Maternal Prenatal Depressive Symptoms, Nicotine Addiction,and Smoking-Related Knowledge, Attitudes, Beliefs,and Behaviors

Suezanne Tangerose Orr • Dan G. Blazer •

Caroline A. Orr

Published online: 24 May 2011

� Springer Science+Business Media, LLC 2011

Abstract Maternal smoking is a key preventable cause of

poor pregnancy outcomes, such as low birthweight. In

many areas of the United States, including Eastern North

Carolina, rates of prenatal smoking are high. Prenatal

depressive symptoms are associated with maternal smok-

ing, but there remains much to learn about this relationship,

especially among Black women, who have double the risk

of poor pregnancy outcomes of White women. In the study

reported in this paper, we investigated the relationship

between maternal prenatal depressive symptoms with

smoking behaviors, beliefs and attitudes, environmental

factors which promote smoking and nicotine addiction.

Pregnant women were enrolled in the study at the first

prenatal visit to the clinics of the Departments of Obstetrics

and Gynecology and Family Medicine of the Brody School

of Medicine, East Carolina University. An interviewer

administered a questionnaire to each woman about smok-

ing, smoking-related attitudes, knowledge, beliefs and

behaviors, nicotine addiction, and home environmental

factors that encourage smoking. The CES-D was used to

measure depressive symptoms. We used the cut-point score

of 23 or greater to indicate elevated depressive symptoms,

which is thought to represent major depressive disorder.

The sample consisted of 810 Black women, of whom 18%

were smokers. CES-D score was associated with nicotine

addiction, not thinking of quitting smoking, and not

expecting support from family and friends if they decided

to quit. Prenatal depressive symptoms may be a barrier to

smoking cessation.

Keywords Prenatal smoking � Pregnancy

Introduction

Maternal prenatal smoking is one of the key preventable

causes of poor pregnancy outcomes, such as low birth-

weight [1–3]. Despite widespread knowledge of the

harmful effects of smoking upon pregnancy outcomes,

many women continue to smoke during pregnancy, espe-

cially in certain areas of the country. In our prior work, we

demonstrated that close to 20% of pregnant Black women

seeking prenatal care in Greenville, North Carolina in

2001–02 smoked during pregnancy [4]. Nationally, during

this time period, 12% of pregnant women smoked [5]. The

Southeastern states, including North Carolina, have among

the highest rates of smoking during pregnancy [6]. In

addition, North Carolina has one of the highest infant

mortality rates in the United States, which is partially

attributable to prenatal smoking and its relationship to low

birthweight and preterm births. Thus, the prevention or

cessation of prenatal smoking is an important public health

issue.

In our prior work, we demonstrated that elevated levels

of maternal depressive symptoms were significantly asso-

ciated with prenatal smoking among pregnant Black

women [4]. (In logistic regression analysis in the prior

S. T. Orr (&)

Department of Health Education and Promotion (ret.),

College of Health and Human Performance,

East Carolina University, 410 Kempton Drive,

Greenville, NC 27834, USA

e-mail: orrs@ecu.edu

D. G. Blazer

Department of Psychiatry and Behavioral Sciences,

Duke University Medical Center, Durham, NC, USA

C. A. Orr

Department of Psychology, Loyola University,

Baltimore, MD, USA

123

Matern Child Health J (2012) 16:973–978

DOI 10.1007/s10995-011-0822-9

study, the following variables were associated with pre-

natal smoking among Black women: elevated depressive

symptoms; being single; belief that smoking will harm

their baby or themselves; living with other smokers in the

home; and allowing smoking in the home [4]).

In the current study, we sought to identify those factors

associated with increased risk of smoking which were also

associated with elevated maternal depressive symptoms.

Such factors include home environmental factors (e.g.,

allowing smoking in the home); attitudes, beliefs and

knowledge about smoking (e.g., knowledge that smoking

could harm their unborn baby); and nicotine addiction

(e.g., number of cigarettes smoked per day). These are the

major factors associated with prenatal smoking [5, 6].

We assessed the relationships between maternal prenatal

depressive symptoms with measures of each of these

factors. Identifying the associations between maternal

depressive symptoms and factors associated with prenatal

smoking might help providers of prenatal care to focus

their educational efforts on the appropriate factors.

Other research has similarly demonstrated associations

between depressive symptoms and smoking, but most prior

research has either not focused upon prenatal smoking [7],

or has focused on White samples [8]. None of the prior

research of which we are aware has explored factors

associated with smoking and depressive symptoms among

Black women. Since Black women have such high rates of

poor pregnancy outcomes, it is especially important to

study smoking among this largely understudied group.

Pregnant Black women are also more likely to be depressed

than their White counterparts [9], so focusing attention on

depression among Black women is significant. Finally,

even though Black pregnant women are less likely to

smoke than pregnant White women, they are more likely to

smoke cigarettes with greater nicotine content, which

increases the danger to the fetus and the risk of addiction

[10–13].

Methods

Pregnant women were enrolled in the study at the time of

their first prenatal visit to the prenatal clinics of the

Departments of Obstetrics and Gynecology and Family

Medicine of the Brody School of Medicine, East Carolina

University, in Greenville, North Carolina. The study was

approved by the Institutional Review Board of East Caro-

lina University. All patients ages 16 years and older were

eligible for participation in the study. Each woman who

presented for her first prenatal visit and was 16 years or

older was approached by a trained research assistant and

invited to participate in the study. Written informed

consent was obtained from the women at this time. Refusal

rates were very low (\5%).

The research assistant administered a face-to-face

interview to all participants while they waited to be seen by

the physician. Interviews took place in a private room, and

anyone accompanying the women were not included in the

interviewing process. Interviews were conducted from

March 2001 until November 2002. The analyses reported

in this paper are focused upon the women who self-iden-

tified their race as Black or African-American (70% of the

total sample).

The study interview included items to assess smoking-

related behaviors, knowledge, attitudes, beliefs and nico-

tine addiction. Most of these items were developed by the

Robert Wood Johnson Foundation’s Smoke-Free Families

program. When this program was initiated, a working

group of experts conducted an exhaustive literature review

and developed the most useful measures for smoking-

related behaviors, knowledge, beliefs, attitudes and nico-

tine addiction [14].

One item was used to assess smoking status. This item

asked women, ‘‘Which statement best describes you

now?’’ There were six response categories which allowed

the women to classify themselves as lifetime nonsmokers;

current smokers who smoked the same (I smoke regularly

now, about the same as before I found out I was preg-

nant.); more (I smoke regularly now, but more than

before I found out I was pregnant); or less than prior to

pregnancy (I smoke some now, but I cut down on the

number of cigarettes I smoke since I found out I was

pregnant); or as having quit smoking prior to (I stopped

smoking before I found out I was pregnant and I am not

smoking now) or after learning of the pregnancy (I

stopped smoking after I found out I was pregnant and I

am not smoking now). Melvin et al. have found that

offering women several response options provides more

valid information about smoking in lieu of a yes/no type

of question [14, 15].

For all women who reported that they had quit smoking,

biochemical confirmation was used to assess validity of

self-reports. Urinary cotinine was used for these assess-

ments. Cotinine is a stable metabolite of nicotine, and

testing for cotinine is a well-accepted standard for vali-

dating reports of smoking cessation [16, 17]. Urinary

cotinine assessment was performed on all women who

reported to the interviewer that they had stopped smoking

within the past 12 months. Urine was obtained at the same

time as the interview and was frozen in tubes that were

labeled with the study ID number and shipped weekly to

J-2 Laboratories of Phoeniz, Arizona. The laboratory used

enzyme multiplied immunoassay technique (EMIT) using

an FDA approved reagent system purchased from Diag-

nostic Reagents, Inc. on an automated chemistry analyzer

974 Matern Child Health J (2012) 16:973–978

123

(Olympus AU 640). The lowest concentration that can be

detected using this technique is 50 ng/ml. The laboratory

was blinded to the smoking status of the women A cut-

point of 80 ng/ml of urinary cotinine was used to identify

levels sufficient to indicate current smoking. Women who

reported themselves to be quitters but who had cotinine

levels above this cut-point were reclassified as smokers

(N = 20 of 102 tested).

The interview also included items to assess beliefs,

knowledge and attitudes about smoking, including beliefs

about the potentially harmful effects of smoking upon their

own health or that of their unborn child; desire to quit

smoking; desire to keep smoking; and number of attempts

at smoking cessation during the pregnancy. Respondents

were also asked about nicotine addiction (age initiated

smoking, number of cigarettes smoked per day, and time

from awakening to the first cigarette) and home environ-

mental factors such as the number of smokers living in

their home and allowing smoking in the home.

Depressive symptoms were assessed using the twenty

item CES-D (Center for Epidemiologic Studies’ Depres-

sion Scale), which was developed by NIMH to allow the

assessment of depression in community-based samples

[18]. The CES-D contains items to assess the presence (in

the past week) of salient symptoms of depression such as

sadness, crying, hopelessness, loss of pleasure, and fatigue.

Scores can range from zero to sixty, with a customary cut-

point score of 16 or higher used to indicate ‘‘elevated’’

levels of depressive symptoms. However, Radloff and

Locke have suggested that a cut-point of 23 or higher

corresponds to major depressive disorder and we used this

cut-point [19] Our rationale included that some symptoms

of pregnancy overlap with symptoms of depression (e.g.,

fatigue, changes is sleep or appetite). However, women

would be very unlikely to score 23 or greater based solely

on symptoms of pregnancy.

The CES-D has been widely used in epidemiologic

studies, and its reliability and validity with diverse samples

are well established [18, 20–22]. CES-D scores correlate

well with clinical assessments, and CES-D scores decline

with treatment of depression [18].

Descriptive analyses were used to identify the smoking,

demographic, and other characteristics of the sample.

Bivariate analyses and the chi squared statistic were used to

evaluate the associations between smoking status (i.e.,

smoker, nonsmoker) with each variable, and CES-D score

with knowledge, beliefs, and attitudes about smoking,

nicotine addiction, and home environmental factors. We

performed all analyses using the cut-point of 23 on the

CES-D. We were unable to perform multivariable analysis

(e.g., logistic regression) because the number of smokers

was small and regression coefficients were unstable due to

lack of power.

Results

A total of 810 Black women comprised the sample for

analysis. Of these women (after inclusion of results of

cotinine analysis), 149 were classified as current smokers

(18.4%), 82 as former smokers (10.1%) and 579 as non-

smokers (71.5%). CES-D score, as shown in Table 1, was

strongly associated with smoking status, with 14.3% of

those women with low CES-D scores and 29.1% of women

with elevated CES-D scores classified as smokers

(P \ .001). In addition, women who did successfully quit

smoking with elevated CES-D scores were significantly

more likely (11.2%) than those with lower CES-D scores

(6.6%) to cease smoking after learning of the pregnancy

rather than prior to learning of the pregnancy. In our prior

work, we demonstrated that women who quit smoking after

learning of the pregnancy were more likely to relapse than

those who quit prior to pregnancy [4].

As shown in Table 1, the women who smoked were

somewhat different from the women who did not smoke.

Women who smoked were significantly more likely to have

lower levels of education, to be single, to have enrolled for

prenatal care in the second or third trimester, to allow

smoking in the home, to live with other smokers, and to

believe that smoking could not harm the baby or themselves

‘‘a lot.’’ Thus, smokers differed from nonsmokers in knowl-

edge, attitudinal and environmental factors about smoking.

About one-quarter of the women had CES-D scores

above the 23 cut-point which was used to indicate elevated

levels of depressive symptoms.

Table 2 shows data from the 149 smokers, comparing

those with elevated and lower levels of depressive symp-

toms. There are missing values for some of these items,

because women who were self-reported former smokers

(‘‘quitters’’) who later were shown, by cotinine testing, to

be current smokers, were not asked these questions. There

were significant differences in the two groups in the

number of cigarettes smoked per day prior to pregnancy,

with 25% of those with elevated CES-D scores smoking 10

or more cigarettes per day compared to 9.2% of those with

lower scores on the CES-D. Women with elevated CES-D

scores also smoked significantly more cigarettes per day

during pregnancy, with 23.3% reporting smoking 10 or

more cigarettes per day, compared to 8.6% of those with

lower CES-D scores.

Also, women with elevated CES-D scores were signifi-

cantly more likely to report not seriously thinking of

quitting smoking during the pregnancy compared to those

with lower scores (20.4 and 10.3%). Finally, women with

elevated CES-D scores were significantly more likely to

report that they would receive no support from family and

friends if they tried to quit smoking (20.8%) compared to

those with lower CES-D scores (7.2%).

Matern Child Health J (2012) 16:973–978 975

123

There were no significant differences by CES-D score in

the age women started smoking, home environmental

factors, or knowledge about the potential harm of smoking

to themselves or their baby (data not shown).

Discussion

Perhaps the most striking finding from our study is the

association between elevated prenatal depressive

symptoms and nicotine addiction. Women with elevated

CES-D scores smoked more than women with lower scores

both before and during pregnancy. Others have similarly

found, in non-pregnant samples, associations between

depression and nicotine addiction [23]. This would suggest

that women with elevated depressive symptoms may be a

particularly difficult group in which to achieve prenatal

smoking cessation, since they are more addicted to nicotine

and smoke more cigarettes. Special and intensive efforts

may be needed to achieve cessation among this group.

Table 1 Characteristics of the

sample, by smoking status

(N = 810)

* P B .05; ** P B .01;

*** P B .001

Variables Smoking status

Smokers Nonsmokers

(N = 149) (N = 661)

N (%) N (%)

Age

\20 30 (20.1) 149 (22.6)

C20 119 (79.9) 511 (77.4)

Education***

\h.s. grad 70 (47.0) 188 (28.4)

h.s. grad 59 (39.6) 271 (41.0)

[h.s. grad 20 (13.4) 202 (30.6)

Marital status**

Married, living with 36 (24.2) 240 (36.3)

Single 113 (75.8) 421 (63.7)

Trimester enrolled for care*

First 67 (45.3) 350 (53.4)

Second, third 81 (54.7) 306 (46.6)

Allowed to smoke in home***

No 43 (28.9) 444 (67.4)

Yes 106 (71.1) 215 (32.6)

Number of smokers in home (Excluding Self)**

None 90 (60.4) 466 (71.8)

1 or more 59 (39.6) 183 (28.2)

Smoking can harm self**

A lot 123 (83.7) 589 (90.6)

Else 24 (16.3) 61 (9.4)

Smoking can harm baby**

A lot 125 (85.6) 600 (92.6)

Else 21 (14.4) 48 (7.4)

Smoked in past 7 days***

No 32 (21.5) 653 (99.1)

Yes 117 (78.5) 6 (.9)

Smoked in past 30 days***

No 27 (18.1) 639 (97.0)

Yes 122 (81.9) 20 (3.0)

CES-D score***

0–22 84 (56.4) 503 (76.1)

C23 65 (43.6) 158 (23.9)

976 Matern Child Health J (2012) 16:973–978

123

It is also of interest that women with higher levels of

depressive symptoms are less likely to express serious

interest in quitting compared to those with lower levels of

CES-D scores. This may also function to make them a

difficult group with which to achieve smoking cessation,

and is likely related to their high levels of nicotine addic-

tion. Similarly, the high percentage of women (21%) with

elevated CES-D scores who perceive that they would

receive no support from friends and family if they decided

to quit smoking suggests that achieving smoking cessation

with this group may be very difficult.

One limitation of our study is the missing information

about smoking-related knowledge, beliefs, and behaviors

from women who self-reported themselves to be quitters

but who were later shown to be smokers through cotinine

testing. Also, we did not test women who claimed to be

lifetime nonsmokers to determine their cotinine levels, due

to financial constraints of testing hundreds of women.

However, this does not allow us to know what percentage

of women who self-reported themselves to be lifelong

nonsmokers were actually smokers.

Another limitation is that our sample may not be rep-

resentative of all pregnant Black women. It would be

helpful to replicate our work in other settings. However,

the sample is large and is very representative of pregnant,

Black clinic-attenders in southern communities. Very few

women in Greenville who are Medicaid recipients attend

other clinics. We also had a very low refusal rate.

A strength of our study is that we had biochemical

confirmation of smoking status for women who claimed to

have quit smoking. This allowed us to reclassify almost

20% of self-reported former smokers as current smokers.

One limitation of this process is that we had no information

about use of nicotine patches or gum, so there may have

been women who were not smoking who were falsely

classified as smokers due to their use of these substances.

Black pregnant women who smoke and have elevated

depressive symptoms clearly are an important group for

health education programs for smoking cessation. Despite

the potential difficulties associated with intervening with

women with high levels of depressive symptoms and nic-

otine addiction, it is important to develop appropriate and

effective programs as a method to reduce the high rates of

low birthweight, preterm birth, and infant mortality among

Blacks. In addition, such programs may help to protect

infants from harmful effects of second-hand smoke in

the postpartum period. It is important to intervene with

smoking cessation programs as early in pregnancy as

possible to reduce harm to the fetus. While treatment of

depression is important, waiting for depression to respond

Table 2 Selected bivariate

relationships between CES-D

score and smoking variables

among smokers

* .05 \ P \ .10; ** P \ .05;

*** P \ .01; **** P \ .001

Variables CES-D score

0–22 C23

N (%) N (%)

Age initiated smoking

B19 64 (83.1) 41 (73.2)

C20 13 (16.9) 15 (26.8)

Number of cigarettes smoked before pregnancy**

\1 pack/day 47 (61.8) 26 (50.0)

1 pack/day 22 (28.9) 13 (25.0)

[1 pack/day 7 (9.2) 13 (25.0)

Number of cigarettes smoked in the past week**

None 24 (25.8) 10 (16.7)

\10 61 (65.6) 36 (60.0)

C10 8 (8.6) 14 (23.3)

Seriously thinking of quitting smoking?*

No 7 (10.3) 10 (20.4)

Yes 61 (89.7) 39 (79.6)

If you decided to quit smoking, how much support

would you get from family and friends?*

None 5 (7.2) 10 (20.8)

Not much 4 (5.8) 3 (6.3)

Some 10 (14.5) 10 (10.8)

A lot 50 (72.5) 25 (52.1)

Matern Child Health J (2012) 16:973–978 977

123

to treatment may delay smoking cessation, so the most

effective approach may be to combine smoking cessation

with treatment of depression.

One might speculate that depression influences behav-

ior, motivation, attitudes and perceptions about smoking. If

this is so, then interventions to reduce depression among

women of childbearing ages might help to reduce smoking

among pregnant women. Such interventions could include

the provision of social support [24], or preconception care

of mental disorders [25]. This would have the potential to

affect both smoking behavior and other risks for women

with depression and their infants. Calls for preconception

health care should include preconception mental health

care and smoking cessation.

Acknowledgments This research was partially supported by Grant

No. 040679 from the Robert Wood Johnson Foundation.

References

1. Behrman, R. E., & Butler, A. S. (Eds.). (2006). Preterm birth:Causes, consequences and prevention. Washington D.C.:

National Academy Press.

2. Martin, J. A, Hamilton, B. E., Sutton, P. D., Ventura, S. J.,

Menacker, F., & Kirmeyer, S. et al. (2007). Births: Final data for

2005. National Vital Statistics Reports, Vol. 56 no. 6. Hyattsville,

MD: National Center for Health Statistics.

3. Savitz, D. A., & Pastore, L. M. (1999). Causes of prematurity. In

M. C. McCormack & J. E. Siegel (Eds.), Prenatal care: Effec-tiveness and implementation (pp. 63–104). Cambridge: Cam-

bridge University Press.

4. Orr, S. T., Newton, E., Tarwater, P. M., & Weismiller, D. (2005).

Factors associated with prenatal smoking among black women in

Eastern North Carolina. Maternal and Child Health Journal, 9,

245–252.

5. CDC. (2004). Smoking during pregnancy—United States,

1990–2002. Morbidity and Mortality Weekly Report, 53,911–914.

6. Cnattingius, S. (2004). The epidemiology of smoking during

pregnancy: Smoking prevalence, maternal characteristics, and

pregnancy outcomes. Nicotine and Tobacco Research, 6, S125–

S140.

7. Ockene, J. K., Emmons, K. M., Mermelstein, R. J., Perkins,

K. A., Bonollo, D., Voorheed, C. C., et al. (2000). Relapse and

maintenance issues for smoking cessation. Health Psychology,19, 17–31.

8. Zhu, S. H., & Valbo, A. (2002). Depression and smoking during

pregnancy. Addictive Behaviors, 27, 649–658.

9. Orr, S. T., Blazer, D. G., & James, S. A. (2006). Racial disparities

in elevated prenatal depressive symptoms among black and white

women in Eastern North Carolina. Annals of Epidemiology, 16,

463–468.

10. English, P. B., Eskenazi, B., & Christenson, R. E. (1994). Black–

White differences in serum cotinine levels among pregnant

women and subsequent effects on infant birthweight. AmericanJournal of Public Health, 84, 1439–1443.

11. Hebert, J. R., & Kabat, G. C. (1989). Menthol cigarette smoking

and oesophageal cancer. International Journal of Epidemiology,18, 37–44.

12. Sidney, S., Tekawa, I., & Friedman, G. D. (1989). Mentholated

cigarette use among multiphasic examinees 1979–86. AmericanJournal of Public Health, 79, 1415–1416.

13. Cummings, K. M., Giovino, G., & Mendicino, A. J. (1987).

Cigarette advertising and racial differences in cigarette brand

preference. Public Health Reports, 102, 698–701.

14. Melvin, C., Tucker, P., and the Smoke-Free Families Common

Evaluation Measures for Pregnancy and Smoking Cessation

Projects Working Group. (2000). Measurement and definition for

smoking cessation intervention research: The smoke-free families

experience. Tobacco Control, 9, iii87–iii90.

15. Melvin, C. L., & Gaffney, C. A. (2004). Treating nicotine use and

dependence of pregnant and parenting smokers: An update.

Nicotine and Tobacco Research, 6, S107–S124.

16. Klebanoff, M. A., Levine, R. J., Clemens, J. D., DerSimonian, R.,

& Wilkins, D. G. (1998). Serum cotinine concentration and self-

reported smoking during pregnancy. American Journal of Epi-demiology, 148, 259–262.

17. Klebanoff, M. A., Levine, R. J., Morris, C. D., Hauth, J. C., Sibai,

B. M., Ben, C., et al. (2001). Accuracy of self-reported cigarette

smoking among pregnant women in the 1990s. Paediatric andPerinatal Epidemiology, 15, 140–143.

18. Radloff, L. S. (1977). The CES-D scale; a self-report depression

scale for research in the general population. Applied Psycholog-ical Measurement, 1, 385–401.

19. Radloff, L. S., & Locke, B. (1986). The community mental health

assessment survey and CES-D scale. In M. M. Weissman &

J. K. Myers (Eds.), Community surveys of psychiatric disorders(pp. 177–187). New Brunswick, NJ: Rutgers University Press.

20. Weissman, M. M., Sholomskas, D., Pottenger, M., Prusoff, B. A.,

& Locke, B. Z. (1997). Assessing depressive symptoms in five

psychiatric populations: A validation study. American Journal ofEpidemiology, 106, 203–214.

21. Husaini, B. A., Neff, J. A., Harrington, J. B., & Hughes, M. D.

(1980). Depression in rural communities: Validating the CES-D

scale. Journal of Community Psychology, 8, 20–27.

22. Markush, R. E., & Favero, R. V. (1974). Epidemiologic assess-

ment of stressful life events, depressed mood, and psychophysi-

ological symptoms—a preliminary report. In B. S. Dohrenwend

& B. P. Dohrenwend (Eds.), Stressful life events: Their natureand effects (pp. 171–190). New York, NY: Wiley.

23. Breslau, N., Peterson, E. L., Schultz, L. R., Chilcoat, H. D., &

Andreski, P. (1998). Major depression and stages of smoking.

Archives of General Psychiatry, 55, 161–166.

24. Berkman, L. F. (1995). The role of social relations in health

promotion. Psychosomatic Medicine, 57, 245–254.

25. Fiore, E. (2002). March of dimes updates: Is early prenatal care

too late? Contemporary Ob/Gyn, 12, 54–72.

978 Matern Child Health J (2012) 16:973–978

123