Background Maternal smoking during pregnancy has an established causal relationship with poor perinatal outcomes including low birthweight and preterm birth. Nonetheless, a significant minority of women, especially those in lower socio-economic groups, continue to smoke throughout their pregnancy despite current interventions to quit. In this group, it has been suggested that harm-reduction may be a more attainable goal. A previous study of low-income pregnant women in a Dublin maternity hospital (The Rotunda) showed that quitting smoking for even a part of pregnancy (partial quitting) resulted in higher birthweight than those who continued to smoke throughout. Further support for this strategy for harm-reduction is required. Our objective was to determine whether the relationship between different smoking patterns among pregnant smokers and perinatal outcomes could be replicated in the UK Southampton Women’s Survey (SWS) cohort.

Methods Women who were smoking at the time of conception (taken as last menstrual period) were categorised according to their smoking status across pregnancy into sustained quitters, partial quitters (who quit smoking in either the first or third trimester alone) or sustained smokers (who continued to smoke throughout pregnancy). Linear regression analyses with birthweight and gestational age as the dependent variables, and smoking status as the exposure were performed. The choice of confounders (child sex, parity, maternal weight and prudent diet score) was guided by a Directed Acyclic Graph (DAG).

Results Of the 3,158 women who became pregnant, 768 were smokers at conception. Of these, 697 (91%) had complete smoking data with 355 (51%) being sustained smokers, 81 (12%) partial quitters and 261 (37%) sustained quitters. Compared with infants born to sustained smokers, infants born to sustained quitters and partial quitters were heavier at birth by 361 g (95% CI: 284 g, 438 g) and 203 g (92 g, 315 g), respectively, adjusted for confounders. Sustained quitters had a longer gestation by 3.5 days (1.7 days, 5.2 days) compared with sustained smokers, adjusted for confounders, but no difference was seen for partial quitters.

Conclusion These results from the SWS, after adjusting for a wide range of available confounders, closely replicated the findings in Dublin, providing further support for partial quitting by pregnant smokers as a harm-reduction strategy for offspring. While sustained quitting is clearly most desired, for women who cannot quit for the duration of their pregnancy, partial quitting should be encouraged as a strategy to reduce some of the harmful effects of smoking on offspring.