From this sibship file, 184,483 linked first- and second-order singleton pregnancies (live births and fetal deaths greater than or equal to 20 weeks; gestation) for Missouri resident women were extracted. This represented 86.8 percent of the attempted linkages in which the targeted first birth was a live birth. There was little difference in match rates by demographic characteristics (race, marital status, birth weight). Maternal smoking status was available for both pregnancies for 176,843 of these sibships (95.9 percent), which represented the final study sample.

Results

More of the Missouri women in this study smoked cigarettes during their second pregnancy (27.4 percent) than during the first (25.8 percent). As Table 1 shows 21.5 percent of women smoked in both first and second pregnancies and 68.4 percent did no t smoke in either pregnancy. While 4.3 percent of the study mothers smoked in their first pregnancy only 5.9 percent smoked in the second pregnancy only.

The pattern of increased smoking in the second pregnancy is strongest for mothers who were under age 20 in their first pregnancy. About 6.2 percent more of these mothers smoked in the second pregnancy than in the first (34.6 percent in first vs. 40.8 p ercent in second). Conversely for women aged 20 or more in the first pregnancy there was little difference in the smoking rates between the first and second pregnancies (22.0 percent vs. 21.8 percent respectively). Black women in their second pregnancy ha d a smoking rate of 29.0 percent or 4.8 percent more than in their first pregnancy. For non-black women the difference was just 1.1 percent (26.0 percent in the first vs. 27.1 percent in the second).

Table 1 shows that the pattern of increased smoking in the second pregnancies is stronger by age than by race. For both black and non-black women there were substantial increases in smoking in the second pregnancy for women under age 20 in the first. F or women aged 20 or more years in the first pregnancy there was little change in the rate of smoking between pregnancies regardless of race. Table 1 also shows much higher smoking rates for non-black women under age 20 in the first pregnancy than black wo men which is the opposite of the pattern or women ages 20 or more.

One exception to the increased smoking in the second pregnancy is for women who experienced a fetal or infant death in the first pregnancy. For these women, more stopped smoking by the second pregnancy (7.3 percent) than those women who did not experie nce a death in the first pregnancy (4.2 percent). The smoking rate for women experiencing a death in the first pregnancy was 30.7 percent in the second pregnancy compared with 32.4 percent in the first. This pattern of reduced smoking in the second pregna ncy did not quite hold for women with a low birth weight infant in the first pregnancy (39.5 percent in first vs. 40.2 percent in second).

Birth weight in the second pregnancy was highly correlated with smoking status in the second pregnancy regardless of smoking status in the first pregnancy. The mean birth weight after adjustment for covariates in the second pregnancy for infants of mot hers who smoked in both pregnancies was 3,280 grams (95 percent confidence interval 3,275-3,285 grams) or 185 grams less than infants of mothers who smoked in neither pregnancy (3,465 + 3 grams). Infants of mothers who smoked in the second pregnanc y only weighed an average of 3,303 + 10 grams, while infants of mothers who smoked in the first pregnancy only weighed the most at 3,489 + 11 grams.

As Table 2 shows, mothers who smoked in the second pregnancy had infants with second order adjusted low birth weight (LBW) and small for gestational age (SGA) rates approximately double those not smoking in the second pregnancy, regardless of smoking s tatus in the first pregnancy. Smoking was more strongly associated with SGA rates than with LBW rates. For those smoking in both pregnancies the relative risk (RR) for SGA was 2.31 and 1.87 for LBW compared with infants of mothers smoking in neither pregn ancy.

The adjusted RRs of smoking in both pregnancies to not smoking in both pregnancies were 1.19 for fetal death 1.37 for neonatal death and 1.90 for postneonatal death. Smoking in the second pregnancy but not the first was associated with a significant ad justed RR of 1.60 for postneonatal death.

Table 3 presents the RRs of levels of smoking in first and second pregnancies to not smoking in either pregnancy for LBW SGA and perinatal mortality. Generally there was a close relationship of smoking level in the second pregnancy to pregnancy outcome s in the second pregnancy. Infants of mothers who smoked at least one pack of cigarettes per day had adjusted RRs for LBW and SGA of more than two to one compared with non-smoking pregnancies regardless of smoking level in the first pregnancy. Pregnancies involving mothers who reduced their smoking level from one or more packs per day in the first pregnancy to less than a pack per day in the second pregnancy had similar RRs for SGA and LBW to those smoking a pack or more per day in the second pregnancy. T he two other groups of mothers smoking less than a pack per day in the second pregnancy had RRs of LBW and SGA between 1.65 and 2.05. Infants of mothers who stopped smoking by the second pregnancy had RRs for LBW and SGA in the second pregnancy between 0. 95 and 1.24.

The RRs of amount smoked to fetal and neonatal mortality were again much lower than for LBW and SGA. The group with the highest adjusted RR of 1.79 for fetal mortality was for those mothers who did not smoke in the first pregnancy, but smoked a pack or more per day in the second pregnancy. The group with the highest adjusted RR for neonatal death was for women who reduced their smoking level in the second pregnancy, but did not stop (1.65). Significant adjusted RRs for postneonatal death were found for five of the six categories involving smoking in the second pregnancy.

Also as with the cross-sectional studies, smoking in the second pregnancy was associated with a decreased birth weight of nearly 200 grams and a doubling of the risk for LBW and SGA. The relationship was stronger for SGA than LBW as smoking apparently affects birth weight more than gestational age. Perinatal mortality was increased by nearly 30 percent among smoking-related pregnancies. This longitudinal study showed that mothers who smoked in the first pregnancy but did not smoke in the second have th e same or lesser risk of a negative pregnancy outcome as those who did not smoke in either pregnancy. Therefore stopping smoking between pregnancies is clearly beneficial.

However, despite the obvious benefits of smoking cessation, most women who smoked in the first pregnancy also smoked in the second, and a large percentage smoked in the second after abstaining in the first. This occurred in spite of the general decline in smoking rates during the 13 years of this study and the fact that there was an average of 3 years between pregnancies in this study. Possible reasons for the increased smoking in the second pregnancy, especially among teenagers, may be underreporting of smoking in the first pregnancy or the possibility that teen pregnancy may be a precursor to substance abuse including cigarette smoking. The type of risky behavior that leads to teen pregnancy may be the same type that leads to smoking.

References:

3Kendrick, J.S., Zahniser, C., Miller, N., Salas, N., Stine, J., Gargiullo, P.M., et. al. Integrating smoking cessation into routine public prenatal care: the smoking cessation in pregnancy project. American Journal of Public Health 1 995; 85: 217-222. 

*95 percent confidence interval does not overlap one.

*95 percent Confidence Interval does not overlap one.

Live births decreased sharply in November as 5,063 Missourians were born compared with 6,816 one year earlier. Irregular reporting is the primary reason for this sharp decrease.

Cumulative births for the 11- and 12-month periods ending with November also show a decrease, but this too may be due to irregular reporting.

Deaths decreased for all three time periods shown below. For January-November, deaths decreased 1 percent from 49,860 to 49,348

The Natural increase for Missouri in November was 981 (5,063 births minus 4,082 deaths). For January-November, the natural increase rate was the same for 1995 and 1996, 3.9 per 1,000 population.

Marriages increased in November, but decreased slightly for the cumulative months ending with November.

Dissolutions of marriage decreased in November, but increased for the cumulative periods ending with November.

The Infant death rate shows virtually no change for the three time periods shown below.

*Rates for live births, deaths, natural increase, marriages and dissolutions are computed on the number per 1000 estimated population. The infant death rate is based on the number of infant deaths per 1000 live births. Rates are adjusted to account for varying lengths of monthly reporting periods.

AN EQUAL OPPORTUNITY/AFFIRMATIVE ACTION EMPLOYER Services provided on a nondiscriminatory basis.

Alternate forms of this publication for persons with disabilities may be obtained by contacting the Missouri Department of Health, Center for Health Information Management & Epidemiology/Bureau of Health Data Analysis, P.O. Box 570, Jefferson City, MO 65102; phone (573) 751-6278. Hearing impaired citizens telephone 1-800-735-2966.