Occupational standing, walking and forward bending during pregnancy and the risk of miscarriage: a Danish nationwide, register-based, cohort study
Hannah Nørtoft Frankel, Camilla Sandal Sejbaek, Esben Meulengracht Flachs, Mette Korshøj, Sandra Soegaard Toettenborg, Jens Peter Ellekilde Bonde, Laura Deen, Jonathan Aavang Petersen, Christina Bach Lund, Luise Mølenberg BegtrupObjective
We aimed to investigate the association between occupational standing, walking and forward bending during pregnancy and the risk of miscarriage.
Methods
This register-based study examined all labour-active pregnant women in Denmark between 2004 and 2018, comprising 803 829 pregnancies among 475 312 women. Occupational exposure to standing, walking and forward bending ≥30° was assessed with a quantitative, pregnancy-specific job exposure matrix, based on the woman’s occupational code in the occupational and industry register, at the year of pregnancy start. Information on miscarriage was retrieved from the Danish National Patient Register (DNPR). Statistics Denmark, the Danish Medical Birth Registry and the DNPR provided information on potential confounders. Absence data, sourced from the Danish Register for Evaluation and Marginalisation (DREAM), were assessed as an effect modifier. We estimated adjusted hazard ratios (aHRs) using Cox regression.
Results
All three exposures were associated with increased hazards for miscarriage. The aHR for miscarriage was 1.03 (95% CI: 1.02 to 1.04) per additional hour of occupational standing , 1.18 (95% CI: 1.15 to 1.22) per additional hour of occupational walking and 1.36 (95% CI: 1.30 to 1.41) per additional hour of occupational forward bending . The aHR was more pronounced among women with absence from work the preceding week than among those without. Post hoc quartile analyses indicated a consistent exposure–response relationship for forward bending, while the associations for standing and walking decreased at the highest quartile.
Conclusions
We found an elevated risk of miscarriage associated with all three occupational exposures. Replication in comparable populations incorporating smoking status and other health information is needed to validate these associations.