Unintended Pregnancy in Opioid-abusing Women

Abstract

The aim of this study was to estimate the prevalence of unintended pregnancy and its three subtypes (mistimed, unwanted, and ambivalent) among opioid-abusing women. In the general population, 31%-47% of pregnancies are unintended; data on unintended pregnancy in opioid- and other drug-abusing women are lacking. Pregnant opioid-abusing women (N = 946) screened for possible enrollment in a multisite randomized controlled trial comparing opioid maintenance medications completed a standardized interview assessing sociodemographic characteristics, current and past drug use, and pregnancy intention. Almost 9 of every 10 pregnancies were unintended (86%), with comparable percentages mistimed (34%), unwanted (27%), and ambivalent (26%). Irrespective of pregnancy intention, more than 90% of the total sample had a history of drug abuse treatment, averaging more than three treatment episodes. Interventions are sorely needed to address the extremely high rate of unintended pregnancy among opioid-abusing women. Drug treatment programs are likely to be an important setting for such interventions.

Full text

Unintended Pregnancy in Opioid-abusing Women
Sarah H. Heil, Ph.D1, Hendree E. Jones, Ph.D2, Amelia Arria, Ph.D3, Karol Kaltenbach,
Ph.D4, Mara Coyle, M.D5, Gabriele Fischer, M.D6, Susan Stine, M.D., Ph.D7, Peter Selby,
M.D8, and Peter R. Martin, M.D9
1 Departments of Psychiatry and Psychology, University of Vermont, Burlington, VT 05401 USA
2 Departments of Psychiatry and Behavioral Sciences and Obstetrics and Gynecology, Johns
Hopkins University School of Medicine, Baltimore, MD 21224 USA
3 Center for Young Adult Health and Development, University of Maryland, College Park, MD
20742 USA
4 Departments of Pediatrics, Psychiatry and Human Behavior, Thomas Jefferson University,
Philadelphia, PA 19107 USA
5 Department of Pediatrics, The Warren Alpert Medical School of Brown University, Providence,
RI 02912
6 Department of Psychiatry and Psychotherapy, Medical University of Vienna, Vienna, AUSTRIA
7 Department of Psychiatry and Behavioral Neurosciences, Wayne State University School of
Medicine, Detroit, MI 48207 USA
8 Addictions Program, Centre for Addiction and Mental Health, Toronto, Ontario M5S 2S1
CANADA
9 Departments of Psychiatry and Pharmacology and Addiction Center, Vanderbilt University
School of Medicine, Nashville, TN 37212 USA
Abstract
The aim of the present study was to estimate the prevalence of unintended pregnancy and its three
subtypes (mistimed, unwanted, ambivalent) among opioid-abusing women. In the general
population, 31–47% of pregnancies are unintended; data on unintended pregnancy in opioid- and
other drug-abusing women are lacking. Pregnant opioid-abusing women (N=946) screened for
possible enrollment in a multi-site randomized controlled trial comparing opioid maintenance
medications completed a standardized interview assessing sociodemographic characteristics,
current and past drug use, and pregnancy intention. Almost 9 of every 10 pregnancies were
unintended (86%), with comparable percentages mistimed (34%), unwanted (27%), and
ambivalent (26%). Irrespective of pregnancy intention, more than 90% of the total sample had a
history of drug abuse treatment, averaging more than 3 treatment episodes. Interventions are
sorely needed to address the extremely high rate of unintended pregnancy among opioid-abusing
women. Drug treatment programs are likely to be an important setting for such interventions.
Corresponding author: Sarah H. Heil, Rm. 1415 UHC, 1 So. Prospect St., Burlington, VT 05401, phone: 802-656-8712, FAX:
802-656-5793, sarah.heil@uvm.edu.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our
customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of
the resulting proof before it is published in its final citable form. Please note that during the production process errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
NIH Public Access
Author Manuscript
J Subst Abuse Treat. Author manuscript; available in PMC 2012 March 1.
Published in final edited form as:
J Subst Abuse Treat
. 2011 March ; 40(2): 199–202. doi:10.1016/j.jsat.2010.08.011.
NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript

Keywords
Pregnancy; intention; family planning; opioid; drug abuse
1. Introduction
Licit and illicit opioid dependence during pregnancy is often complicated by a multitude of
other factors, including low socioeconomic status, poor nutrition, lack of prenatal care,
family instability, interpersonal violence, homelessness, psychological problems, and other
drug use (Center for Substance Abuse Treatment, 1993). In the perinatal period, these
intertwined factors can contribute to a number of adverse maternal and infant outcomes
including, but not limited to, premature delivery, low birth weight, and neonatal abstinence
syndrome (see Kaltenbach et al., 1998 for a review). In the longer term, bearing a child in
such disadvantaged circumstances has been shown to significantly diminish the future
wellbeing of both the mother and the child (Graham 2007, 2009; Mishel et al., 2009).
Further compounding these difficult circumstances, opioid-dependent women become
pregnant more often than women in the general population. In a seminal study of the
reproductive health of opioid-dependent women, 54% reported having 4 or more
pregnancies in their lifetime compared to 14% of a nationally representative sample of US
women (Armstrong et al., 1999). These authors also observed that almost 5 times as many
opioid-dependent women reported ever having an abortion compared to women in the
national sample (57% vs. 12%), suggesting that many pregnancies among opioid-dependent
women were not intended.
To our knowledge, there is just one small study estimating unintended pregnancy among
opioid-dependent women. The results of this study indicated that 67% (24/36) of pregnant
women enrolled in a New York City methadone maintenance program reported they did not
plan the pregnancy (Selwyn et al., 1989). As a first step toward developing interventions to
reduce unintended pregnancy among opioid-dependent women, the present study sought to
estimate the prevalence of unintended pregnancy and its three subtypes (mistimed,
unwanted, and ambivalent) in a much larger sample of pregnant women reporting opioid
abuse.
2. Methods
2.1. Participants
Data were obtained from 946 opioid-abusing pregnant women screened for potential
enrollment in the MOTHER (Maternal Opioid Treatment: Human Experimental Research)
trial. This multi-site trial, performed at eight diverse U.S. and international clinical sites and
settings, was designed to compare the safety and efficacy of methadone and buprenorphine
for the treatment of opioid-dependence during pregnancy (Jones et al., 2008).
2.2. Screening Assessment
Participants who provided informed consent were screened for eligibility either at the time
of treatment entry or at the time they considered a change from their established drug
treatment program. Interviews were conducted with all potential participants to determine
eligibility for the study; at some sites, some information was collected by chart review prior
to the interview. Demographic information collected included age, education level, race, and
marital status. Drug use and treatment variables assessed included frequency of current
opioid and cocaine use and the number and type of prior treatment episodes.
Heil et al. Page 2
J Subst Abuse Treat. Author manuscript; available in PMC 2012 March 1.
NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript

Pregnancy intention of the current pregnancy was assessed by the question “When did you
intend to become pregnant?” Response options were “sooner”, “now”, “later”, “never”, and
“don’t know/unsure”. Women who responded that they intended to become pregnant
“sooner” or “now” were classified as having intended pregnancies. Women who responded
“later” were classified a having mistimed pregnancies. Women who responded “never” were
classified as having unwanted pregnancies. Women who responded “don’t know/unsure”
were classified as having ambivalent pregnancies (Mohlajee et al., 2007).
2.3. Data Analyses
Two types of analyses were performed to examine between-group differences. First,
analyses examined demographic differences between women with intended pregnancies and
women with unintended pregnancies. Statistically significant differences in continuous and
dichotomous variables were evaluated using t-tests, and z-tests, respectively. Second,
differences between groups on drug use and other factors were evaluated using logistic
regression models in which each variable of interest was entered separately into a logit
model controlling for age, race and site location.
3. Results
3.1. Pregnancy Intentions
Of 946 opioid-abusing women screened, 129 (14%) reported having intended pregnancies
and 817 (86%) reported having unintended pregnancies. As a percentage of all pregnancies,
323 (34%) were mistimed, 252 (27%) were unwanted, and 242 (26%) were ambivalent
pregnancies.
3.2. Pregnancy Intention and Maternal Demographic Characteristics and Drug Use
No significant differences were observed on the 5 maternal demographic characteristics
compared between women with intended vs. unintended pregnancies (top of Table 1).
Regarding the subtypes of unintended pregnancy, women with mistimed pregnancies were
significantly younger compared to women with intended pregnancies (t(450) = 2.1, p <
0.05). Women with unwanted pregnancies were significantly older (t(379) = 4.8, p < 0.001)
and less likely to be White (t(378) = 2.9, p < 0.01) compared to women with intended
pregnancies. Women with ambivalent pregnancies were significantly older (t(368) = 3.3, p =
0.001), less likely to be White (t(366) = 2.7, p < 0.01) and employed (t(354) = 2.8, p < 0.01)
compared to women with intended pregnancies.
Regarding maternal drug use, women with unintended pregnancies were more likely to have
used cocaine in the 30 days prior to screening compared to women with intended
pregnancies (adjusted odds ratio = 1.6, p < 0.05). Regarding the subtypes of unintended
pregnancy, women with mistimed pregnancies were less likely to have used cocaine in the
past 30 days compared to women with intended pregnancies (adjusted odds ratio = 1.8, p <
0.05). Women with ambivalent pregnancies were more likely to report prior medication-
assisted treatment compared to women with intended pregnancies (adjusted odds ratio = 0.5,
p < 0.05). [Table 1 about here]
4. Discussion
Unintended pregnancy was highly prevalent in this sample; nearly 9 of every 10 women
screened reported that the current pregnancy was unintended. This rate is 2–3 times the rate
observed in the general population (Chandra et al., 2005; Mohllajee et al., 2007; Williams et
al., 2006). In addition, the occurrence of unintended pregnancy in the current sample was
Heil et al. Page 3
J Subst Abuse Treat. Author manuscript; available in PMC 2012 March 1.
NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript

nearly 20% higher than previous estimates in pregnant women with opioid problems
(Selwyn et al., 1989).
To our knowledge, this is the first report of the rates of the three subtypes of unintended
pregnancy in opioid-abusing pregnant women. The percentage of women reporting
mistimed, unwanted or ambivalent pregnancies in the present sample were fairly
comparable, with each representing about one-third of the total sample. The percentage of
women reporting an unwanted pregnancy was nearly 3 times higher in the present study
compared to the general population and the percentage of women reporting ambivalence,
more than 4 times higher (Mohllajee et al., 2007). These figures dramatically underscore the
need to develop interventions to bring contraceptive use in line with conception desires
among opioid-abusing women.
Although there were few differences between women with intended vs. unintended
pregnancies, more differences emerged when women with unintended pregnancies were
disaggregated into the three subtypes of unintended pregnancy and compared to women with
intended pregnancies. Consistent with the literature on pregnancy intention in the general
population, women with mistimed pregnancies were younger (D’Angelo et al., 2004;
Mohlajee et al., 2007). A lower percentage of these women also reported recent cocaine use
compared to women with intended pregnancies. In studies of the general population, women
with mistimed pregnancies report more smoking, but less drinking compared to women with
intended pregnancies (D’Angelo et al., 2004; Mohlajee et al., 2007), suggesting some
variability in drug use among women with mistimed pregnancies.
Consistent with the literature in the general population, women with unwanted and
ambivalent pregnancies were older and less likely to be White compared to women with
intended pregnancies (D’Angelo et al., 2004; Mohlajee et al, 2007). Women with
ambivalent pregnancies were also more likely to be unemployed and a higher percentage
reported prior medication-assisted treatment. Overall, the greatest number of differences was
observed between women with ambivalent vs. intended pregnancies. This is in contrast to
the general population literature, where women with ambivalent pregnancies tend to be most
similar to women with intended pregnancies in terms of demographic characteristics as well
as maternal and infant outcomes (Mohlajee et al., 2007). Additional studies will be needed
to replicate this pattern of results and to determine the implications of such differences.
Although there were no differences as a function of pregnancy intention on this variable, it
is notable that more than 90% of the total sample had a history of prior drug treatment,
averaging more than 3 episodes. These data suggest that drug abuse treatment programs may
be an important setting for interventions to reduce the very high rate of unintended
pregnancy in this population. In the late 1980’s, the Centers for Disease Control funded
several demonstration projects designed to improve access to reproductive health services
for women at high risk of unintended pregnancy and HIV infection, including women with
substance use disorders (see Armstrong et al., 1999). One strategy for doing so involved
integrating free family planning services into drug treatment programs. The limited results
reported from these projects suggests that women who received family planning services,
including inexpensive referral services, in their drug treatment program were more likely to
be using contraception at follow-up than women who didn’t (CDC, 1995). These findings
suggest that this is a promising model that should be further developed and rigorously tested
as part of efforts to reduce unintended pregnancy among drug-abusing women.
The present study has notable strengths. The data were systematically collected across eight
diverse U.S. and international clinical sites and settings and represent the largest dataset to
date on the topic of pregnancy intention in pregnant women with substance use disorders.
Heil et al. Page 4
J Subst Abuse Treat. Author manuscript; available in PMC 2012 March 1.
NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript

The study also has limitations. The format of the pregnancy intention question differed from
the format used in national surveys (e.g., the National Survey on Family Growth, Pregnancy
Risk Assessment Monitoring System) and has not been formally validated in women with
substance use disorders. Also, it is possible that women who were screened for potential
study participation may not be representative of the larger population of opioid-dependent
women. Nevertheless, the results of the present study clearly document the extremely high
rate of unintended pregnancy among a large sample of opioid-abusing women and
underscore the need for a greater scientific attention to this serious problem.
Acknowledgments
Funding for this study was provided by NIDA research grants RO1 DA 015738, 015741, 015764, 015778, 015832,
017513, 018410, and 018417. We thank Laura Garnier for assistance with statistical analyses.
References
Armstrong KA, Kennedy MG, Kline A, Tunstall C. Reproductive health needs: comparing women at
high, drug-related risk of HIV with a national sample. Journal of the American Medical Women’s
Association 1999;54:65–70.
Centers for Disease Control. What we have learned … 1990–1995. Retrieved from
http://www.cdc.gov/std/research/older/wwhl-1990-1995/learn0.htm
Center for Substance Abuse Treatment. Treatment Improvement Protocol (TIP) Series 5. Rockville,
MD: Substance Abuse and Mental Health Services Administration; 1993. Improving treatment for
drug-exposed infants.
Chandra A, Martinez GM, Mosher WD, Abma JC, Jones J. Fertility, family planning, and reproductive
health of U.S. women: Data from the 2002 National Survey of Family Growth. Vital and Health
Statistics 2005:1–160. [PubMed: 16285217]
D’Angelo DV, Gilbert BC, Rochat RW, Santelli JS, Herold JM. Differences between mistimed and
unwanted pregnancies among women who have live births. Perspectives on Sexual and
Reproductive Health 2004;36:192–197. [PubMed: 15519961]
Graham, H. Unequal lives: health and socioeconomic inequalities. Berkshire, England: Open
University Press; 2007.
Graham H. Women and smoking: Understanding socioeconomic influences. Drug and Alcohol
Dependence 2009;104(Suppl 1):S11–16. [PubMed: 19345520]
Jones HE, Martin PR, Heil SH, Kaltenbach K, Selby P, Coyle MG, Stine SM, O’Grady KE, Arria AM,
Fischer G. Treatment of opioid dependent pregnant women: Clinical and research issues. Journal of
Substance Abuse Treatment 2008;35:245–259. [PubMed: 18248941]
Kaltenbach K, Berghella V, Finnegan L. Opioid dependence during pregnancy. Effects and
management. Obstetrics and Gynecology Clinics of North America 1998;25:139–151. [PubMed:
9547764]
Mishel, Ll; Berstein, J.; Shierholz, H. The state of working America, 2008/2009. Ithaca, NY: ILR
Press; 2009.
Mohllajee AP, Curtis KM, Morrow B, Marchbanks PA. Pregnancy intention and its relationship to
birth and maternal outcomes. Obstetrics and Gynecology 2007;109:678–686. [PubMed:
17329520]
Selwyn PA, Carter RJ, Schoenbaum EE, Robertson J, Klein RS, Rogers MF. Knowledge of HIV
antibody status and decisions to continue or terminate pregnancy among intravenous drug users.
JAMA 1989;261:3567–2571. [PubMed: 2724503]
Williams, L.; Morrow, B.; Shulman, H.; Stephens, R.; D’Angelo, D.; Fowler, CI. PRAMS 2002
Surveillance Report. Retrieved from
http://www.cdc.gov/PRAMS/2002PRAMSSurvReport/Index.htm
Heil et al. Page 5
J Subst Abuse Treat. Author manuscript; available in PMC 2012 March 1.
NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript

NIH-PA Author Manuscript NIH-PA Author Manuscript NIH-PA Author Manuscript
Heil et al. Page 6
Table 1
Maternal Demographic Characteristics and Drug Use by Pregnancy Intention
Total N=946^Intended n=129 (14%) Unintended n=817 (86%)
Unintended Pregnancy Subtypes
Mistimed n=323 (34%) Unwanted n=252 (27%) Ambivalent n=242 (26%)
Demographic characteristics
Mean (SD) age 27.9 (5.9) 27.0 (5.4) 28.1 (5.9) 25.8 (5.4)*30.1 (6.1)*29.0 (5.5)*
% White 78 82 77 89 69*70*
Mean (SD) years of education 11.1 (1.8) 11.2 (2.1) 11.1 (1.8) 11.1 (1.7) 11.2 (1.9) 11.1 (1.9)
% married 11 13 11 9 11 13
% employed 11 15 11 13 12 6*
Drug use‡
% with prior drug treatment 91 91 91 90 90 95
% with prior medication-assisted
treatment 88 87 88 84 89 92*
Mean (SD) number of times treated
for drug abuse in lifetime 3.2 (3.6) 3.2 (4.0) 3.2 (3.5) 2.9 (2.9) 3.4 (4.2) 3.3 (3.2)
Mean (SD) years of age at 1st
medication-assisted treatment 24.8 (5.5) 23.8 (5.1) 25 (5.6) 22.9 (4.9) 26.3 (5.8) 26.2 (5.4)
% with daily illicit/non-medical
opioid use in the 30 days prior to
screening
83 72 85 74 91 93
% with cocaine use in the past 30
days 40 40 40*28*48 49
^Ns vary by characteristic due to missing data and range from n=726–945
*Significantly different (p<.05) from intended pregnancy group
‡Analyses controlled for age, race, and site
J Subst Abuse Treat. Author manuscript; available in PMC 2012 March 1.