Original article

Scand J Work Environ Health 2021;47(1):70-77    pdf full text

https://doi.org/10.5271/sjweh.3921 | Published online: 08 Sep 2020, Issue date: 01 Jan 2021

Exposure to psychosocial work strain and changes in smoking behavior during pregnancy – a longitudinal study within the Danish National Birth Cohort

by Mattsson K, Hougaard KS, Sejbaek CS

Objective Knowledge of the relationship between psychosocial strain in the work environment and smoking during pregnancy is scarce. This study aimed to examine the association between psychosocial job strain and change in smoking behavior during pregnancy.

Methods The cohort included 65 645 pregnancies from the Danish National Birth Cohort (1996–2002), where pregnant women were interviewed on job factors and lifestyle during the first and third trimesters. Smoking was categorized into non-, non-daily, and daily smoking at each interview. Psychosocial job strain was categorized into four groups based on the concept of Karasek’s demand–control model: low strain (reference), passive, active and high strain. Associations between psychosocial strain and change in smoking status between the first and second interviews were analyzed by multinomial logistic regression, separately for each smoking category at first interview.

Results Non-smoking women exposed to high strain work were more likely to become daily smokers [adjusted odds ratio (ORadj) 1.41, (95% confidence interval (CI) 1.08–1.83)] compared to non-smoking women exposed to low strain work. Non-smoking women exposed to passive work were more likely to become both non-daily and daily smokers [ORadj 1.59 (95% CI 1.21–2.08) and ORadj 1.32 (95% CI 1.03–1.70), respectively]. Daily smoking women exposed to high strain work were less likely to decrease their smoking [ORadj 0.57 (95% CI 0.32–0.99)] compared to daily smoking women exposed to low strain work.

Conclusions Psychosocial strain influenced the women’s smoking behavior during pregnancy, especially in job types with low control.

This article refers to the following text of the Journal: 2013;39(4):335-342