A new study assessed how seasonality is associated with the risk of extremely preterm birth, defined as live births at 22–28 weeks of gestation. The researchers found that the risk of extremely preterm birth was slightly higher in the autumn versus winter. They hope that their research can eventually help to elucidate why some women give birth extremely early in pregnancy and help to prevent this happening.
Although premature babies get better treatment than ever in Denmark, preterm birth is the main cause of death among children younger than 5 years. The severe subtypes of preterm birth are especially associated with moderate-to-severe neurodevelopmental disabilities, and the adverse health effects appear to persist into adulthood. Identifying preventable risk factors for preterm birth, especially extremely preterm birth, is therefore very important for public health. In a new study, researchers from Statens Serum Institut in Copenhagen assessed seasonality as a risk factor in Denmark.
“We found an association between seasonality and extremely preterm birth. Somewhat surprisingly, our study shows that the fewest extremely premature babies are born in the winter and the most in the autumn. In fact, about 25% more babies were born extremely preterm in the autumn compared with the winter,” says Anders Hviid, Professor, Department of Epidemiology Research, Statens Serum Institut, who also emphasises the importance of differentiating between subtypes of preterm birth:
“In parallel with our findings, we observed that this seasonal association is less strong for the less severe preterm birth subtypes – that is, the longer before full term the mother gives birth, the stronger the association with seasonality.”
Began with COVID-19 study
The researchers behind the study suggest that the underlying causes of preterm birth are complex and that only a few preventable risk factors have been identified. However, the researchers were alerted to the potential of seasonality when colleagues from Statens Serum Institut observed a marked decline in the incidence of extremely preterm birth in Denmark during the first COVID-19-related lockdown in early 2020.
“We naturally wondered why this happened. Could COVID-19 be the cause, and could measures that reduce transmission have a protective effect? It could also be related to many people being less stressed because they did not have to commute to work or that people had more time to exercise. Many factors could explain this,” says Anders Hviid.
Data from more than 1 million pregnancies
The researchers analysed data from more than 1 million pregnancies in Denmark from 1997 to 2016 and categorised preterm birth into subtypes, especially focusing on extremely preterm birth of less than 28 weeks of gestation.
“We investigated the severe cases of extremely preterm birth in our historical registry data and then tried to determine whether seasonality had any influence. This stemmed from the idea that some things that were discussed in connection with the lockdown such as limiting transmission, lifestyle patterns and weather could be related to seasonality, since the seasons also affect our lifestyles, and the incidence of various infectious diseases also varies by season,” explains Anders Hviid.
Multifactorial causes
The findings indicate that some seasonal factors may be associated with the risk of extremely preterm birth, and Anders Hviid advocates further research to identify preventable risk factors associated with extremely preterm birth. Examining seasonality more closely may inspire further research on risk factors such as infections, weather, physical activity and nutrition. However, the researchers will not speculate further about such factors:
“The associations between seasonality and extremely preterm birth are currently very hypothetical and multifactorial,” says Anders Hviid.
Differentiating between subtypes
Anders Hviid says that the study may inspire other researchers in the field to more closely examine the subtypes of preterm birth.
“As our study indicates, some risk factors might only apply to one subtype, and we may miss something by treating all the subtypes alike,” concludes Anders Hviid.