Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2016 Apr 23.
Published in final edited form as: Prev Med. 2015 Apr 18;80:23–31. doi: 10.1016/j.ypmed.2015.04.008

Contraceptive use and method choice among women with opioid and other substance use disorders: A systematic review

Mishka Terplan a,b,*, Dennis J Hand c,d, Melissa Hutchinson a,b, Elizabeth Salisbury-Afshar a,b, Sarah H Heil c,d,e
PMCID: PMC4842019  NIHMSID: NIHMS777425  PMID: 25900803

Abstract

Aim

To systematically review the literature on contraceptive use by women with opioid and other substance use disorders in order to estimate overall contraceptive use and to examine method choice given the alarmingly high rate of unintended pregnancy in this population.

Method

Pubmed (1948–2014) and PsycINFO (1806–2014) databases were searched for peer-reviewed journal articles using a systematic search strategy. Only articles published in English and reporting contraceptive use within samples of women with opioid and other substance use disorders were eligible for inclusion.

Results

Out of 580 abstracts reviewed, 105 articles were given a full-text review, and 24 studies met the inclusion criteria. The majority (51%) of women in these studies reported using opioids, with much smaller percentages reporting alcohol and cocaine use. Across studies, contraceptive prevalence ranged widely, from 6%–77%, with a median of 55%. Results from a small subset of studies (N = 6) suggest that women with opioid and other substance use disorders used contraception less often than non-drug-using comparison populations (56% vs. 81%, respectively). Regarding method choice, condoms were the most prevalent method, accounting for a median of 62% of contraceptives used, while use of more effective methods, especially implants and intrauterine devices (IUDs), was far less prevalent 8%.

Conclusions

Women with opioid and other substance use disorders have an unmet need for contraception, especially for the most effective methods. Offering contraception services in conjunction with substance use treatment and promoting use of more effective methods could help meet this need and reduce unintended pregnancy in this population.

Keywords: Contraceptive use, Unintended pregnancy, Substance use disorders, Women, Opioids

Introduction

Drug and alcohol use are common among women of reproductive age (15–44 years). In the US, 11% report illicit drug use in the past month, with marijuana and opioids accounting for the overwhelming majority of this use (Substance Abuse and Mental Health Services Administration, 2014). Twenty-five percent of US women of reproductive age also report binge alcohol use in the past month, with 5% meeting criteria for heavy alcohol use (Substance Abuse and Mental Health Services Administration, 2014). Worldwide, nearly 16 million are estimated to be dependent on illicit drugs (Degenhardt et al., 2013) and an estimated 63 million have an alcohol use disorder (Rehm et al., 2009).

Drug and alcohol use by women of reproductive age are associated with many negative outcomes, but one that has received relatively little attention is the high rate of unintended pregnancy. For example, unplanned pregnancy rates approach 8 of every 10 pregnancies among opioid-using women (Black et al., 2012; Heil et al., 2011; Jones et al., 2011) and women with substance use disorders have significantly greater odds of an unintended pregnancy as compared to non-drug-using women (Than et al., 2005; Wellings et al., 2013). Together, these results suggest that women with substance use disorders may have an unmet contraceptive need.

Contraceptive need can be met by a variety of contraceptive methods, however the effectiveness of these methods varies widely. For example, the World Health Organization divides contraceptive methods into four tiers of effectiveness (WHO, 2007). In the top tier are the “very effective” methods of implants, intrauterine devices (IUDs), and tubal ligation. Fewer than 1% of women using one of these methods will become pregnant in the first year of typical use because these methods do not require any additional effort on the part of the user to maintain maximum effectiveness. [Of note, because implants and IUDs provide protection for up to 10 years, but removal results in a rapid return of fertility, they are often grouped together and referred to as long-acting reversible contraceptives or LARCs.] In the second tier are the “effective” methods of depot injections, oral contraceptive pills, vaginal rings, and transdermal patches. Between 1 and 9% of women using one of these methods will become pregnant because their effectiveness is dependent upon continued effort by user, from taking a pill every day to getting an injection every three months, to prevent a return to fertility. In the third tier are the “moderately effective” methods of condoms, diaphragms, sponges, and fertility awareness methods (10–25% pregnancy rates) and in the fourth tier, the “less effective” methods of spermicides and withdrawal (≥26% pregnancy rates). The methods in the third and fourth tiers are less effective than methods in higher tiers because their effectiveness is dependent upon continued effort by the user and this effort is needed at or around the time of intercourse, when contraceptive decision-making is likely impaired by sexual arousal (e.g., Ariely and Loewenstein, 2006).

Assessing overall contraceptive use and specific method choice among women with opioid and other substance use disorders is difficult using standard epidemiological approaches. National-level surveys of drug use epidemiology rarely capture contraception data, and contraceptive use and method choice surveillance surveys rarely include questions about drug or alcohol use. Hence, a systematic review of the published literature was undertaken to estimate overall contraceptive use and to examine method choice among women with opioid and other substance use disorders given the alarmingly high rate of unintended pregnancy in this population.

Methods

Pubmed and PsycINFO were searched through August 21, 2014. The keywords and search strategy are described in Table 1. The first aim of the review was to describe prevalence of contraceptive use among women with opioid and other substance use disorders. The second aim was to describe method choice among contraceptive-using women in this vulnerable population. Thus, to be included in the present review, studies must have 1) reported contraceptive use, and 2) had a population comprised of at least 50% women with opioid and other substance use disorders, meaning actively using drugs and/or alcohol or in drug treatment. So as not to underestimate contraceptive use, contraceptive prevalence was calculated from the proportion of the study population who were at risk of pregnancy (i.e., those not pregnant, nor premenopausal, and without a history of hysterectomy at the time when contraceptive use was assessed). Comparison prevalence (either for contraceptive use or method choice) was reported when included in the original study. Although no geographic restrictions were placed, the search strategy was limited to English language articles.

Table 1.

Literature search terms.

Database Years included Search terms
Pubmed 1948 to August 2014

  1. Contraception Behavior/ or Contraception/ or Contraception, Barrier/ or contraception.mp.

  2. family planning services.mp. or Family Planning Services/

  3. 1 or 2

  4. substance related disorders.mp. or Substance-Related Disorders/

  5. substance abuse.mp. or Substance-Related Disorders/

  6. Alcohol Drinking/ or Behavior, Addictive/ or Alcoholism/

  7. Cocaine-Related Disorders/ or Heroin Dependence/ or addiction.mp.

  8. 4 or 5 or 6 or 7

  9. 3 and 8

  10. limit 9 to (english language and humans)
PsycINFO 1806 to August 2014

  1. family planning.mp. or Family Planning/

  2. birth control.mp. or Birth Control/

  3. contraception.mp.

  4. 1 or 2 or 3

  5. Drug Addiction/ or Addiction/ or Heroin Addiction/ or addiction.mp.

  6. alcoholism.mp. or Alcoholism/

  7. drug usage/ or drug self administration/ or needle sharing/

  8. Crack Cocaine/ or Cocaine/

  9. 5 or 6 or 7 or 8

  10. 4 and 9

  11. limit 10 to (human and english language)
Open in a new tab

All authors independently screened all abstracts for inclusion, and any disagreement led to the retrieval of the full text article. Full text articles were screened and extracted into premade data inclusion sheets by one author and verified by another author. All disagreements were resolved by consensus. PRISMA guidelines were followed for the review (Moher et al., 2009).

Meta analysis was precluded by heterogeneity in how contraceptive use and method choice were reported across studies. Medians and ranges are presented since the data were not normally distributed.

Results

Literature search

As shown in Fig. 1, our search strategy retrieved 580 abstracts of which 105 were selected for full text review (103 from the abstract list and 2 from the authors’ personal collections that were not identified in the search). Twenty-four articles were included in the final review. The most common reason for exclusion at both the abstract and full-text review stage was lack of data on contraceptive use.

Fig. 1.

Fig. 1

Open in a new tab

Search strategy.

Study publication date, location, population and population size, details of drug use, and comparison population (where applicable) are presented in Table 2. The 24 included studies assessed more than 5000 women with drug and alcohol problems and were published over a 40-year period between 1972 and 2012, with two reports in the first decade, two in the second decade, five in the third decade, and 15 in the fourth decade. The studies took place in seven different countries: United States (10), Australia (5), England (3), Canada (2), France (2), Finland (1), and Russia (1). Nine studies collected data from women in substance abuse treatment (including White et al., 1993 whose population also include needle exchange clients), seven from women using drugs or alcohol but not in drug treatment (including a subset of women from Toffol et al., 2011), and eight from other populations where >50% of the group reported drug and alcohol use (i.e., street-based female sex workers, chronically homeless women, incarcerated women, or women living with HIV or hepatitis C). Of note, fourteen studies included details about the specific type of drug(s) used, with more than half (51%) of the women in these studies reporting opioid use and much smaller percentages, alcohol and cocaine.

Table 2.

Population studied and date of publication.

Author (year) Country Study population N Drug or alcohol use in study population, N (%) Comparison population, N
Densen-Gerber et al. (1972) US Residential community treatment clients 57 57 (100) heroin
Eldred and Washington (1975) US Drug treatment clients 79 79 (100) heroin
Ralph and Spigner (1986) US Methadone maintenance clients 115 115 (100) opioid dependent 1982 National Survey of Family Growth, N not reported
Armstrong et al. (1991) US Drug treatment clients 599 599 (100) in drug treatment Family planning clinic visits in Philadelphia, PA in 1989, N = 55,223
 405 (68) injecting drug users
 154 (25) injected heroin in last 4 wks
 93 (16) injected cocaine in last 4 wks
Kouzi et al. (1992) US Injecting drug users with at least one sexual partner during the typical month 99 99 (100) injecting drug users
White et al. (1993) UK Drug treatment and syringe exchange clients 44 44 (100) injecting drug users
Morrison et al. (1995) UK Methadone maintenance clients 201 201 (100) opioid dependent
Dudish and Hatsukami (1996) US Non-treatment seeking women who responded to research recruitment 88 88 (100) crack cocaine
41 (47) alcohol
5 (6) marijuana
12 (14) alcohol and marijuana
3 (3) other
Harcourt et al. (2001) Australia Street-based sex workers 48 40 (83) injecting drug users Brothel-based sex workers attending Sydney Sexual Health Centre, N = 679
 35 (88) heroin
 2 (5) amphetamines
 2 (5) cocaine
 14 (29) alcohol
Gelberg et al. (2002) US Chronically homeless women of reproductive age 229 105 (46) — alcohol abuse (lifetime)
140 (61) — drug abuse (lifetime)
Banwell et al. (2003) Australia Women who self-reported being hepatitis C positive 462 217 (47) current injecting drug users 1995 Australian National Health Survey, N not reported
166 (36) past injecting drug users
Harding and Ritchie (2003) Australia Methadone maintenance clients 23 23 (100) opioid dependent
Harvey et al. (2003) US 14–30 years old injecting drug users and partners of male injecting drug users 94 69 (73) injecting drug users
 52 (56) heroin
 25 (27) cocaine
 38 (41) speed
 15 (16) speedball
79 (84) using drugs non-IV
 63 (67) marijuana
 36 (38) cocaine/crack
 37 (39) speed
 15 (16) heroin
 2 (2) speedball
Vidal-Trecan et al. (2003) France 25–34 years old injecting drug users in drug treatment 81 81 (100) injecting drug users 25–34 year old Parisian subsample of the National French Survey of Sexual Behaviour, N = 130
Weber et al. (2003) Canada Injecting drug users ≤50 years old 311 311 (100) injecting drug users
Carrieri et al. (2006) France Injecting drug users infected with HIV who also reported sex with an occasional partner of unknown HIV serostatus 90 90 (100) injecting drug users
Clarke et al. (2006) US Incarcerated women ≥18 years old 484 253 (52) heroin, other opiates, cocaine
165 (34) alcohol
Creighton et al. (2008) UK Street-based sex workers 25 24 (96) used any drug
 24 (96) crack
 7 (28) heroin
 3 (12) cocaine
 2 (8) benzodiazepines
Sharpe and Velasquez (2008) US Women 18–44 years old, of low-income, in urban jail, drug treatment, and healthcare facilities who reported using illicit drugs 2672 2000 (75) polydrug users Low-income women from same study populations who did not report using illicit drugs in the past 6 months, N = 1369
Olsen et al. (2009) Australia Hepatitis C positive women 109 85 (78) injecting drug users
Abdala et al. (2011) Russia 18–42 years old injecting drug users 78 78 (100) injecting drug users
Duff et al. (2011) Canada Street-based sex workers 211 103 (49%) heroin
71 (34%) cocaine
Toffol et al. (2011) Finland General population survey; alcohol abuse and dependence determined from Composite International Diagnostic Interview 2310 10 (b1) alcohol abusers
84 74 (3) alcohol dependent
Black et al. (2012) Australia Opioid-agonist maintenance clients 204 204 (100) opioid dependent
Open in a new tab

Prevalence of contraceptive use

Women with opioid and other substance use disorders

Table 3 outlines time frames of assessment and prevalence of contraceptive use. Prevalence of contraceptive use was assessed using seven different time frames: current (7), past/typical month (5), past 3 months (3), past 6 months (2), past year (2), lifetime (1), or not reported (4). As noted previously, prevalence of contraceptive use in each study was calculated for women with contraceptive need and details regarding these calculations are also presented in Table 3. Overall, prevalence of contraceptive use varied greatly between studies. The lowest reported prevalence was 6%, though this small (n = 18) study only reported the proportion of women who were using “reliable contraception” and noted that “condoms were occasionally used by all women, although unprotected intercourse was the norm” (Creighton et al., 2008). The highest reported contraceptive use prevalence was 77%. The remaining studies reported between 25% and 74% of women with opioid and other substance use disorders were using any contraceptive, with an overall median of 55%.

Table 3.

Contraceptive use prevalence.

Author (year) Women with opioid and other substance use disorders
Comparison population
Time frame of contraception assessment Comments
Contraceptive needs, N Contraceptive use, N (%) Contraceptive needs, N Contraceptive use, N (%)
Densen-Gerber et al. (1972) 50 26 (52) Lifetime 7 women did not provide contraception data
Eldred and Washington (1975) 73 23 (31) Not reported

  1. 6 women were intending to conceive

  2. Contraceptive use reported as either “all of the time” (23%) or “sometimes” (8%)
Ralph and Spigner (1986) 115 30 (26) Not calculable (49) Not reported N not reported for comparison group and thus could not be calculated
Armstrong et al. (1991) 368 143 (39) 42,411 41,265 (97) Last 1 month 231 women with substance use disorders problems and 12,812 comparison women were infertile, pregnant, or not sexually active
Kouzi et al. (1992) 99 45 (45)
White et al. (1993) 28 16 (57) Last 1 month

  1. 16 women were intending to conceive or were not sexually active

  2. Contraceptive use reported as “always” (46%) and “sometimes” (11%)
Morrison et al. (1995) 135 82 (61) Last 1 year 66 women were infertile, menopausal, pregnant, nor sexually active or intending to conceive
Dudish and Hatsukami (1996) 77 26 (34) Current 11 women were pregnant or had hysterectomies
Harcourt et al. (2001) 42 31 (74) 679 663 (98) Not reported 6 women were transgendered
Gelberg et al. (2002) 229 151 (66) Last 1 year Contraceptive use was reported as “always” (35%) or “sometimes” (31%).
Banwell et al. (2003) 399 137 (34) Not calculable (67) Current

  1. 63 women with substance use disorders were b18 and >49 and were excluded to match age range of comparison population

  2. N could not be calculated for comparison group as it was not reported
Harding and Ritchie (2003) 12 8 (67) Current 11 women were pregnant, menopausal, or not sexually active
Harvey et al. (2003) 91 62 (68) Last 1 month 3 women were not sexually active
Vidal-Trecan et al. (2003) 81 62 (77) 130 109 (84) Current
Weber et al. (2003) 262 162 (62) Current 49 women were infertile, pregnant, or not sexually active
Carrieri et al. (2006) 90 57 (63) Last 6 months
Clarke et al. (2006) 250 70 (28) Last 3 months

  1. 234 women were infertile, menopausal, intending to conceive, or not sexually active

  2. 72% reported “inconsistent” use of contraceptives, which included non-use in the last 3 months
Creighton et al. (2008) 18 1 (6) Last 3 months

  1. 7 women were pregnant

  2. Contraceptive use was defined as using “reliable” contraception.
Sharpe and Velasquez (2008) Not calculable (72) Not calculable (78) Not reported 1313 women across the polydrug using and comparison groups were sexually active, not pregnant, and not intending to conceive. Authors reported percentages of contraceptive use by such women in each group, but not the n for each group.
Olsen et al. (2009) 109 47 (43) Current
Abdala et al. (2011) 78 53 (68) Last 3 months
Duff et al. (2011) 196 88 (45) Last 6 months 15 women had a hysterectomy
Toffol et al. (2011) 84 21 (25) Current Contraceptive use only reported for women aged 30–54, and only for OCP and LNG-IUS
Black et al. (2012) 106 64 (60) Last 1 month 98 women were infertile, pregnant, trying to conceive, not sexually active, or provided incomplete data
Open in a new tab

Women with opioid and other substance use disorders vs. comparison populations

Six studies reported prevalence of contraceptive use for a comparison population (see Table 3). The comparison populations for three studies were national survey data from the countries where the studies were conducted and the remaining three studies used local comparison populations. All six of these studies reported lower prevalence of contraceptive use among the population with opioid and other substance use disorders than among their respective comparison populations, with a median difference of 25% (56% vs. 81%).

Contraceptive method choice

Women with opioid and other substance use disorders

Twenty of the 24 studies included in the review reported on the specific contraceptive methods used. The prevalence of different methods and details about their reporting are presented in Table 4.

Table 4.

Method choice among those using contraception.

Author (year) WHO Tier Women with opioid and other substance use disorders N(%) Comparison population N(%) Comments
Densen-Gerber et al. (1972) Tier 1 IUD: 7 (27) More than one method could be reported
Tier 2 OCP: 20 (77)
Tier 3 Condom: 2 (8)
Tier 3/4 Diaphragm/foam: 3 (12)
Ralph & Spigner (1986) Tier 1 TL: (24) TL: (30)

  1. More than one method could be reported in both populations

  2. N not calculable for comparison population
Tier 2 OCP: (32) OCP: (24)
Tier 3 Condom: (3)
Diaphragm: (18)		 Condom: (11)
Diaphragm: (7)
Tier 4 Foam: (3) Foam: (2)
Armstrong et al. (1991) Tier 1 TL: 160 (>100)
IUD: 9 (6)		 TL: 2,235 (5)
IUD: 446 (1)

  1. More than one method could be reported by women with substance use disorders, but only one method could be reported in comparison population

  2. Not all contraceptive users were sexually active; TL method prevalence not used in summary evaluations in results because prevalence >100%
Tier 2 OCP: 30 (21) OCP: 26,791 (65)
Tier 3 Condom: 89 (62)
Sponge: 6 (4)		 Condom: 9,506 (23)
Sponge: 453 (1)
Unk. Other: 23 (16) Other: 1834 (4)
Kouzi et al. (1992) Tier 2 OCP: 4 (9)

  1. More than one method could be reported

  2. Two women reported using contraception, but the exact methods were not specified More than one method could be reported
Tier 3 Condom: 39 (87)
White et al. (1993) Tier 1 TL: 1 (6)
IUD: 1 (6)
Tier 2 OCP/injection: 12 (75)
Tier 3 Condom: 4 (25)
Diaphragm: 0 (0)
Morrison et al. (1995) Tier 1 TL 13 (16)
IUD: 11 (13)

  1. More than one method could be reported

  2. Women with hysterectomies inextricable from those with a TL
Tier 2 Injection: 5 (6)
OCP: 18 (22)
Tier 3 Condom: 36 (44)
Natural family planning: 1(1)
Dudish and Hatsukami (1996) Tier 1 IUD: 0 (0)
Implant: 4 (15)		 Ns back-calculated
Tier 2 Injection: 15 (58)
OCP: 2 (8)
Tier 3 Condom/diaphragm: 5 (19)
Condom/diaphragm + foam: 0 (0)
Rhythm method: 0 (0)
Tier 4 Foam: 0 (0)
Harcourt et al. (2001) Tier 2 OCP: (12) OCP: (42)

  1. More than one method could be reported

  2. 26% of women with substance use disorders and 2% of comparison group reported “inadequate contraception”
Tier 3 Condom: (76) Condom: (56)
Banwell et al. (2003) Tier 1 TL/sterilization: 24 (17) TL/sterilization: (34) Sterilization included vasectomy
Tier 2 Injection: 10 (8)
OCP: 36 (26)		 Injection: not reported
OCP: (40)
Tier 1/3 IUD/condom/diaphragm: 67 (49) IUD/condom/diaphragm: (21)
Harding & Ritchie (2003) Tier 1/2 TL/injection/OCP: 5 (63)
Tier 3 Condom: 3 (38)
Harvey et al. (2003) Tier 3 Condom (any use): 43 (69)

  1. “Birth control” was not explicitly defined

  2. Ns back-calculated
Unk. Birth control (any use): 62 (100)
Vidal-Trecan et al. (2003) Tier 1 IUD: 5 (8) IUD: 20 (18)

  1. More than one method could be reported

  2. Tier 1/2 + 3 indicates dual use

  3. Other methods could include diaphragm, spermicides, withdrawal, and periodic abstinence.

  4. Ns back-calculated
Tier 2 OCP: 19 (31) OCP: 91 (83)
Tier 1/2+3 Condom + OCP/IUD: 2 (3) Condom + OCP/IUD: 1(1)
Tier 3 Condom: 52 (84) Condom: 13 (12)
Unk. Other methods: 12 (19) Other methods: 8 (7)
Weber et al. (2003) Tier 1 TL: 47 (30)
Tier 2 Injection: 10 (6)
OCP: 4 (3)
Tier 3 Condom: 101 (64)
Carrieri et al. (2006) Tier 2 OCP: 10 (17)

  1. “Other” category included withdrawal, IUD, spermicide, diaphragm, and vaginal douching

  2. Tier 2 + 3 indicates dual use

  3. Ns back-calculated
Tier 2+3 Condom + OCP: 4 (7)
Tier 3 Condom: 37 (65)
Tier 1/3/4 Other: 6 (11)
Clarke et al. (2006) Tier 3 Condom (consistent use): 50 (71)

  1. “Birth control” was not explicitly defined

  2. Ns back-calculated
Unk. Birth control (consistent use): 70 (100)
Olsen et al. (2009) Tier 1 TL: 8 (17)
Partner’s vasectomy: 3 (6)
IUD: 1 (2)		 Tier 1/2 + 3 indicates dual use
Tier 1/2 Implant/injection: 11 (23)
Tier 2 OCP: 6 (13)
Emergency Contraception: 0 (0)
Tier 1/2+3 Condom + pill/implant: 7 (15)
Tier 3 Condom: 10 (21)
Diaphragm: 1 (2)
Abdala et al. (2011) Tier 1 IUD: 0 (0)
Tier 2 OCP: 0 (0)
Tier 3 Condom: 40 (75)
Tier 4 Spermicide: 1 (2)
Douching/withdrawal: 12 (23)
Duff et al. (2011) Tier 1 TL: 35 (40)
IUD: 3 (3)		 139 reported using condoms only with clients
Tier 2 Injection: 18 (21)
OCP: 2 (2)
Tier 3 Condom: 30 (34)
Female condom: 0 (0)
Toffol et al. (2011) Tier 1 IUD: 6 (29) IUD was specifically levonorgestrel-releasing system
Tier 2 OCP: 15 (71)
Black et al. (2012) Tier 1/2 TL/IUD/implant//injection: 24 (38) One woman reported using contraception, but did not specify a method
Tier 2 OCP: 10 (16)
Tier 3 Condom: 29 (46)
Open in a new tab

Notes: WHO Tier = World Health Organization’s tiers of contraceptive effectiveness (WHO, 2007). Tier 1 = very effective (<1% of women using one of these methods will become pregnant in the first year of typical use), Tier 2 = effective (1–9% pregnancy rates), Tier 3 = moderately effective (10–25% pregnancy rates), and Tier 4 = less effective (26–32% pregnancy rates). TL = tubal ligation, IUD = intrauterine device, Injection = depot injection, OCP = oral contraceptive pills, unk. = unknown.

Very effective methods

IUD use ranged from 2%–29% with a median of 7% across eight studies. Use of tubal ligation ranged from 6%–40% with a median of 17% across seven studies. One study reported 15% of their contraceptive-using population used implants.

Effective methods

Oral contraceptive pill use ranged from 2%–77% with a median of 17% across 15 studies. Injection use ranged from 6%–58% with a median of 8% across five studies. No studies assessed vaginal ring or transdermal patch use.

Moderately effective methods

Seventeen studies presented data on condom use. Among these studies, condom use ranged from 3%–87% with a median of 62%. Interestingly, studies containing predominantly injecting drug users reported a median condom use prevalence of 65% compared to 44% in studies containing alcohol and/or non-injecting drug users. Diaphragm use ranged from 2–18% across two studies with a median of 10%. Sponge and natural family planning were reported by one study each, with very low percentages of use (≤4%).

Less effective methods

Foam was reported in one study, with a very low percentage of use (3%). Another study reported that 23% of contraceptive users were relying on vaginal douching or withdrawal as their primary contraceptive method.

Method combination

Three studies reported dual use, that is condoms in addition to another contraceptive method, with prevalence ranging from 3%–15% and a median of 7%.

Women with opioid and other substance use disorders vs. comparison populations

Five studies reported prevalence for contraceptive method choice in their target and comparison populations (see Table 4).

Very effective methods

Only two studies provided comparisons of IUD use and reported discrepant results. Armstrong et al. (1991) reported higher prevalence of IUD in their study population (6% vs. 1%), while Vidal-Trecan et al. (2003) reported lower prevalence of IUD use in their study population (8% vs. 18%). Tubal ligation was also less common among women with opioid and other substance use disorders in two studies (approximately 21% vs. 32%). None of the studies with comparison populations reported on use of implants.

Effective methods

Four of the five studies reported lower prevalence of oral contraceptive use among their study populations, with a median difference of 30% (24% vs. 54%). Only one of the five studies (Ralph and Spigner, 1986) reported higher prevalence of oral contraceptive use among their study population versus their comparison population (32% vs. 24%).

Moderately effective methods

Three of the four studies that reported on condom use separately in the study and comparison populations noted higher prevalence of condom use among women with opioid and other substance use disorders, with a median difference of 53% (76% vs. 23%). Ralph and Spigner (1986) reported that only 3% of women in a methadone maintenance program used condoms compared to 11% of women of similar economic status within a nationally representative sample, while Banwell et al. (2003) did not report condom use separately, but did state that 49% of their population used either condoms, IUDs, or diaphragms compared to 21% of their comparison population.

Less effective methods or method combination

None of the studies with comparison populations reported on use of less effective methods or method combinations.

Discussion

This manuscript describes results of a systematic review of the published literature to estimate overall contraceptive use and examine method choice among women with opioid and other substance use disorders and is the first effort to do so, to our knowledge. While the studies included were published over an approximately 40-year period, the observation that opioids were by far the most commonly used drug make the results especially relevant to the present day, with opioid use at epidemic levels (Maxwell, 2011) and much attention focused on the adverse consequences and costs of infants exposed in utero (e.g., Patrick et al., 2012). Recent data also clearly demonstrate that nearly all of these pregnancies are unintended (Black et al., 2012; Heil et al., 2011; Jones et al., 2011), underscoring the need to better understand contraceptive use and method choice among women with opioid and other substance use disorders.

Results of this review suggest that only about half of women with opioid and other substance use disorders use contraception. In the subset of comparison studies reviewed, contraceptive use prevalence was 25% lower among women with opioid and other substance use disorders suggesting unmet need. Perhaps more alarmingly, the data on contraceptive method choice indicate that condoms were by far the most commonly used method (approximately 62%), while prevalence of the very effective methods (tubal ligation, implants, and IUDs) was typically much lower (approximately 8%). Injecting drug users appear to rely on condoms more often than non-injecting drug users, likely as a result of public health efforts to reduce sexually transmitted infections (STIs) in this population. With continued high prevalence of HIV among drug users and a narrowing of the disparate HIV infection rates between intravenous and non-intravenous drug users (Des Jarlais et al., 2007), promoting condom use among drug users remains important for preventing STIs. However, the distressingly high rates of unintended pregnancy among drug-using women underscore the urgent need to promote the use of more effective contraceptives.

Contraceptive method choice counseling usually takes risk for STIs into consideration and often recommends condoms for women at risk for both unintended pregnancy and STIs. For example, the first recommendation made for women at risk of unintended pregnancy and STIs in the WHO contraceptive method decision-making tool (2005) is condoms, followed by combining condom use with another contraceptive method (i.e., dual use). Results from this review suggest that dual use is uncommon, with only about 7% of women endorsing dual protection in the three studies that reported it. One way to increase dual protection may be to separate contraceptive recommendations from recommendations about condom use to prevent STIs. That is, a woman would be counseled to choose a contraceptive method based solely on her contraceptive needs and regardless of her STI risk. Then, if she is at risk for STIs, she would be encouraged to use condoms in addition to the contraceptive method. This approach may increase the chances that a woman at risk for unintended pregnancy and STIs will choose one of the very effective or effective contraceptive methods, rather than defaulting to moderately effective condom use.

Regarding the very effective methods, tubal ligation has long been the gold standard female contraceptive, but the permanence of the method combined with the vulnerability of the population has raised concerns about promoting tubal ligation among women with opioid and other substance use disorders (cf., Lucke and Hall, 2012 and associated commentaries). Fortunately, implants and IUDs have equally low, if not lower, pregnancy rates (WHO, 2007), but their contraceptive actions are reversible with a rapid return to fertility upon removal (Zieman and Hatcher, 2013), significantly reducing ethical concerns. More generally, the benefits of using LARC methods are so substantial that the American College of Obstetricians and Gynecologists now recommends LARCs as first-line contraceptives for all women (ACOG, 2009), including other populations at increased risk for unintended pregnancy like adolescents (ACOG, 2012).

While the benefits of LARCs are well established, they are a more difficult method of contraception to obtain. Implants and IUDs can cost $650 or more depending on insurance coverage and must be inserted by trained providers (Zieman and Hatcher, 2013). In an effort to address the persistent 50% unintended pregnancy rate in the general population (Mohllajee et al., 2007), a recent large study set out to promote LARCs among a wide range of women at risk for unintended pregnancy in part by providing the methods free of charge. The authors reported that 75% of participants chose a LARC and that rates of abortion, repeat abortion and teen births in the catchment area decreased significantly in subsequent years (Peipert et al., 2012). Regarding insertion by trained providers, women who enter substance abuse treatment are in frequent contact with health care professionals, suggesting that providing family planning services as part of treatment may be especially efficacious with this population. Observational studies that offered free contraceptive supplies and provided family planning services in treatment settings have reported promising results (Armstrong et al., 1991; Elko and Jansson, 2011). In addition, results of a small randomized trial of a similar approach conducted by some of the authors of this review pointed to increased use of more effective contraceptives among opioid-maintained women at risk for unintended pregnancy (Heil et al., 2013; in preparation).

The availability of different contraceptive methods has been shown to vary substantially over time and by country (Ross and Stover, 2013). Time was undoubtedly part of the reason that none of the studies in the review reported on vaginal ring or transdermal patch use, as both of these methods are relatively new to the market (2001 for ring, 2002 for patch). As another example, the availability of implants has differed at times in the US due to changes in manufacturers and formulations. Wyeth Pharmaceuticals produced Norplant from approximately 1990–2002 and Merck and Co. produced Implanon from approximately 2006–2012 and then transitioned to Nexplanon from 2011 through the present day. Method availability and other factors, like cost and public opinion, also likely contribute to variability in method choice. For example, rates of IUD use are much lower in the US as compared to most European countries and this is often attributed in part to significant problems with one brand of IUD, the Dalkon Shield, in the US in the 1970s (Sonfield, 2007). All of these aspects likely contributed to the wide variability observed in the prevalence of contraceptive use and in the use of specific methods across the studies, but point to a benefit of systematic reviews, namely that looking across a large number of studies may help bring out larger patterns that are difficult to glean from individual studies. This approach may be especially important in the study of contraceptive use and method choice.

The results of this review also suggest that there is increasing interest in assessing contraceptive use prevalence and method choice among women with opioid and other drug problems, with two papers published in first decade of the review, but 15 by the fourth decade. Future assessments of contraceptive practices would benefit from standardizing how contraceptive use prevalence is assessed and calculated. First, characteristics of the population must be considered when calculating contraceptive prevalence. Generally, contraceptive prevalence should be calculated only among women who are at risk of becoming pregnant and are not actively trying to conceive. Second, the time frame for assessing contraceptive method use should be carefully considered and clearly reported to help prevent confusion over methods a woman has ever used vs. methods she is currently using. Finally, use of each specific contraceptive method should be reported separately with Ns as well as percentages. If grouping contraceptive methods is necessary, the groupings should be made with regard to the effectiveness of the methods. Dual method use should also be explicitly described and presented separately from use of each individual contraceptive method. More attention to these sorts of details will strengthen this literature and may permit more rigorous analysis (e.g., meta analysis) in the future.

While more detailed information about contraceptive use prevalence and method choice will help refine estimates of both, the results of this review already strongly suggest that women with opioid and other substance use disorders have unmet contraceptive need and that those who do use contraception rely too heavily on methods that are only moderately effective. Providing family planning services, including promotion of more effective methods of contraception, as part of substance abuse treatment has potential to improve reproductive health and to address the problem of unintended pregnancy in this vulnerable population.

Acknowledgments

Preparation of this manuscript was supported by research grants R01 DA036670 and T32 DA07242 from the National Institute on Drug Abuse. The study sponsor had no involvement in the preparation of the manuscript.

Footnotes

Conflict of interest statement

The authors declare that there is no conflict of interest.

References

  1. Abdala N, Kershaw T, Krasnoselskikh TV, Kozlov AP. Contraception use and unplanned pregnancies among injection drug-using women in St Petersburg, Russia. J Fam Plann Reprod Health Care. 2011;37(3):158–164. doi: 10.1136/jfprhc-2011-0079. http://dx.doi.org/10.1136/jfprhc-2011-0079. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. American College of Obstetricians and Gynecologists (ACOG) Increasing use of contraceptive implants and intrauterine devices to reduce unintended pregnancy. ACOG Committee Opinion No 450. Obstet Gynecol. 2009;114:1434–1438. doi: 10.1097/AOG.0b013e3181c6f965. [DOI] [PubMed] [Google Scholar]
  3. American College of Obstetricians and Gynecologists (ACOG) Adolescents and long-acting reversible contraception: implants and intrauterine devices. ACOG Committee Opinion No 539. Obstet Gynecol. 2012;120:983–988. doi: 10.1097/AOG.0b013e3182723b7d. [DOI] [PubMed] [Google Scholar]
  4. Ariely D, Loewenstein G. The heat of the moment: the effect of sexual arousal on sexual decision making. J Behav Decis Mak. 2006;19(2):87–98. http://dx.doi.org/10.1002/bdm.501. [Google Scholar]
  5. Armstrong KA, Kenen R, Samost L. Barriers to family planning services among patients in drug treatment programs. Fam Plan Perspect. 1991;23(6):264–266. 270–271. [PubMed] [Google Scholar]
  6. Banwell C, Bammer G, Main N, Gifford SM, O’Brien M. Disturbingly low levels of contraception among women living with hepatitis C. Aust N Z J Public Health. 2003;27(6):620–626. doi: 10.1111/j.1467-842x.2003.tb00609.x. [DOI] [PubMed] [Google Scholar]
  7. Black KI, Stephens C, Haber PS, Lintzeris N. Unplanned pregnancy and contraceptive use in women attending drug treatment services. Aust N Z J Obstet Gynaecol. 2012;52(2):146–150. doi: 10.1111/j.1479-828X.2012.01413.x. http://dx.doi.org/10.1111/j.1479-828X.2012.01413.x. [DOI] [PubMed] [Google Scholar]
  8. Carrieri MP, Rey D, Serraino D, et al. Oral contraception and unprotected sex with occasional partners of women HIV-infected through injection drug use. AIDS Care. 2006;18(7):795–800. doi: 10.1080/09540120500431584. http://dx.doi.org/10.1080/09540120500431584. [DOI] [PubMed] [Google Scholar]
  9. Clarke JG, Hebert MR, Rosengard C, Rose JS, DaSilva KM, Stein MD. Reproductive health care and family planning needs among incarcerated women. Am J Public Health. 2006;96(5):834–839. doi: 10.2105/AJPH.2004.060236. http://dx.doi.org/10.2105/AJPH.2004.060236. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Creighton S, Tariq S, Perry G. Sexually transmitted infections among UK street-based sex workers. Sex Transm Infect. 2008;84(1):32–33. doi: 10.1136/sti.2007.026443. http://dx.doi.org/10.1136/sti.2007.026443. [DOI] [PubMed] [Google Scholar]
  11. Degenhardt L, Whiteford HA, Ferrari AJ, et al. Global burden of disease attributable to illicit drug use and dependence: findings from the Global Burden of Disease Study 2010. Lancet. 2013;382(9904):1564–1574. doi: 10.1016/S0140-6736(13)61530-5. http://dx.doi.org/10.1016/S0140-6736(13)61530-5. [DOI] [PubMed] [Google Scholar]
  12. Densen-Gerber J, Wiener M, Hochstedler R. Sexual behavior, abortion, and birth control in heroin addicts: legal and psychiatric considerations. Contemp Drug Probl. 1972;1(4):783–793. [Google Scholar]
  13. Des Jarlais DC, Arasteh K, Perlis T, et al. Convergence of HIV seroprevalence among injecting and non-injecting drug users in New York City. AIDS. 2007;21(2):231–235. doi: 10.1097/QAD.0b013e3280114a15. http://dx.doi.org/10.1097/QAD.0b013e3280114a15. [DOI] [PubMed] [Google Scholar]
  14. Dudish SA, Hatsukami DK. Gender differences in crack users who are research volunteers. Drug Alcohol Depend. 1996;42(1):55–63. doi: 10.1016/0376-8716(96)01266-5. [DOI] [PubMed] [Google Scholar]
  15. Duff P, Shoveller J, Zhang R, Alexson D, Montaner JSG, Shannon K. High lifetime pregnancy and low contraceptive usage among sex workers who use drugs — an unmet reproductive health need. BMC Pregnancy Childbirth. 2011;11:61. doi: 10.1186/1471-2393-11-61. http://dx.doi.org/10.1186/1471-2393-11-61. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Eldred CA, Washington MN. Female heroin addicts in a city treatment program: the forgotten minority. Psychiatry. 1975;38(1):75–85. doi: 10.1080/00332747.1975.11023836. [DOI] [PubMed] [Google Scholar]
  17. Elko A, Jansson LM. Contraception in drug-dependent women: a novel approach. Soc Work Ment Health. 2011;9:445–455. [Google Scholar]
  18. Gelberg L, Leake B, Lu MC, et al. Chronically homeless women’s perceived deterrents to contraception. Perspect Sex Reprod Health. 2002;34(6):278–285. [PubMed] [Google Scholar]
  19. Harcourt C, van Beek I, Heslop J, McMahon M, Donovan B. The health and welfare needs of female and transgender street sex workers in New South Wales. Aust N Z J Public Health. 2001;25(1):84–89. doi: 10.1111/j.1467-842x.2001.tb00556.x. [DOI] [PubMed] [Google Scholar]
  20. Harding C, Ritchie J. Contraceptive practice of women with opiate addiction in a rural centre. Aust J Rural Health. 2003;11(1):2–6. doi: 10.1046/j.1440-1584.2003.00477.x. http://dx.doi.org/10.1046/j.1440-1584.2003.00477.x. [DOI] [PubMed] [Google Scholar]
  21. Harvey SM, Bird ST, De Rosa CJ, Montgomery SB, Rohrbach LA. Sexual decision making and safer sex behavior among young female injection drug users and female partners of IDUs. J Sex Res. 2003;40(1):50–60. doi: 10.1080/00224490309552166. http://dx.doi.org/10.1080/00224490309552166. [DOI] [PubMed] [Google Scholar]
  22. Heil SH, Jones HE, Arria A, et al. Unintended pregnancy in opioid-abusing women. J Subst Abus Treat. 2011;40(2):199–202. doi: 10.1016/j.jsat.2010.08.011. http://dx.doi.org/10.1016/j.jsat.2010.08.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Heil SH, Hand DJ, Sigmon SC, Meyer MC, Higgins ST. Improving effective contraceptive use among opioid-maintained women. Poster Presented at the Annual Meeting of the College on Problems of Drug Dependence; San Diego, CA. 2013. [Google Scholar]
  24. Jones HE, Berkman ND, Kline TL, et al. Initial feasibility of a woman-focused intervention for pregnant African-American women. Int J Pediatr. 2011;2011:389285. doi: 10.1155/2011/389285. http://dx.doi.org/10.1155/2011/389285. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Kouzi AC, Des Jarlais DC, Tross S, Abdul-Quader A, Friedman SR. Contraceptive behavior among intravenous drug users at risk for AIDS. Psychol Addict Behav. 1992;6(2):135–139. [Google Scholar]
  26. Lucke JC, Hall WD. Under what conditions is it ethical to offer incentives to encourage drug-using women to use long-acting forms of contraception? Addiction. 2012;107(6):1036–1041. doi: 10.1111/j.1360-0443.2011.03699.x. http://dx.doi.org/10.1111/j.1360-0443.2011.03699.x. [DOI] [PubMed] [Google Scholar]
  27. Maxwell JC. The prescription drug epidemic in the United States: a perfect storm. Drug Alcohol Rev. 2011 May;30(3):264–270. doi: 10.1111/j.1465-3362.2011.00291.x. [DOI] [PubMed] [Google Scholar]
  28. Moher D, Liberati A, Tetzlaff J, Altman DG PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med. 2009;6(7):e1000097. doi: 10.1371/journal.pmed.1000097. http://dx.doi.org/10.1371/journal.pmed.1000097. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Mohllajee AP, Curtis KM, Morrow B, Marchbanks PA. Pregnancy intention and its relationship to birth and maternal outcomes. Obstet Gynecol. 2007;109(3):678–686. doi: 10.1097/01.AOG.0000255666.78427.c5. http://dx.doi.org/10.1097/01.AOG.0000255666.78427.c5. [DOI] [PubMed] [Google Scholar]
  30. Morrison CL, Ruben SM, Beeching NJ. Female sexual health problems in a drug dependency unit. Int J STD AIDS. 1995;6(3):201–203. doi: 10.1177/095646249500600311. [DOI] [PubMed] [Google Scholar]
  31. Olsen A, Banwell C, Dance P. Internal or infernal devices: experiences of contraception among Australian women living with hepatitis C. Health Care Women Int. 2009;30(6):456–474. doi: 10.1080/07399330902797591. http://dx.doi.org/10.1080/07399330902797591. [DOI] [PubMed] [Google Scholar]
  32. Patrick SW, Schumacher RE, Benneyworth BD, Krans EE, McAllister JM, Davis MM. Neonatal abstinence syndrome and associated health care expenditures: United States, 2000–2009. JAMA. 2012;307(18):1934–1940. doi: 10.1001/jama.2012.3951. http://dx.doi.org/10.1001/jama.2012.3951. [DOI] [PubMed] [Google Scholar]
  33. Peipert JF, Madden T, Allsworth JE, Secura GM. Preventing unintended pregnancies by providing no-cost contraception. Obstet Gynecol. 2012;120(6):1291–1297. doi: 10.1097/aog.0b013e318273eb56. http://dx.doi.org/10.1097/AOG.0b013e318273eb56. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ralph N, Spigner C. Contraceptive practices among female heroin addicts. Am J Public Health. 1986;76(8):1016–1017. doi: 10.2105/ajph.76.8.1016. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Rehm J, Mathers C, Popova S, Thavorncharoensap M, Teerawattananon Y, Patra J. Global burden of disease and injury and economic cost attributable to alcohol use and alcohol-use disorders. Lancet. 2009;373(9682):2223–2233. doi: 10.1016/S0140-6736(09)60746-7. http://dx.doi.org/10.1016/S0140-6736(09)60746-7. [DOI] [PubMed] [Google Scholar]
  36. Ross J, Stover J. Use of modern contraception increases when more methods become available: analysis of evidence from 1982–2009. Glob Health Sci Pract. 2013;1(2):203–212. doi: 10.9745/GHSP-D-13-00010. http://dx.doi.org/10.9745/GHSP-D-13-00010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Sharpe TT, Velasquez MM. Risk of alcohol-exposed pregnancies among low-income, illicit drug-using women. J Women’s Health. 2008;17(8):1339–1344. doi: 10.1089/jwh.2008.0828. http://dx.doi.org/10.1089/jwh.2008.0828. [DOI] [PubMed] [Google Scholar]
  38. Sonfield A. Popularity disparity: attitudes about the IUD in Europe and the Unites States. Guttmacher Policy Rev. 2007;10(4):19–24. [Google Scholar]
  39. Substance Abuse and Mental Health Services Administration. Results From the 2013 National Survey on Drug Use and Health: Summary of National Findings. Substance Abuse and Mental Health Services Administration; Rockville, MD: 2014. NSDUH Series H-48, HHS Publication No. (SMA) 14-4863. [Google Scholar]
  40. Than LC, Honein MA, Watkins ML, Yoon PW, Daniel KL, Correa A. Intent to become pregnant as a predictor of exposures during pregnancy: is there a relation? J Reprod Med. 2005;50(6):389–396. [PubMed] [Google Scholar]
  41. Toffol E, Heikinheimo O, Koponen P, Luoto R, Partonen T. Hormonal contraception and mental health: results of a population-based study. Hum Reprod. 2011;26(11):3085–3093. doi: 10.1093/humrep/der269. http://dx.doi.org/10.1093/humrep/der269. [DOI] [PubMed] [Google Scholar]
  42. Vidal-Trecan G, Warszawski J, Coste J, et al. Contraceptive practices of non-HIV-seropositive injecting drug users. Eur J Epidemiol. 2003;18(9):863–869. doi: 10.1023/a:1025643421939. [DOI] [PubMed] [Google Scholar]
  43. Weber AE, Tyndall MW, Spittal PM, et al. High pregnancy rates and reproductive health indicators among female injection-drug users in Vancouver, Canada. Eur J Contracept Reprod Health Care. 2003;8(1):52–58. [PubMed] [Google Scholar]
  44. Wellings K, Jones KG, Mercer CH, et al. The prevalence of unplanned pregnancy and associated factors in Britain: findings from the third National Survey of Sexual Attitudes and Lifestyles (Natsal-3) Lancet. 2013;382(9907):1807–1816. doi: 10.1016/S0140-6736(13)62071-1. http://dx.doi.org/10.1016/S0140-6736(13)62071-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. White D, Phillips K, Mulleady G, Cupitt C. Sexual issues and condom use among injecting drug users. AIDS Care. 1993;5(4):427–437. doi: 10.1080/09540129308258012. http://dx.doi.org/10.1080/09540129308258012. [DOI] [PubMed] [Google Scholar]
  46. World Health Organization (WHO) Family Planning: A Global Handbook for Providers. CCP and WHO; Baltimore, MD and Geneva: 2007. World Health Organization Department of Reproductive Health and Research (WHO/RHR) and Johns Hopkins Bloomberg School of Public Health/ Center for Communication Programs (CCP), INFO Project. [Google Scholar]
  47. Zieman M, Hatcher RA. Managing Contraception. Bridging the Gap Foundation; Tiger, GA: 2013. [Google Scholar]

RESOURCES