Traditional epidemiological studies come with a variety of biases: recall bias, social desirability bias, and selection bias to name a few. CLIC partners from the United Kingdom, Drs. Audrey Bonaventure, Eve Roman, and Eleanor Kane, were able to eliminate recall bias from their recent study of maternal illnesses using medical records from mothers’ primary-care and obstetrics files. This is a reliable method to get less biased data and therefore stronger results.
With a population of 2,885 cases and 5,499 controls mother-child pairs that were a part of the United Kingdom Childhood Cancer Study (UKCCS), the researchers looked at the relationship between childhood cancers (leukemia, lymphoma, CNS or embryonal tumors) and maternal infections or illnesses during pregnancy (UTI, influenza, genital infection, chicken pox, preeclampsia, gestational hypertension, Polyhydramnios, vomiting or hyperemesis, diabetes, and anemia). With statistical analysis the team was also able to remove more data bias by controlling for confounding variables like sex, maternal age at birth, pregnancy order, UK region, and socioeconomic status measure. The results showed maternal anemia was associated with higher likelihood of childhood acute myeloid leukemia (AML), medulloblastoma, retinoblastoma and embryonal rhabdomyosarcoma. Additionally, there were weak associations between maternal urinary tract infections (UTIs) and childhood non-Hodgkin lymphoma (NHL), preeclampsia and NHL, and polyhydramnios with both AML and NHL.
CLIC scientists around the world continue to use medical records and national health registries to improve our research on the causes of childhood cancer. We believe it is important to study the causes in order to help influence and improve the treatments and potentially prevent children from getting cancer. We strive to have unbiased, diverse studies with our global partners and continue to advance the field with our collaboration.
Citations:
Article Title: Maternal illnesses during pregnancy and the risk of childhood cancer: A medical-record based analysis (UKCCS)
Authors: Bonaventure A, Simpson J, Kane E, Roman E.
Published In: Int J Cancer. 2025 Mar 1;156(5):920-929. doi: 10.1002/ijc.35166. Epub 2024 Nov 13. PMID: 39535336; PMCID: PMC11701404.