Background: Maternal hormonal contraception use has been associated with childhood leukemia risk. However, studies are few and often based on self-reported information.
Methods: Using registry data from Denmark, Norway, and Sweden, we identified 3,183,316 children (born 1996-2018) and followed them from birth until leukemia diagnosis, censoring (death, emigration, other cancer, 20th birthday) or study closure (December 31st, 2017, 2018 or 2020). We estimated hazard ratios (HRs) and 95 % confidence intervals (CIs) for childhood leukemia (any, lymphoid and non-lymphoid) associated with maternal recent use (≤ 3 months before or during pregnancy) or previous use (before recent use) of hormonal contraception overall and by type, compared to no use.
Results: During 29,455,528 person-years, 1701 children developed leukemia (no use: 518, previous use: 974, recent use: 209). Maternal recent use of hormonal contraception was associated with an increased leukemia risk in children (HR 1.22, 95 % CI 1.04-1.44; incidence rate per 1,000,000 person-years [IR] 65), compared to no use (IR 53). The association was strongest for non-lymphoid leukemia (HR 1.69, 95 % CI 1.20-2.37) and mainly driven by the oral combined products, both for any leukemia (HR 1.29, 95 % CI 1.05-1.59) and non-lymphoid leukemia (HR 1.75, 95 % CI 1.17-2.62). Additionally, non-lymphoid leukemia was associated with recent use of the non-oral progestin-only products (HR 2.10, 95 % CI 1.28-3.44).
Conclusions: Although the absolute risk was low, maternal hormonal contraception use up to or during pregnancy was associated with an increased childhood leukemia risk, particularly non-lymphoid leukemia, and mainly driven by oral combined and non-oral progestin-only products.
Keywords: Childhood cancer; Cohort study; Hormonal contraception; Leukemia; Pharmacoepidemiology; Scandinavia.
Copyright © 2024 Elsevier Ltd. All rights reserved.