This Web site is a component of the SAMHSA Health Information Network

Enter Keywords Choose Website This Site Mental Health at HHS All Mental Health

Online Publications Order Publications National Library of Medicine National Academies Press Publications Homepage

printer friendly page e-mail this page bookmark this page shopping cart  current or new account

This Web site is a component of the SAMHSA Health Information Network.

Clinical Preventive Services in Substance Abuse and Mental Health Update: From Science to Services

V. Pregnant Women

The literature provides strong evidence that substance use disorder (tobacco, alcohol, and use of illicit drugs) services for pregnant women can substantially reduce premature births, neonatal deaths, birth defects, and the need for neonatal intensive care. Alcohol use that would not be considered physically problematic for nonpregnant women is medically contraindicated during pregnancy. Effective interventions to address tobacco and alcohol use in pregnancy yield benefits in excess of program costs within 12 months of program initiation. Preventing use of illicit drugs during pregnancy may generate similar benefits, but studies have not been done to definitively confirm or deny this impression. The health care cost savings achieved within 12 months of program initiation will be due to reduction in use of newborn intensive care unit (NICU) services.

The evidence base for the recommended tobacco-related and alcohol-related universal interventions for pregnant women is very strong and includes well-designed, randomized controlled trials. The evidence base for services related to illicit drugs does not include randomized controlled trials because ethical and practical considerations preclude such studies. (Randomized studies would require purposely denying care for substance abuse to half the women in the study.) Despite this limitation, the data from currently available nonrandomized studies fully justify vigorous efforts to identify and address illicit drug use by pregnant women.

The literature specific to depression during pregnancy was insufficient to justify pregnancy-specific depression screening because it does not seem to be of value in preventing postpartum depression (Hayes, Muller, & Bradley, 2001).

Screening pregnant women for use of tobacco, alcohol, and illicit drugs during pregnancy may be considered in the context of similar interventions for all adolescents and all adults. Special emphasis is given to pregnant women in this section of this monograph because such screening usually can be relied upon to be cost-effective by offsetting reductions in health care costs within 12 months of providing the screening service.

Yet another factor is the well-documented increased responsiveness to such screening and counseling during pregnancy, when women appear more sensitive to such screening. After delivery of the infant, they are likely to relapse into previously established patterns of substance use disorder. This relapse, although undesirable, does not negate the value of their abstinence from substance use disorder during pregnancy.

A. Tobacco

Robust research suggests that tobacco screening and follow-up be classified as essential for all pregnant women in all health care settings. The immediate benefit (direct outcome) is reduction of tobacco use for the duration of pregnancy. The indirect but definitive benefit is reduction in the percentage of women delivering lowbirthweight infants who are at high risk of requiring neonatal intensive care (NICU) services and reduction of infant mortality.

Tobacco-related programming for pregnant women has a very high probability of being cost-effective by reducing the need for NICU and other hospital services. This is true even with very low quit rates because of the extremely high cost of NICU and other hospital services.

Within the doctor-patient interface, tobacco control for pregnant women is perhaps best delivered in the context of tobacco and alcohol screening and related services for pregnant women. The primary intervention takes place at the first prenatal visit, when a full history is taken and substantial counseling is provided.

From the perspective of the health care system, the initial screening and follow-up services are best developed in the context of a well-established array of related services for all life-cycle groups, with links to community-based support services.

Interventions
General information on screening, follow-up, and data gathering are presented in Appendix D of this monograph.

The literature provides evidence that every pregnant woman should be asked whether she smokes or uses any other form of tobacco. If so, she may be counseled to quit—at least for the duration of the pregnancy—for the benefit of the unborn child. This may be reinforced at every outpatient visit.

Intervention issues specific to tobacco and pregnancy are as follows:

• Research studies indicate that more intensive smoking cessation programming for pregnant women has not been shown to be more effective than less intense interventions (unlike studies for nonpregnant adult smokers).
• Adequate data are not available to recommend for or against the use of nicotine-replacement products in pregnant women.

Review of Literature
A more general review of the tobacco and health literature is presented in the discussion of tobacco in the Adults (19 Years and Older) section of this report. The following review is limited to literature specific to pregnant women.

Evidence of Need
According to the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services (USPSTF, 1996)—

… Smoking during pregnancy causes about 5 percent to 6 percent of perinatal deaths, 17 percent to 26 percent of low-birthweight births, and 7 percent to 10 percent of preterm deliveries (DHHS, 1989; Centers for Disease Control and Prevention [CDC], 1990), and it increases the risk of miscarriage and fetal growth retardation. It may also increase the risk for sudden infant death syndrome (SIDS) (Mitchell, Ford, Steward, et al., 1993; Schoendorf & Kiely, 1992)… .

Pregnant women who stop smoking by the 30th week of gestation have infants with higher birthweights than infants born to women who smoke throughout pregnancy (CDC, 1990).

Effectiveness: Evidence Base for Intervention
In two of the earlier randomized clinical trials, tobacco cessation counseling with selfhelp materials increased mean birthweight and decreased the incidence of intrauterine growth retardation (Ershoff, Quinn, Mullen, & Lairson, 1990; Sexto & Hebel, 1984).

Studies indicate that asking pregnant women about tobacco use, combined with physician counseling and supplementary smoking cessation programming can increase tobacco-abstinence rates 5–23 percent, comparing intervention to control groups (Ershoff et al., 1990; Sexto & Hebel, 1984; Hjalmarson, Hahn, & Svanberg, 1991; Windsor, Lowe, Perkins, et al., 1993; Mayer, Hawkins, & Todd, 1990).

Since the mid-1980s, every major healthrelated organization that has addressed this issue has recommended routine clinician counseling of adults, pregnant women, parents, and adolescents to avoid or discontinue smoking and use of smokeless tobacco (USPSTF, 1996; American College of Physicians Health and Public Policy Committee, 1986; American Academy of Family Physicians [AAFP], 1994; American Academy of Pediatrics [AAP], 1994, 1988; American College of Obstetricians and Gynecologists [ACOG], 1993; Manley, Epps, Husten, et al., 1991; American Medical Association [AMA], 1993, 1994a; American Dental Association [ADA], 1992; Canadian Task Force on the Periodic Health Examination, 1994b; National Institutes of Health [NIH], 1989, 1994; American Academy of Otolaryngology—Head and Neck Surgery, 1992; Green, ed., 1994).

Strong evidence for the efficacy and costefficiency of tobacco-related interventions for pregnant women can be found in multiple randomized controlled trials and metaanalyses. Four are briefly reviewed below.

The first set of randomized controlled trials was published by Ershoff et al., from Kaiser Permanente, in Los Angeles (Ershoff et al., 1990; Ershoff, Mullen, & Quinn, 1989). These studies explored the benefits of various intensities of smoking cessation programming for pregnant women in an HMO, representing a wide range of socioeconomic classes and racial and ethnic diversity. Women who were welfare clientele or who did not speak English were not included in these studies.

The first trial included 126 cases and 116 controls. The experimental intervention consisted of one-time counseling and a set of eight short self-help booklets distributed by mail at weekly intervals, with the women committed to completion of activity assignments within the booklets. The control group received the initial counseling, a twopage brochure, and usual physician counseling. No attempt was made to modify the physician counseling or to provide other health education to the intervention group. This intervention resulted in a 22.2 percent quit rate in the study group, compared with an 8.6 percent quit rate in controls. Compared with the control group, the selfhelp groups were 45 percent less likely to deliver a low-birthweight infant. Within the studied population, mean cost per full-term birth, without intrauterine growth retardation, was $695. Mean cost per preterm birth was $6,213. Benefit-cost ratio, based on data limited to the infants’ initial hospitalization, was estimated at about 3:1.

In 1995, Ershoff et al. published data from 171 pregnant women who quit smoking prior to pregnancy, then relapsed during pregnancy (Ershoff, Quinn, & Mullen, 1995). These women were provided the same interventions noted above (simultaneous with the study noted above). In the intervention group, 16 percent relapsed, compared with 20 percent in the control group—a difference too small to be of statistical significance.

In 1996 and 1997, the Ershoff team ran another smoking cessation trial among pregnant women. This study, published in 1999 (Ershoff et al., 1999), randomized 390 English-speaking women, 18 years of age and older, into three groups. The first received usual physician counseling and a self-help book. The second also was given telephone access to a computerized telephone cessation program based on interactive voice response technology. The third received the booklet, usual counseling, plus proactive telephone counseling from nurse educators using motivational interviewing techniques and strategies. All three groups achieved the approximate 20 percent quit rate achieved in the earlier study, but the more intensive interventions provided no additional benefit. In all three groups, cessation rates among initially heavy smokers were strikingly low. Within each of the groups, approximately two thirds of the women made at least one serious attempt to quit smoking, at least for the duration of pregnancy. Most were unable to do so. Mean reductions in cigarette smoking among those who continued to smoke were modest, averaging a reduction from 8.3 cigarettes per day to 7.8 cigarettes per day.

Windsor et al. reported on a preliminary and more definitive trial conducted in a public health clinic population in Birmingham, Alabama (Windsor, Warner, & Cutter, 1988; Windsor et al., 1993). The initial study randomized 309 pregnant smokers into three groups. Group 1, the control, received information in a nonfocused interaction on smoking and pregnancy requiring approximately 5 minutes at the first prenatal visit. Group 2 received the standard clinic information plus a copy of Freedom From Smoking in 20 Days, a self-help manual published by the American Lung Association (ALA). They also received an ALA informational packet entitled “Because You Love Your Baby” and a 10-minute educational session by a baccalaureate-trained health education specialist at the initial prenatal visit. The third group received the Group 2 intervention, but with a pregnancy-specific self-help manual, A Pregnant Woman’s Self- Help Guide To Quit Smoking. No smoking cessation interventions were used in any of the three groups after the first prenatal visit. Smoking status was confirmed midpregnancy and at the end of pregnancy using patient self-reports and saliva thiocyanate tests. The quit rates were 2 percent, 6 percent, and 14 percent for the three groups, respectively.

In the follow-up study, published in 1993 (Windsor et al., 1993), the Windsor team randomized 814 pregnant smokers from the same clinic setting to case and control groups. The control group received an intervention similar to that of Group 2 from the earlier study. The experimental group received more extensive written materials and counseling, with follow-up and reinforcement at each subsequent clinic visit. Quit rates in the two groups were approximately the same as the quit rates in the earlier study—8.5 percent and 14.3 percent in the two groups, respectively. Quit rates were higher for African Americans than for Whites in both control and experimental groups (10.7 percent and 18.7 percent for African Americans, compared with 5.2 percent and 10.0 percent for Whites).

In a study similar to the second Windsor study but conducted in a Women, Infants, and Children (WIC) clinic in Grand Rapids, Michigan, Mayer et al. (1990) demonstrated quit rates of 11 percent among the experimental group and 3 percent among the controls. When measured 4.7 weeks postpartum, the quit rates within the two groups were 7 percent and 0 percent, respectively.

The strength of this evidence base and benefits of such screening were reaffirmed in a 2002 meta-analysis by Melvin et al. (Melvin, Dolan-Mullen, Windsor, Whiteside, & Goldenberg, 2000). Another extensive literature review published that same year (Lumley, Olver, & Waters, 2000) noted that smoking cessation programs in pregnancy appeared to reduce smoking, low-birthweight and preterm birth, but no effect was detected for very low birthweight or perinatal mortality. Five trials of (postpartum) smoking relapse prevention showed no significant benefit (Lumley et al., 2000).

Efficacy and Program Implementation Issues
A meta-analysis by Mullen (1999) provides a summary of the available literature and implementation-related issues to be considered by individual managed care plans. Important program implementation points include the following:

• Smoking during pregnancy is a substantial health hazard to the fetus/infant and mother.
• These hazards appear to be best avoided by having the woman quit smoking prior to pregnancy; but if that has not been achieved, substantial benefits may be secured by having her quit, or at least substantially reduce cigarette consumption during pregnancy.
• Available interventions only offer limited quit rates (5–23 percent).
• Prevalence of smoking is higher and response to smoking-cessation programming is less substantial in lowincome and otherwise economically and socially vulnerable women.
• Estimating both current smoking rates and quit rates in a given population can be problematic because smokers who know they should not smoke often lie. The better studies (such as all those referenced above) supplement the women’s statements with laboratory measures of tobacco exposure. Laboratory confirmable quit rates tend to run much lower than the rates suggested by interviews of smokers. (Editorial note: such laboratory confirmation, measuring cotinine or thiocyanate used in research studies, is not suggested for routine clinical practice.)
• Studies show that pregnant women seem to respond differently to smokingcessation programming, compared with other adults who smoke. In other adults, more intensive programming with more frequent personal contact increases quit rates, as does use of nicotine replacement products. With pregnant women, basic physician counseling, supplemented by limited interventions, such as self-help materials, appears to generate maximal benefit, while more intensive programming does not increase quit rates.
• High-quality data on the efficacy of nicotine replacement products are not available for pregnant women.

The one issue of greatest concern not addressed by Mullen is the level of benefit, according to quit rate, that is needed to generate cost-effectiveness within 12 months of program initiation. This issue is addressed in a cost-benefit/cost-effectiveness analysis of such programming published by Marks and his team at the Centers for Disease Control and Prevention (CDC) in 1990 (Marks, Koplan, Hogue, & Dalmat, 1990). This analysis, based on the studies referenced previously in this report and a number of similar studies by other authors, demonstrates an average savings of $3.31 for each dollar spent on effective smoking cessation programming. This estimate assumes a quit rate of approximately 15 percent, with the cost calculations limited to prenatal care and the initial hospitalization at time of birth of the infant. Considering the cost of care for the infant in subsequent years, the benefit exceeds $6 per dollar spent on smoking cessation programming for pregnant women. According to these limited calculations, a program with a quit rate of only 5 percent could pay for itself within a year. These cost-benefit calculations do not include costs averted relative to respiratory illness in mother and infant or any of the other smoking-related costs, some of which can be substantial.

One other study of note is that of Latts et al. (Latts, Prochaska, Salas, & Young, 2002) in a Denver, Colorado, managed care plan. In this study, the sponsoring plan staff from participating physician offices were trained and paid $150 for each pregnant woman counseled. This study, reported as an uncontrolled pilot study, failed to increase the number of smokers counseled.

Program implementation issues deal with the social and cultural milieu of the pregnant woman, her educational and socioeconomic status, and the dedication of both the physician and health care system to tobacco control. The Ershoff (Ershoff et al., 1999), Windsor (Windsor et al., 1988, 1993), and Mullen (1999) studies referenced above provide information on providing effective and cost-efficient smoking-cessation services to pregnant women in conventional HMO settings (Ershoff et al., 1999) and indigent care clinics (Windsor et al., 1988, 1993). The Mullen study (Mullen, 1999) provides excellent guidance on issues to be addressed in the design of such programs.

In the studies where this has been documented, more than half the women who quit smoking during pregnancy resume smoking after the birth of the infant (CDC, 2002). Thus, screening of pregnant women for tobacco use and provision of antismoking programming does not eliminate the need for the pediatrician to address these same issues after birth of the infant, for the benefit of both mother and child.

Data Needs Specific to Tobacco and Pregnancy
Refer to Appendix D, Procedures for Implementation and Evaluation of Preventive Services, for a discussion of issues related to screening, follow-up, and data gathering.

Assessment of Need for Programming and Assessment of Program Efficacy
Collecting the following data would help health plans track and evaluate the impact of tobacco interventions:

• Medical records data showing use or suspicion of use of tobacco before and during pregnancy
• The number and percentage of these women who quit prior to the first prenatal visit
• Rates of NICU utilization and other hospital services during the first 30 days of life
• Perinatal death rates (infant death rates during the first 30 days of life)
• Comparison of fetal/infant illness, death, and health care utilization through the first 30 days of life, comparing mothers who quit, those who did not, and nonusers (as ascertained by interview and recorded in the medical record)

Summary of Tobacco Use and Pregnancy
Tobacco use during pregnancy is a major cause of prematurity, low birthweight, and neonatal death. The robust literature indicates that all pregnant women—and those comtemplating becoming pregnant—should be screened for use of tobacco and advised to quit. In response to such screening and follow-up, quit rates from 5 to 30 percent can be expected. Even a 5 percent quit rate is likely to pay for itself in reduced utilization of intensive care for premature infants within 12 months of program initiation.

B. Alcohol

Screening pregnant women for alcohol use is classified as “general.” This means that extensive research suggests programming is beneficial to all pregnant women in all health care settings. The direct outcome is reduced alcohol use during pregnancy. The immediate benefit is a dramatic reduction in Fetal Alcohol Spectrum Disorders (FASD), including the most debilitating form, Fetal Alcohol Syndrome (FAS), and a modest reduction in prematurity. Given the relative rarity of FAS and FASD in most health care settings, and the nature and quality of the literature available, the primary measurable benefit to reducing alcohol use in pregnancy relates to the reduction in prematurity and low birthweight. The absence of claims for FAS and FAE does not suggest a lack of need for alcohol control programming for pregnant women.

Alcohol-related programming for pregnant women has a very high probability of being cost-effective by reducing the need for NICU services. This is true even with very low abstinence rates because of the extremely high cost of premature births and underweight newborns.

At the doctor-patient interface, alcoholcontrol programming for pregnant women is probably best delivered in the context of tobacco and illicit drug screening and related services for pregnant women. The primary intervention takes place at the first prenatal visit, when a full history is taken and substantial counseling is provided.

From the perspective of the health care system, the initial screening and the followup services may be best developed in the context of a well-established array of such services for all life-cycle groups, with links to community-based support services.

Interventions
A general discussion of factors related to screening, follow-up, and data gathering appears in Appendix D, Procedures for Implementation and Evaluation of Preventive Services. The literature provides strong evidence that every pregnant woman should be asked about alcohol consumption and should be urged to abstain, at least for the duration of the pregnancy for the benefit of the unborn child. Similarly, research suggests that those who historically have consumed alcohol would benefit from having this message reinforced at every outpatient visit.

Intervention-Related Issues Specific to Alcohol and Pregnancy
Information adapted from the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services(USPSTF, 1996) suggests that—

• All pregnant women be screened for evidence of problem drinking or risk drinking (two drinks or more per day or binge drinking), especially during the first trimester of pregnancy.
• All pregnant women and all women contemplating pregnancy be informed of the harmful effects of alcohol on the fetus and be advised to cease drinking.
• Women who both smoke and drink be advised that their risk of low-birthweight infants is greatest.
• Patients with evidence of alcohol abuse or hazardous drinking be offered brief advice and counseling.
• Patients with evidence of alcohol dependence be referred to appropriate clinical specialists or community programs.
• Physician education: Because of the difficulty in ascertaining alcohol use in many women, use of facilitators, as suggested later in this report, or use of videotape-augmented training of obstetric care practitioners may be considered. A group in New Mexico has demonstrated the value of the videotapeaugmented training in a randomized controlled trial (Handmaker, Hester, & Delaney, 1999).

In a 2002 review of alcohol problemrelated screening questionnaires, the National Institute on Alcohol Abuse and Alcoholism (NIAAA, 2002) stated—

… Two questionnaires are available that are appropriate for pregnant women, both derived in part from CAGE (Cut Down/Annoyed/Guilty/Eye opener) (Chan et al., 1994), T-ACE Tolerance-Annoyed/Cut down/Eye Opener (Sokol, Martier, & Ager, 1989) takes approximately 1 minute to complete and is more accurate than AUDIT (Alcohol Use Disorder Identification Test) for detecting current alcohol consumption and risky drinking, as well as history of past alcoholism; however, it is less specific (Chang, 2001). The five-item TWEAK (Tolerance/Worried/Eye opener/Amnesia/K(c)ut down) (Russell, Martier, & Sokol, 1991) performs similarly to T-ACE (Chang, 2001) and can be used to detect a range of drinking levels from moderate to high-risk consumption (Dawson, Das, Faden, et al., 2001).

Details on these and other alcohol-related screening tests can be found on the NIAAA Web site at www.niaaa.nih.gov. Additional information and sample questionnaires for CAGE and AUDIT are provided in the discussion about alcohol in this monograph.

Literature Review
More substantial reviews of the alcohol-andhealth literature can be found in the sections of this monograph related to selected children, adolescents, and adults. The discussion on adults and alcohol includes presentation and discussion of the most important alcohol screening questionnaires.

Evidence of Need
According to further information in the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services (USPSTF 1996)—

The proportion of pregnant women who report drinking has declined steadily in the U.S. (Serdula, Williamson, Kendrick, et al., 1991). Recent surveys indicated 12–14 percent of pregnant women continue to consume some alcohol (Goodwin, Bruce, Zahniser, et al., 1994; CDC, 1994b), with most reporting only occasional, light drinking (median: four drinks per month) (Serdula et al., 1991). Binge drinking or daily risk drinking (usually defined as two drinks per day or greater) is reported by 1–2 percent of pregnant women (Goodwin et al., 1994; CDC, 1994b, 1995a), but higher rates (4–6 percent) have been reported in some screening studies (Sokol et al., 1989; Russell, Martier, Sokol et al., 1994).

Excessive use of alcohol during pregnancy can produce fetal alcohol syndrome (FAS), a constellation of growth retardation, facial deformities, and central nervous system dysfunction (microcephaly, mental retardation, or behavioral abnormalities) (Rosett, Weiner, & Edelin, 1983). Other infants display growth retardation or neurologic involvement in the absence of full FAS (i.e., fetal alcohol effects [FAE]) (NIAAA, 1993). FAS has been estimated to affect approximately one in 3,000 births in the U.S. (1,200 children annually), making it a leading treatable cause of birth defects and mental retardation (Abel & Sokol, 1991; CDC, 1993b).

The level of alcohol consumption that poses a risk during pregnancy remains controversial (NIAAA, 1993; Russell, 1991). FAS has only been reported in infants born to alcoholic mothers, but the variable incidence of FAS among alcoholic women (from 3 to 40 percent) (Abel & Sokol, 1991) suggests that other factors … may influence the expression of FAS (NIAAA, 1993)…. Most studies report an increased incidence of FAE among mothers who consume 14 drinks per week or more (Russell, 1991; Virji, 1991; Forrest, Florey, et al., 1991; Verkerk, Noord-Zaadstra, Florey, et al., 1993), but the effects at lower levels have been inconsistent (Russell, 1991; Jacobson, Jacobson, Sokol, et al., 1993; Streissguth, Barr, & Sampson, 1990). Modest developmental effects have been attributed to light drinking (seven drinks per week) in some studies, but underreporting by heavy drinkers and confounding effects of other important factors (nutrition, environment, etc.) make it difficult to prove or disprove a direct effect of light drinking (NIAAA, 1993; Russell, 1991; Knupfer, 1991). Timing of exposure and pattern of drinking may be important, with greater effects proposed for exposure early in pregnancy and for frequent binge drinking (NIAAA, 1993).

Effectiveness Evidence Base for Intervention
According to the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services (USPSTF, 1996)—

There are no definitive controlled trials of treatments for excessive drinking in pregnancy (Schorling, 1993). In several uncontrolled studies, a majority of heavy-drinking pregnant women who received counseling reduced alcohol consumption (Rosett et al., 1983; Larson, 1983; Halmesmaki, 1988) and reductions in drinking were associated with lower rates of FAS (Rosett et al., 1983; Halmesmaki, 1988). Many women spontaneously reduce their drinking while pregnant, however, and women who continue to drink differ in many respects from women who cut down (e.g., heavier drinking, poorer prenatal care, and nutrition). As a result, it is difficult to determine precisely the benefit of screening and counseling during pregnancy. In two trials that employed a control group, the proportions of women abstaining or reducing consumption were similar in intervention and control groups (Waterson & Murray-Lyon, 1990; Meberg, Halvorsen, Holter, et al., 1986).

The U.S. Surgeon General (Surgeon General, 1981) and the American Academy of Pediatrics (AAP) and American College of Obstetricians and Gynecologists (ACOG) (AAP/ACOG, 1992; American Academy of Pediatrics Committee on Substance Abuse and Committee on Children with Disabilities, 1993) advise counseling all women who are pregnant or planning pregnancy that drinking can be harmful to the fetus and that abstinence is the safest policy. The Canadian Task Force (CTF) recommends that all women be screened for problem drinking and advised to reduce tobacco use during pregnancy (CTF on the Periodic Health Examination, 1994a).

Efficacy and Program Implementation Issues
In the case of alcohol control during pregnancy, the major program implementation issue will relate to the sociodemographic profile of the membership and issues that will need to be addressed relative to cultural sensitivity. The overall community tolerance for alcohol consumption, use, and abuse will be a significant factor.

A major part of the problem is identifying alcohol use in pregnant women, since many will not admit such use. Several studies have demonstrated the value of structured questionnaires as an effective means of ascertaining alcohol use (Chang et al., 1998; Chang, Goetz, Wilkins-Haug, & Berman, 1999; Midanik, Zahnd, & Klein, 1998; Bull, Kvigne, Leonardson, Lacina, & Welty, 1999; Chasnoff, Neuman, Thornton, & Callaghan, 2001).

Another part of the problem is the limited utility of interventions, especially in heavier drinkers and those who do not access early prenatal care. As noted below, results are mixed and not well documented in controlled studies. The better controlled studies did not address the cost-benefit or cost-efficiency of treatment options.

Despite this lack of firm evidence, the hazard posed by alcohol consumption during pregnancy and the apparent ease by which alcohol consumption can be reduced in many pregnant women would seem to indicate that all health care providers should address this issue.

Although they did not provide new findings or evidence, two recent reviews nicely summarized literature more recent than the USPSTF Guide (USPSTF, 1996). These are a 1999 review in the Milbank Quarterly (Frohna, Lantz, & Pollack, 1999) and a 2000 review from a group at Wayne State University in Detroit (Hankin, McCaul, & Heussner, 2000).

Data Needs Specific to Pregnancy and Alcohol
Collecting the following data would help health plans track the impact of their alcohol- screening intervention. Refer to Appendix D.

• Numbers of cases of FAS and FASD diagnosed in prior year
• Evidence of alcohol-related problems in other members of the managed care plan that might suggest a community-wide alcohol problem
• Medical records data showing use or suspicion of use of alcohol before and during pregnancy
• The number and percentage of these women who quit prior to the first prenatal visit n Rates of NICU utilization and other hospital services during the first 30 days of life
• Perinatal death rates (infant death rates during the first 30 days of life)
• Comparison of fetal/infant illness, death, and health care utilization through the first 30 days of life, comparing mothers who quit, those who did not, and nonusers (as ascertained by interview and documented in the medical record)

Summary of Alcohol Use and Pregnancy
The robust literature indicates that all pregnant women—and those contemplating becoming pregnant—should be screened for the use of alcohol and advised to abstain while pregnant.

C. Illicit Drugs

Screening pregnant women for use of illicit drugs is classified as “general.” This means that strong research supports this for all pregnant women in all managed care and other health care settings. With the exception of withdrawal symptoms at time of delivery, no studies have successfully separated the effects of the illicit drugs on the fetus/infant from the effects of concurrent tobacco and alcohol use and lack of prenatal care. The literature clearly indicates that pregnant women using illicit drugs have poor pregnancy outcomes, but separating the influence of the drug itself from these other risk factors has proven practically impossible (USPSTF, 1996). There are no published studies in which the woman has been given drug treatment without concurrent prenatal care.

The benefits to be pursued are reduction of illicit drug use during pregnancy and elimination of maternal, fetal, and infant complications of such use. At the doctorpatient interface, programming for pregnant women using illicit drugs is probably best delivered in the context of tobacco and alcohol screening and related services for pregnant women. The primary intervention takes place at the first prenatal visit, when a full history is taken and substantial counseling is provided. From the perspective of the health care system, the services are best developed within the context of established services for all life-cycle groups with links to community-based support services.

Intervention
Robust research supports asking every pregnant woman about use of illicit drugs and urging pregnant women to abstain, at least for the duration of the pregnancy, for the benefit of the unborn child. Similarly, the literature provides strong evidence that this message should be reinforced at every outpatient visit for those who historically have used such drugs.

Service-Related Issues Specific to Illicit Drugs and Pregnancy
Information adapted from the recommendation in the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services (USPSTF, 1996) suggests that—

• Every managed care organization has access to psychiatrists and/or other professional staff who are expert in the diagnosis and management of women who engage in the use of illicit drugs (marijuana, cocaine, heroin, and others) during pregnancy.
• All clinicians in managed care settings that participate in the provision of prenatal care be trained to recognize signs and symptoms that suggest use of illicit drugs during pregnancy and how best to interview such patients.
• All pregnant women be advised of the potentially adverse effects of drug use on the development of the fetus.
• Routine (blood and urine) screening of pregnant women for illicit drug use is only justified when dealing with populations known to have a high prevalence of use of such drugs (more than 2 percent of pregnant women as ascertained by record review and/or claims data). There is no need for such a screening program in most managed care organizations.
• Organizations dealing with a high prevalence of use of illicit drugs or an otherwise exceptionally high-risk population for such substance abuse are virtually assured of encountering high rates of tobacco use and alcohol abuse. Such organizations can consider their options for screening through modification of one of the alcoholrelated screening instruments, and adoption of follow-up of such screenings patterned after their alcohol-control programming.

Review of Literature
Evidence of Need
According to the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services (USPSTF, 1996)—

A national probability sample of 2,613 women giving birth in 1992–93 estimated that 5.5 percent used some illicit drug during pregnancy: the most frequently used drugs were marijuana (2.9 percent) and cocaine (1.1 percent) (National Institute on Drug Abuse [NIDA], 1994c). Anonymous urine testing of nearly 30,000 women giving birth in California in 1992 detected illicit drugs in 5.2 percent: marijuana (1.9 percent), opiates (1.5 percent), and cocaine (1.1 percent) were the most frequently detected substances (Vega, Kolodny, Hwang, & Noble, 1993). Prevalence of drug use generally is higher among mothers who smoke or drink, are unmarried, are not working, have public or no insurance, live in urban areas, or receive late or no prenatal care (NIDA, 1994c; Vega et al., 1993; Moser, Jones, & Kuthy, 1993). Anonymous urine testing detected cocaine use in 7–15 percent of pregnant women from high-risk, urban communities (Schulman, Morel, Karmen, et al., 1993) and in 0.2 percent to 1.5 percent of mothers in private clinics and rural areas (Sloan, Gay, & Snyder, 1992; Burke & Roth, 1993).

Drug use during pregnancy has been associated with a variety of adverse outcomes, but problems associated with drug use (e.g., use of alcohol or cigarettes, poverty, poor nutrition, inadequate prenatal care) may be more important than the direct effects of drugs (Mayes, Granger, Borstein, et al., 1992; Robins & Mills, eds., 1993). Regular use of cocaine and opiates is associated with poor weight gain among pregnant women, impaired fetal growth, and increased risk of premature birth; cocaine appears to increase the risk of abruptio placentae (Volpe, 1992). The effects of social use of cocaine in the first trimester are uncertain (Graham, Dimitrakoudis, Pellegrini, et al., 1989; Chasnoff, Griffith, MacGregor, et al., 1989). Cocaine has been blamed for some congenital defects (Robins et al., 1993), but the teratogenic potential of cocaine has not been definitively established. Infants exposed to drugs in utero may exhibit withdrawal symptoms due to opiates, or increased tremors, hyperexcitability, and hypertonicity due to cocaine (Robins et al., 1993; Hutchings, 1982). Possible long-term neurologic effects of drug exposure are difficult to separate from the effects of other factors that influence development among vulnerable children (Robins et al., 1993; Frank, Bresnahan, & Zuckerman, 1993; Chasnoff, Griffith, Freier, & Murray, 1992). The effects of marijuana on the fetus remain controversial (Zuckerman, Frank, Hingson, et al., 1989; Day & Richardson, 1991; Bell & Lau, 1995).

Effectiveness: Evidence Base for Intervention
Although the risk of drug use to the mother and fetus is clear, the evidence base for effective interventions during pregnancy is largely limited to observational studies showing a decrease in the risk of low birthweight with increasing numbers of prenatal visits (Chasnoff et al., 1989; Zuckerman et al., 1989).

Two studies published since the 1996 Guide reaffirmed that substance abuse in pregnancy continues to be a significant problem (Butz, Lears, O’Neil, & Lukk, 1998; Richardson, Hamel, Goldschmidt, & Day, 1999). Our literature search also identified five clinical trials relating to treatment to secure discontinuation of illicit drug use in pregnancy (Elk, Mangus, Rhoades, Andres, & Grabowski, 1998; Eisen, Keyser-Smith, Dampeer, Sambrano, 2000; Schuler, Nair, Black, & Kettinger, 2000; Jansson et al., 1996; Svikis et al., 1997). All were controlled to some degree, with study populations ranging from 12 (Elk et al., 1998) to 658 (Eisen et al., 2000). Taken together, these studies reaffirm previously established impressions that aggressive provision of basic prenatal care is of substantial value for these women, but supplementary programs for illicit drug use in pregnant women are of only marginal value. In the only one of these studies to address this issue (Eisen, et al., 2000), it was noted that none of the reductions in use of alcohol or illicit drugs was maintained through 6 months postpartum.

Given this circumstance, the recommendation of the American College of Obstetricians and Gynecologists is limited to “a thorough history of substance use and abuse in all obstetric patients, and remain alert to signs of substance use disorder in all women” (USPSTF, 1996; ACOG, 1994).

Efficacy: Program Implementation Issues
According to the 1996 Second Edition of the U.S. Preventive Services Task Force’s Guide to Clinical Preventive Services (USPSTF, 1996)—

The diagnostic standard for drug abuse and dependence is the careful diagnostic interview (USPSTF, 1996; APA, 1994). … There are few data to determine whether or not the use of standardized screening questionnaires can increase the detection of potential drug problems among patients. Brief alcohol screening instruments such as the CAGE or MAST [Michigan Alcoholism Screening Test] can be modified to assess the consequences of drug use in a standardized manner (Trachtenberg & Fleming, 1994; Skinner, 1982), but these instruments have not been compared with routine history of clinician assessment. Questionnaires … [that] identify adolescents at increased risk for drug use … have not been validated in prospective studies (Schwartz & Wirtz, 1990). Other instruments such as the Addiction Severity Index (McLellan, Luborsky, Woody, et al., 1980) are useful for evaluating treatment needs but are too long for screening.

… Drug testing is frequently performed without informed consent in the clinical setting on the grounds that it is a diagnostic test intended to improve the care of the patient. Because of the significance of a positive drug screen for the patient, however, the rights of patients to autonomy and privacy have important implications for screening of asymptomatic persons (Merrick, 1993). If confidentiality is not ensured, test results may affect a patient’s employment, insurance coverage, or personal relationships (Rosenstock, 1987). Testing during pregnancy is especially problematic because clinicians may be required by State laws to report evidence of potential harmful drug or alcohol use in pregnant patients.

There is a single recent paper suggesting that primary care clinicians can ask three questions in the context of a prenatal health evaluation to target women for referral to a full clinical assessment for drug and alcohol use (Chasnoff et al., 2001). The three questions are—

1. Have you ever drunk alcohol?
2. How much alcohol did you drink in the month before pregnancy?
3. How many cigarettes did you smoke in the month before pregnancy?

The screen is intended for use by primary practitioners to sort women by risk category.

In at least one high-prevalence population where this issue was addressed in a recent study in Pittsburgh, women commonly denied their use of tobacco, alcohol, and cocaine. Interviews detected only about half of the women whose urine tests were positive for one or more of these substances (Markovic et al., 2000).

There are few controlled trials of interventions for pregnant women who use illicit drugs (USPSTF, 1996). The lack of randomized and controlled studies is not accidental. It is due to the perception by investigators that it would be unethical to deny pregnant women treatment believed to be beneficial (Burkett, Gomez-Martin, Yasin, & Martinez, 1998). As a result, there is a continuing flow of observational studies (Kukko & Halmesmaki, 1999; Newschaffer, Cocroft, Hauck, Fanning, & Turner, 1998; Berkowitz, Brindis, & Peterson, 1998; Clark, Dee, Bale, & Martin, 2001; Corse & Smith, 1998) and one controlled but not randomized study (Burkett et al., 1998) that showed substantial benefit to mother and fetus/infant. These studies suggest, but do not confirm, that detection of substance use disorder in pregnant women should be costeffective within 12 months of program initiation through reduction in need for NICU services.

The AMA and most other medical organizations endorse urine testing when there is reasonable suspicion of substance use disorder, but none of these groups recommends routine drug screening in the absence of clinical indications (USPSTF, 1996).

Program Implementation Issues:
How To Manage the Intervention So That It Succeeds in Securing Desired Benefits
In most health care settings, issues relative to substance use disorders among pregnant members will be limited to assurance that clinicians engaged in prenatal care have the capacity to recognize such cases and have the capacity to refer such members to appropriate specialists. In those few plans with a prevalence of use of illicit drugs likely to be more than 2 percent of pregnant women, substance use disorder screening and follow-up can be managed in a manner patterned after what should already be welldeveloped alcohol control programming in those managed care plans.

Data Needs Specific to Illicit Drugs and Pregnancy
The following data should help health plans track and assess the impact of their intervention. Refer to Appendix D.

• Numbers of cases of illicit drug use diagnosed in prior year in pregnant women and newborn infants
• Data from the local criminal justice system that might suggest a communitywide drug problem or specific problems within geographically or demographically defined subpopulations
• Use of NICU services for infants

Summary: Use of Illicit Drugs During Pregnancy

All pregnant women should be asked about their use of illicit drugs and advised to abstain. Those who report using drugs during pregnancy need follow-up, supplementary case management, and counseling to receive optimal medical care.

TOC | Previous | Next