Associations between multiple perinatal exposures and risk of childhood hospitalisation with infection: a registry-based study in two countries.

Abstract

Several perinatal exposures impact severe infection risk in offspring. However, most studies consider exposures in isolation rather than the impact of multiple co-occurring perinatal exposures. Here, we investigated the risk of hospitalised childhood infections associated with combinations of multiple perinatal exposures. We studied liveborn, singleton births in Denmark between 1997 and 2019 (n = 1,113,708) and Norway between 2008 and 2018 (n = 470,270) through national registry linkage. Seven perinatal exposures were evaluated including maternal antibiotic use during pregnancy, maternal smoking, maternal diabetes, hypertensive disorders of pregnancy, caesarean section birth, offspring being small for gestational age (SGA), and preterm birth. The outcome was hospitalised infections before 5 years age. Using Cox regression, we estimated hazard ratios (HR) of hospitalised infections for: (1) each exposure individually; (2) the cumulative number of exposures; (3) each unique combination of the seven exposures; (4) pairwise interactions between exposures. Results were combined through meta-analysis. Each exposure was individually associated with greater hospitalised infection risk with pooled HR estimates ranging from 1.08 (95% CI 1.08-1.09) for SGA to 1.45 (95% CI 1.43-1.46) for preterm birth. Cumulative risk was observed for increasing number of exposures from a pooled HR of 1.17 (95% CI 1.17-1.18) for one exposure to 1.88 (95% CI 1.78-1.99) for five or more exposures. Analyses of unique combinations of the exposures showed a particularly increased risk among preterm children with one or more additional exposures. Our findings can inform targeted strategies for the prevention and management of childhood infections. We highlight a simple, intuitive method for measuring risk associated with co-occurring exposures.