Gregory Colman

Trends in smoking before, during, and after pregnancy in ten states

BACKGROUND While trends in smoking prevalence during pregnancy are known, little is known about trends in quitting during pregnancy and resuming smoking after pregnancy. OBJECTIVES This study examined the trends in and correlates of quitting during pregnancy and resuming smoking after pregnancy. METHODS We used population-based random surveys of recent mothers in ten U.S. states (total of 115,000 women) conducted between 1993 and 1999. RESULTS Although the prevalence of smoking 3 months before pregnancy was stable at around 26%, quitting during pregnancy rose from 37% to 46% between 1993 and 1999. Adjusted for maternal and state characteristics, the odds of quitting during pregnancy increased 51% between 1993 and 1999 (odds ratio [OR]=1.51; 95% confidence interval [CI]=1.08-2.12). Approximately half of the women who quit smoking during pregnancy resumed smoking within 6 months postpartum. Primiparous, privately insured, college-educated women are more likely to quit and least likely to resume smoking after delivery, compared to multiparous, Medicaid-insured, and high school-educated women. Teenaged women are more likely to quit, but also more likely to resume smoking than older women. CONCLUSIONS The increase in quit rates during pregnancy is encouraging, but the lack of any change in smoking before pregnancy or in postpartum relapse rates suggests that permanent changes in maternal smoking will require additional focus.

Impact of newborn follow-up visit timing on subsequent ED visits and hospital readmissions: an instrumental variable analysis.

OBJECTIVE To determine whether newborn first outpatient visit (FOV) within 3 days of discharge is associated with reduced rates of emergency department (ED) visits and hospital readmissions. METHODS Retrospective cohort analysis was performed of all newborns who were born and had outpatient follow-up within a large academic medical center to determine whether they had ED visits or hospital readmission within 2 weeks after hospital discharge. Multivariable regression using an instrumental variable for timing of FOV was conducted to estimate the relationship between FOV within 3 days of discharge and ED visits and hospital readmissions within 2 weeks of discharge, adjusting for potential confounders. Stratified analyses assessed this relationship in subpopulations with medical or social risk factors. RESULTS Of 3282 newborns, 178 (5%) had 1 or more ED visits or hospital readmissions within 2 weeks of hospital discharge. FOV within 3 days was not significantly associated with ED visits and readmissions in the instrumental variable analysis (IVA) (-0.035, P = .11) or the ordinary least squares analysis (OLS) (0.006, P = .52). The difference in coefficients between these analyses, however, suggests that IVA successfully adjusted for some unmeasured bias. In stratified analyses, only newborns born to African American mothers or discharged by family medicine providers demonstrated a significant relationship between FOV within 3 days and reduced odds of ED visits and readmissions. CONCLUSIONS No significant relationship between outpatient visit timing and ED visits and hospital readmissions was found. Further study is needed to assess the impact of early outpatient visits on other newborn outcomes.