Smoking among pregnant women : epidemiology and health consequences

Smoking during pregnancy is an important, preventable risk factor for late fetal death and even SIDS. There is a strong dose-response relationship between cigarette smoking and spontaneous abortion, reduction in birth weight, abruptio placentae, placenta previa and bleeding during pregnancy. Ten years ago, the prevalence of smoking among Norwegian pregnant women was between 35 and 40%. During the last 8 years there has been a dramatic change and in 1995 the prevalence seems to be around 20%. THE EPIDEMIOLOGY OF SMOKING IN PREGNANCY In Norway 89% of all women are pregnant at least once up to the age of 35 years (1). During the last ten years we have seen a constantly increasing interest for smoking cessation strategies among pregnant women in Norway. The reasons are the following: • in the middle of the 1980's pregnant women in Norway were among those who had the highest smoking prevalence in the world (Table 1), • only a small proportion of daily smoking women stopped smoking on their own when they got pregnant (5), • a majority of daily-smoking pregnant women want to stop smoking (10), • smoking during pregnancy is probably the single most important threat to a successful pregnancy outcome (2,11), • a doubling of the risk for a health problem late in life may not prevent many women from continuing to smoke, but a doubling of the risk for early miscarriage probably will (12), • women as a group never have more frequent contact with the health care system than during pregnancy. In 1993, the Norwegian Ministry of Health and Social Affairs declared that the prevalence of smoking among pregnant women ought to be 20% or less within the year 1996. This was an ambitious aim, but today it seems to be realistic. In Trondheim (Norway), the smoking prevalence among pregnant women who came to routine ultrasound examination at 18 weeks of pregnancy decreased from 34% in 1987 to 22% in 1994 (9). A recent multicenter study from Tromsø, Trondheim, Bergen, Lillehammer and Oslo indicates that the results from Trondheim can be generalized to the whole country (unpublished data from 1995, publication in preparation). Half of pregnant teenagers in Norway and Sweden are daily smokers compared with 25-29% of mothers aged 25 or more (10,11). Adolescent mothers are at risk from biological, social, educational, and economic factors associated with childbearing at a young age. In addition, their infants also risk the complications associated with maternal smoking (17). Smoking during pregnancy is more common among women whose parents have been smokers, among those whose husbands are smokers, among women who smoked more than 10 cigarettes per day before they became pregnant, and women who started to smoke at an early age (2,10,16). High parity, not living with infant's father, and daily passive smoking at home are also associated with a significantly increased rate for continued smoking during pregnancy, whereas a high level of education and a high age at the onset of smoking decrease the rate (6). The smoking cessation rate during pregnancy varies considerably between different regions throughout the world. Reports from the USA indicate that more than one third of women smoking before pregnancy quit smoking during pregnancy (13). This is comparable with the most recent results from Norway. In the United States, the smoking prevalence rate among pregnant women and adult women in general has decreased 0.3-0.5% per year since 1969 (14). A remarkable change has occurred among black teenagers

• in the middle of the 1980's pregnant women in Norway were among those who had the highest smoking prevalence in the world (Table 1), • only a small proportion of daily smoking women stopped smoking on their own when they got pregnant (5), • a majority of daily-smoking pregnant women want to stop smoking (10), • smoking during pregnancy is probably the single most important threat to a successful pregnancy outcome (2,11), • a doubling of the risk for a health problem late in life may not prevent many women from continuing to smoke, but a doubling of the risk for early miscarriage probably will (12), • women as a group never have more frequent contact with the health care system than during pregnancy.
In 1993, the Norwegian Ministry of Health and Social Affairs declared that the prevalence of smoking among pregnant women ought to be 20% or less within the year 1996.This was an ambitious aim, but today it seems to be realistic.In Trondheim (Norway), the smoking prevalence among pregnant women who came to routine ultrasound examination at 18 weeks of pregnancy decreased from 34% in 1987 to 22% in 1994 (9).A recent multicenter study from Tromsø, Trondheim, Bergen, Lillehammer and Oslo indicates that the results from Trondheim can be generalized to the whole country (unpublished data from 1995, publication in preparation).
Half of pregnant teenagers in Norway and Sweden are daily smokers compared with 25-29% of mothers aged 25 or more (10,11).Adolescent mothers are at risk from biological, social, educational, and economic factors associated with childbearing at a young age.In addition, their infants also risk the complications associated with maternal smoking (17).
Smoking during pregnancy is more common among women whose parents have been smokers, among those whose husbands are smokers, among women who smoked more than 10 cigarettes per day before they became pregnant, and women who started to smoke at an early age (2,10,16).High parity, not living with infant's father, and daily passive smoking at home are also associated with a significantly increased rate for continued smoking during pregnancy, whereas a high level of education and a high age at the onset of smoking decrease the rate (6).
The smoking cessation rate during pregnancy varies considerably between different regions throughout the world.Reports from the USA indicate that more than one third of women smoking before pregnancy quit smoking during pregnancy (13).This is comparable with the most recent results from Norway.In the United States, the smoking prevalence rate among pregnant women and adult women in general has decreased 0.3-0.5% per year since 1969 (14).A remarkable change has occurred among black teenagers in Missouri, whose smoking-during-pregnancy rate decreased from 36% in 1978 to 7% in 1990 ( 15).The relapse rates in women who quit in pregnancy appear to be somewhat lower than those among quitters in the general population.Fingerhut et al reported a one-year postpartum relapse rate of 66% among women who had quit in early pregnancy compared with a one-year relapse rate of 80% in the general population (18).In a German study, smoking cessation rates resulting in long-term abstinence were about three times higher during the year after childbirth and the year before than in other years (19).However, childbirth led to long-term abstinence from smoking in only a small minority of smoking mothers and fathers (19).

MONOXIDE
Nicotine rapidly crosses the placenta to affect the fetus (20).Fetal hypertension due to blood vessel contraction and bradycardia due to stimulation of the carotid and aortic bodies has been demonstrated in rhesus monkeys (21).Placental intervillous blood flow decreases with as much as one fourth during smoking (22) and the fetal heart rate increases after smoking one cigarette (23).Although CO diffuses across the placenta relatively slowly, the fetal CO-Hgb levels reflects that of the mother and is 10% higher than maternal levels (24).
Polycyclic aromatic hydrocarbons (PAHs, known as mutagens and carcinogens), such as benzopyrene, reach the placenta and fetus.The placental concentra-tion is highly correlated with the number of cigarettes smoked (25).

BIRTH WEIGHT
Babies born to women who have smoked during pregnancy are, on average, 170 grams (5%) lighter than babies born to non-smoking women (8,(26)(27)(28)(29).There is a dose-response relationship; the more the woman smokes during pregnancy, the greater the reduction in birth weight (27).Mothers who smoke around the time of conception, nearly double their risk of smallfor-gestational age (SGA) birth (30).In an American study, the risk of having a low-birth-weight baby under 2,500 grams was 49 per thousand for nonsmokers, 76 per thousand for smokers of less than twenty cigarettes per day and 114 per thousand for smokers of twenty cigarettes or more per day (31).Women older than 30 years who smoke, enter pregnancy underweight and have a previous lowbirthweight delivery are at greatest risk of delivering a SGA baby (30,32).Smokers' babies are smaller than non-smokers' in all dimensions: length and head, chest and shoulder circumference (33,34).However, if a woman gives up smoking within her fourth month in pregnancy, her risk of delivering a SGA baby is almost similar to that of non-smokers (35).

ABORTION AND PERINATAL MORTALITY
Some studies have demonstrated a strong doseresponse relationship between maternal cigarette smoking and the risk of spontaneous abortion (36).The increased risk for spontaneous abortion ranges from 1.1 to 1.8 (37,38).The underlying mechanism seems to be due to complications of pregnancy rather than to any fetal abnormality (39).The relative increase in risk associated with maternal smoking is highest (1.7 times that of non-smokers) at the youngest maternal ages (40).
A strong dose-dependent relationship between cigarette smoking and abruptio placentae, placenta previa, bleeding during pregnancy, and risk of perinatal mortality has also been demonstrated (11,41,42).
Smoking seems to be the most important preventable risk factor for late fetal death (11).Smokers aged under 35 face a relative risk of late fetal death ranging from 1.1 to 1.6, while the risk is doubled if the mother is aged 35 years or more and smoke (11).

SUDDEN INFANT DEATH SYNDROME (SIDS)
In a prospective, Swedish study based on 279,938 infants, elevated relative risks of SIDS were associated with low maternal age, multiparity, maternal smoking, and male infants (43).Smoking doubled the risk and a clear dose-response relationship with maternal smoking was observed after controlling for sociodemographic variables.Maternal smoking also seemed to influence the time of death, as infants of smokers died at an earlier age.The authors conclude that smoking may be the single most important preventable risk factor for SIDS in Sweden.

CHILDHOOD CANCER
Benzopyrene from tobacco smoke is known to pass the placenta (25).The relationship of maternal smoking during pregnancy to the incidence of cancers in children has been investigated.A dose-response relationship has been found between the number of cigarettes smoked per day by the mother during pregnancy and cancer risk of the offspring (44).When all tumours were considered, the cancer risk was 50% higher for the most exposed group than for the controls.The risk was doubled for non-Hodgkin lymphoma, acute lymphoblastic leukaemia, and Wilms' tumour.

SMOKING AND LACTATION
There is a negative correlation between smoking and duration of lacta tion with a dose-response effect of the number of cigarettes smoked (45).The same conclusion was made in two other recent studies (46,47).

SMOKING INTERVENTION AMONG PREGNANT WOMEN
Brief counselling and improved access to educational materials allows a low-cost, pregnancy-specific selfhelp smoking cessation programme to be integrated easily into routine prenatal care (48).Such programmes have proved health-care cost saving (49,50).
Pregnant smokers are more likely to stop smoking if they are provided with systematic interventions (7).In a Norwegian intervention study among pregnant women the point prevalence abstinence rate one year after delivery (18 months after the initial intervention) was 15% and 7% in the intervention and the control group, respectively (51).In an American intervention study, which was rather similar to the Norwegian study, 28% in the intervention group and 16% in the control group reported quitting at 32-to 36-week visits.At the postpartum visit, 9% and 10% were nonsmokers, respectively (52).
In another Norwegian study 150 pregnant smokers were assigned to three different smoking intervention programmes (53).Group 1 participated in a smoking cessation programme offering several self-help methods; group 2 attended a doctor's information meeting at the hospital; and group 3 received a pamphlet on the deleterious effects of smoking in pregnancy.Twelve months later, 16%, 6% and 8%, respectively, remained abstinent in the three groups.In another intervention study, the proportion who had ceased smoking early in pregnancy, late in pregnancy and at delivery was 9%, 11.8% and 10.6% in the intervention group versus 2,6%, 4.3% and 4,7% (NS) in the control group (54).
In a follow-up study of pregnant women unwilling to quit smoking, the quitting rate of the negatively motivated women proved similar to that obtained in a stop-smoking intervention protocol used among positively motivated women who also had received additional information (55).
Training family practitioners to provide smoking cessation advice during pregnancy seems to be effective (56).Training resulted in significant chan ges in the advice provided, with greater emphasis on gaining a commitment to smoking behaviour change, but not in the average time providing the advice, approximately three minutes (56).
Nicotine replacement therapy among pregnant women is a controversial subject.Due to official recommendations in Norway, all kinds of nicotine replacement therapy are contraindicated during pregnancy and the lactation period (Felleskatalogen 1994/95).However, the benefits of nicotine replacement therapy to aid pregnant women who have tried to stop smoking, substantially outweigh the risk of continued smoking, particularly those who smoke 20 cigarettes or more per day (57).

CONSEQUENCES FOR PUBLIC HEALTH
A successful smoking intervention programme designed for pregnant women can result in at least six important consequences for public health (Figure 1).Since 89% of Norwegian women are pregnant at least once (1), communicating with daily smoking pregnant women will reach a majority of daily smoking women in young adult age.Smoking cessation research has demonstrated that daily smoking women, like men, are more concerned about the harmful effects of "here and now" rather than harmful effects later in life (12).Focusing the harmful effects on the fetus and the newborn baby might increase the success of the intervention.
Available data indicate that a majority of pregnant women who stop smoking during pregnancy, resume smoking within the first year postpartum (18).Helping women to remain nonsmokers rather than to resume smoking after the birth is important, perhaps even more important than getting them to stop during pregnancy.Therefore, smoking intervention during pregnancy can have a major impact on the future health of women as well as on the health of their offspring.

Figure 1 .
Figure 1.Spreading consequences of smoking cessation among pregnant women.

Table 1 .
Smoking habits in early pregnancy in different studies.
* Collaborative Group on Drug Use in Pregnancy, 22 countries.