Back

Spatiotemporal patterns of African infant hydrocephalus are predicted by prenatal environment and ancestral genomics

Newbury, L.; Mulondo, R.; Sasanami, M.; Movassagh, M.; Nsubuga, B. K.; Natukwatsa, D.; Magombe, J.; Nalule, E.; Wabukoma, M.; Ssentongo, P.; Paulson, J. N.; Mbabazi-Kabachelor, E.; Warf, B. C.; Ssenyonga, P.; Onen, J.; Ericson, J. E.; Kiwanuka, J.; Morton, S. U.; Diggle, P. J.; Whalen, A. J.; Fronterre, C.; Schiff, S. J.

2026-03-11 public and global health
10.64898/2026.03.10.26347974 medRxiv
Show abstract

Infant hydrocephalus is the leading indication for neurosurgery in early childhood in sub-Saharan Africa. The causes and consequences of hydrocephalus generate a high rate of morbidity and mortality. Postinfectious hydrocephalus and neural tube defects together comprise the majority of infant hydro-cephalus in such settings, and both conditions are in principle preventable. In this analysis of data including more than 5000 individuals treated for infant hydrocephalus over 19 years in Uganda, we find strong evidence that prenatal environmental exposures predict the risk of hydrocephalus. The risk of postinfectious hydrocephalus was positively correlated with poverty and local rainfall (with maximal rainfall effect 14 days prior to birth), and negatively correlated with geographical ancestral admixture and elevation. The risk of neural tube defects was positively correlated with geographical ancestral admixture and negatively correlated with vegetation index (with maximal vegetation effect 8 months prior to birth). These findings associating potential prenatal gene-environment interactions with infant hydrocephalus lay the foundation for more targeted and optimized public health policies to reduce the rates of these devastating conditions.

Matching journals

The top 5 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.