical trial could also specify instructions regarding smoking behaviour post-pregnancy (i.e. whether to resume smoking), which can modify the effect of smoking during pregnancy on childhood outcomes. To illustrate these distinctions and their importance in practice, we analyze multiple trial analogues of the effect of quitting smoking during pregnancy on childhood health outcomes in the Generation R Study, each answering a different causal question and requiring a different analysis. The trial analogues address the effect of smoking cessation during pregnancy among women who smoked pre-pregnancy, but vary the instructions on women’s smoking behaviors post-pregnancy. Framing perinatal epidemiologic studies as target trials can clarify which study designs and analyses correspond to which public health interventions.

Background and Aims

Active perinatal management is associated with survival without moderate-severe neurodevelopmental disability (NDD) at 2 years of age for extreme preterm (EP: <27 weeks gestational age (GA)) births. Subtle deficits become apparent later. We investigated 5.5 year outcomes according to intensity of perinatal care in EP-born children.

Methods

Births in EPIPAGE-2, a 2011 French national prospective cohort, from 22 to 26 weeks GA with a live fetus at maternal admission to a level 3 hospital were included. Intensity of perinatal care was assigned according to hospital of birth, categorised into low, medium and high intensity groups using the ratio of 24–25 weeks GA babies admitted to neonatal intensive care to fetuses of the same GA alive at maternal admission. Mild and moderate-severe NDD (motor, sensory, cognitive, and behavioural deficits) were assessed at 5.5 years.

Results

126 of 358 (35.2%) children from low, 140 of 380 (36.8%) from medium and 207 of 374 (55.3%) from high intensity hospitals survived. Children born in high compared to low intensity hospitals had less mild (36.1% versus 47.4%, OR=0.44, 95% CI 0.23 – 0.85) or moderate-severe (21.2% versus 26.6%, OR=0.44, 95% CI 0.19 – 1.03) NDD and higher Movement Assessment Battery for Children v2 (mean difference 1.4, 95% CI 0.3 – 2.5) and IQ scores (mean difference 4.4, 95% CI 0.1 – 8.6). There were no differences between medium and low intensity hospitals, or for behaviour.

Conclusions

Higher intensity of perinatal care is associated with less NDD and better motor and cognitive abilities in 5.5 year old EP-born children.