Objective The aim was to investigate the association between night work during pregnancy and risk of having a small for gestational age (SGA) child.

Methods This cohort study had payroll data with detailed information on working hours for employees in all Danish administrative regions (primarily hospital employees) between 2007 and 2015, retrieved from the Danish Working Hour Database. Pregnancies, covariates and outcome were identified from the national birth registry. We used logistic regression to investigate the association between intensity and duration of night work during the first 32 pregnancy weeks and SGA. The adjusted model included age, body mass index, socioeconomic status and smoking. Using quantitative bias analysis and G-estimation, we explored potential healthy worker survivor bias (HWSB).

Results The final cohort comprised 24 548 singleton pregnancies in 19 107 women, primarily nurses and medical doctors. None of the dimensions of night work were associated with an increased risk of SGA. We found a tendency towards higher risk of SGA in pregnancies where the women stopped having night shifts during pregnancy. Using G-estimation we found an OR<1 for the association between night work and SGA if all workers continued having night work during pregnancy compared with daywork only.

Conclusion We found no increased risk of SGA in association with night work during pregnancy among healthcare workers. G-estimation was not precise enough to estimate the observed indication of HWSB. We need better data on pregnancy discomforts and complications to be able to safely rule out HWSB.