What happens to women's self-reported cigarette consumption and urinary cotinine levels in pregnancy?

AIMS: To describe the pattern of self-reported cigarette consumption and nicotine consumption, measured by urinary cotinine concentration, in a cohort of pregnant women who did not stop smoking.
DESIGN: Cohort study.
SETTING: Randomly selected general practices from the West Midlands, UK. PARTICIPANTS: Five hundred and fifty-nine pregnant women in a clinical trial who were enrolled at booking for maternity care (about 12 weeks of gestation), who were followed up in mid and late pregnancy and 10 days post-natal and who did not stop smoking during that period. MEASUREMENTS: Retrospectively collected self-reported cigarette consumption prior to pregnancy and contemporaneously collected self-reported cigarette consumption and urinary cotinine concentrations at booking for maternity care, 20 weeks of gestation, 30 weeks of gestation and 10 days post-natal.
FINDINGS: Women reported smoking a median of 10-19 cigarettes per day prior to pregnancy and a median of 5-9 cigarettes per day at booking for maternity care. At booking, women reported consuming a mean [95% confidence interval (CI)] of 6.3 (5.6-7.0) cigarettes per day. At 20 weeks of pregnancy this had risen to mean (95% CI) 11.5 (10.9-12.2), and remained at 11 cigarettes per day when measured again at 30 weeks of gestation and 10 days post-natal. Mean (95% CI) urinary cotinine levels at booking were 6.0 (5.4-6.6) microg/mL, and did not change much through pregnancy. There were statistically significant associations between urinary cotinine and reported cigarette consumption at all time points except at booking.
CONCLUSIONS: Women smokers report lower cigarette consumption at booking for maternity care than they do prior to pregnancy or from mid pregnancy onwards, but cotinine data imply that their intake of toxins does not change throughout pregnancy. Reports suggesting many women reduce their smoking in pregnancy have probably been over-optimistic.