Reproductive Health Knowledge,Attitudes,and Behaviors Among Women Enrolled in Medication-Assisted Treatment for Opioid Use Disorder

Abstract

Background:

The lifetime prevalence of unintentional pregnancy among women enrolled in medically assisted treatment (MAT) for opioid use disorder (OUD) has been reported to exceed 80%. Consistent voluntary contraception use by women enrolled in MAT programs can reduce their risk of unintended pregnancies and increase their opportunity to plan the family size they want, yet little is understood about past and current contraception use or associated barriers and facilitators for this population of women.

Materials and Methods:

Women enrolled in treatment for OUD at two MAT clinics in East Tennessee were invited to participate in an anonymous survey about contraception. We describe the sociodemographic and health characteristics of the women (N = 287; estimated response rate of 81.0%), report on their contraception knowledge and attitudes, and assess how these characteristics are associated with current contraception use.

Results:

Most of the 287 women had been sexually active in the past 12 months (88%), were pregnant at least once (98%), and were insured (58%). Women having undergone bilateral tubal ligation had a greater average number of pregnancies (including those while on birth control), lower mean levels of depressed mood, and higher mean level of agreement that avoiding pregnancy now was important and that pregnant women should not use painkillers, compared to the women not using contraception.

Conclusions:

Strategies to ensure all women enrolled in MAT for OUD have knowledge of effective contraception and affordable, quality contraception care that is responsive to their complicated and oftentimes traumatic histories are necessary to enhance the women's reproductive health.

Introduction

Very little is understood about facilitators or barriers to effective contraceptive use among women receiving medically assisted treatment (MAT) for opioid use disorders (OUDs). Along with behavioral therapy, women in MAT receive daily doses of opioid agonist pharmacotherapies to prevent withdrawal symptoms. Barriers to effective contraception use in the broader population of women with OUD include misinformation, mistrust of healthcare systems, fear of child protection services, cost, and denial or shame about their substance use;^1

–4 it is not known whether these same barriers apply to women receiving MAT.^5,6 The few studies that have asked women in MAT about barriers to contraceptive use focused their questions around cost, side effects, and intimate partner violence.^4,7 Although women endorsed cost and side effects as barriers, they did not endorse intimate partner violence, and large percentages of women checked “other reasons” or did not answer the question, suggesting they were presented with an incomplete list of potential barriers.

Two factors make it vital to better understand the barriers to and facilitators of effective contraceptive use for women enrolled in MAT for OUD. First, women enrolled in MAT report a high incidence and lifetime prevalence of unintended pregnancy (75% and over 80%, respectively).^4,7,8 This fig ure far exceeds the 2011 rate of 45% for the general U.S. population.⁹ We located two studies of current contraception use among women enrolled in MAT, one in the United States⁴ and the other in Australia.⁸ Respectively, the percentage of women not wanting to conceive, who used any form of contraception (including sterilization), was 66%⁴ and 60%,⁸ compared to the 2014 U.S. national rate of 89%.¹⁰ In the U.S. study, 43% of women using contraception reported a bilateral tubal ligation (BTL), 23% used an intrauterine device (IUD), and 4% had a progestin implant.⁴ In the Australian study, 37% of women using contraception used one of these three methods, and the rest used a user-dependent method.⁸ This low use rate of IUDs and progestin implants, both long-acting reversible contraceptives (LARCs), is notable given that women using user-dependent methods (e.g., pills, patches, and rings) are up to 20 times more likely to have an unplanned pregnancy than those who use a LARC.¹¹ Together, these findings indicate a disparity in LARC access for women receiving MAT.

Second, chronic prenatal use of opioids can have adverse effects on the newborn. MAT is the recommended treatment option for pregnant women with OUD; it regularizes prenatal dosing of opioids, helps avoid relapse, and improves compliance with prenatal care.¹² Nonetheless, MAT is associated with low birth weight, smaller neonate head circumference, and neonatal abstinence syndrome (NAS), a constellation of symptoms caused by disturbances in gastrointestinal, autonomic, and central nervous systems of infants as they withdraw from opioid dependence.^12,13 In 2015, the year of data collection, MAT was the primary source of prenatal opioid exposure for 40% of the 410 newborns diagnosed with NAS in Tennessee.¹⁴ Between January 1 and December 16, 2017, MAT was the primary source of opioid exposure for 69% of the 681 infants diagnosed with NAS.¹⁵

This study's data collection occurred in a unique social and political context regarding both OUD and contraception. Southern Appalachia, including East Tennessee, has a long history of substance misuse and abuse, including alcohol, tobacco, and nonmedical use of prescription drugs, due, in part, to widespread economic hardship and behavior modeled across generations.¹⁶ East Tennessee is also part of the Bible Belt, which has been associated with restricting women's reproductive choices.¹⁷

In April 2014, Tennessee enacted Public Chapter 820, a law that allowed women to be prosecuted for illegal narcotic use if their child was born dependent on a narcotic or demonstrably harmed by illegal prenatal narcotic use.¹⁸ Simultaneously, Tennessee did not adopt the Medicaid expansion component of the Affordable Care Act (ACA), which would have increased access to reproductive health services and drug treatment reimbursement for low-income residents. Instead, Tennessee's public insurance in 2015 covered these services only for low-income women of childbearing age who were pregnant and in first three postpartum months, or the caregiver of a family member.

To better understand the barriers and facilitators of effective contraception use, including sterilization, for women receiving MAT for OUD, we analyzed survey responses of 287 women of reproductive age (18–45) receiving MAT from the two East Tennessee MAT clinics. We describe the women's demographic and health characteristics; report on their contraception knowledge, attitudes, and behavior; and assess how these variables are associated with current contraception use.

Materials and Methods

Data collection

Survey data were collected from 291 respondents at the two East Tennessee clinics that specialize in treating OUD and offer maintenance treatment of methadone or buprenorphine. All women, aged 18–45, were eligible for the study if they were enrolled in one of the clinics between January 12 and February 6, 2015. The Institutional Review Board of the University of Tennessee, Knoxville, approved the study.

Women were recruited through posters and flyers in the MAT clinics and by in-person invitations from clinic and research staff. Recruitment took place Monday through Friday, from 6 to 10 a.m. Trained female research staff verbally reviewed the active consent form with all eligible and interested women in a private setting and answered all questions about the research. Women who chose to participate then signed the consent and completed the 98-item anonymous electronic or paper survey (25%; n = 71), as preferred by each woman. Research staff were available to answer any questions during the process. Upon survey completion, each woman received a $25 gift card to a local retailer. A total of 291 women participated, with 4 receiving a gift card on 2 occasions. These four repeat cases were not identifiable and may or may not have been included in the analysis. Four women responded to less than half of the survey questions and were excluded from the analyses; thus, the final analysis reflects 287 women. One of the two clinics provided the number of eligible women, indicating an 81% participation rate for this particular clinic.

Measures

When possible, survey items were drawn from previous studies to maximize comparability across populations (Tables 1 –4). This section highlights measures that required calculation or recoding.

Table 1.

Participants' Sociodemographic Characteristics (N = 287)

	n (%)/mean ± SD
Age, years	33.9 ± 6.2
18–25	25 (8.7)
26–35	132 (46.0)
36–45	105 (36.6)
Not reported	25 (8.7)
Social situation
Married	108 (37.6)
Living with a partner	56 (19.5)
Separated, divorced, or widowed	46 (16.0)
Single	62 (21.6)
No response	15 (5.2)
Number of living children	2.3 ± 6.2
No births	52 (18.1)
None	2 (.9)
One or more	217 (75.6)
Not reported	16 (5.6)
Child ever in Department of Children's Services' custody
Yes	37 (12.9)
No, has children	181 (63.1)
No, no births	52 (18.1)
Not reported	17 (5.9)
Education
Less than high school	59 (20.6)
High school diploma or equivalent	110 (38.3)
Postsecondary, no degree	73 (25.4)
Certificate, license, or associate's degree	13 (4.5)
4-year college degree or more	17 (5.9)
Not reported	15 (5.2)
Employment
Employed	133 (46.3)
Unemployed	32 (11.1)
Not in labor force	50 (17.4)
Not reported	72 (25.1)
Annual household income
$0–$10,000	72 (25.1)
$10,001–$20,000	70 (24.4)
$20,001–$30,000	56 (19.5)
$30,001–$50,000	39 (13.6)
Over $50,000	15 (5.2)
Do not know	17 (5.9)
Not reported	18 (6.3)
Difficulties paying rent, mortgage, or utilities in the past year²²
Yes	175 (61.0)
No	107 (37.3)
No response	5 (1.7)
Moved in with friend or family in last year because had no other choice²³
Yes	95 (33.1)
No	185 (64.5)
No response	7 (2.4)
Insurance
Private insurance	44 (15.3)
Public	121 (42.2)
No insurance	118 (41.1)
Not reported	4 (1.4)

SD, standard deviation.

Table 2.

Participants' Self-Reported Health Status and Substance Use (N = 287)

	n (%)/mean ± SD
Would you say in general your health is…²⁴ (range: 1–5, n = 277)	2.9 ± 0.9
Poor	17 (5.9)
Fair	78 (27.2)
Good	121 (42.4)
Very good	51 (17.8)
Excellent	10 (3.5)
Not reported	10 (3.5)
Depressed mood¹⁹ (range 1–4; n = 279)	2.2 ± 0.7
Ever had nonconsensual sexual intercourse
Yes	93 (32.4)
No	188 (65.5)
Not reported	6 (2.1)
Experienced physical abuse (hit, kicked, slapped, etc.) in last year
Yes	49 (17.1)
No	230 (80.1)
Not reported	8 (2.8)
Physically abused (hit, kicked, slapped, etc.) by romantic partner in the last year²⁵
Yes	18 (6.3)
No	262 (91.3)
Not reported	7 (2.4)
Source of first opioid use
Your own prescription	121 (42.2)
A friend who had the drug	101 (35.2)
A family member's prescription	36 (12.5)
Street contact who sold the drug	24 (8.4)
Not reported	5 (1.7)
Mean age of first opioid use (range 12–43; N = 268)	20.3 ± 5.4
Years since first opioid use (range 0.2–30.9; N = 235)	4.2 ± 6.7
Number of intakes for medically assisted treatment
One	186 (64.8)
More than one	83 (29.9)
Not reported	18 (6.3)
Number of cigarettes typically smoked in a day
0	52 (18.1)
1–9	28 (9.8)
10–19	96 (34.9)
20–29	80 (29.1)
30+	19 (6.6)
Not reported	12 (4.2)
Uses smokeless tobacco
Yes	26 (9.1)
No	249 (86.8)
Not reported	12 (4.2)
Uses e-cigarettes
Yes	114 (39.7)
No	163 (56.8)
Not reported	10 (3.5)
T-ACE alcohol risk scale (range 0–5; N = 129)²⁰
0–1	80 (27.8)
2 or greater	49 (17.1)
Not reported	158 (55.1)

Table 3.

Reproductive Health Outcomes, Knowledge, Attitudes, and Behaviors, N = 287

	n (%)/mean ± SD
Sexual activity in the last 12 months
Yes	252 (87.8)
No	83 (11.5)
No response	2 (0.7)
Had vaginal sex without a condom in the last 2 years
Yes	232 (80.8)
No	54 (18.8)
Unsure	1 (0.3)
Birth control methods used during the past 5 years (if selected by 10 or more women). Select all that apply (range 1–8 items selected)	1.8 ± 1.1
Condom	113 (39.4)
BTL and/or hysterectomy	106 (36.9)
Oral contraceptives (pill)	69 (24.0)
Depo	58 (20.2)
Implant	58 (20.2)
Withdrawal	53 (18.5)
IUD	25 (8.7)
Plan B	11 (3.8)
Partner's vasectomy	10 (3.5)
No response	32 (11.1)
Current birth control use
Currently pregnant	6 (2.1)
Not using birth control	75 (26.1)
Using birth control	100 (34.8)
Has a BTL	106 (36.9)
Type of current birth control use (if selected by 10 or more women). Select all that apply (range 1–4 items selected)	1.3 ± 0.6
BTL and/or hysterectomy	106 (36.9)
Condom	61 (21.3)
Withdrawal	33 (11.5)
Depo	21 (7.3)
IUD	20 (7.0)
Oral contraceptives (pill)	19 (6.6)
Partner's vasectomy	13 (4.5)
Implant	13 (4.5)
No response	13 (4.5)
In the past 12 months, have there been times when you have not used birth control because: (check all that apply). (If selected by 10 or more women) (range 1–5 items selected)	1.3 ± 0.8
Had tubes tied or had a hysterectomy	94 (32.8)
No insurance	31 (10.8)
You do not think you need to	24 (8.4)
Did not have sex with a man	22 (7.7)
You have always used birth control in the past 12 months	22 (7.7)
You were pregnant	17 (5.9)
You do not have a medical provider	17 (5.9)
Only occasionally have sex	17 (5.9)
It is okay if you get pregnant	16 (5.6)
Expensive	13 (4.5)
Safety concerns about birth control	12 (4.2)
No response	40 (13.9)
Reasons for missing or not making a birth control appointment in the past 12 months. (Check all that apply; range 1–12 items selected)^26,27	1.4 ± 1.2
Items selected by 10 or more women
Tubes tied or hysterectomy	83 (28.9)
No insurance	47 (16.4)
Cost	24 (8.4)
Forgot	22 (7.7)
Never missed appointment	21 (7.3)
Examination is uncomfortable	19 (6.6)
Kept appointment in past 12 months	14 (4.9)
Transportation	11 (3.8)
No response	120 (41.8)
Ever pregnant
Yes	280 (97.6)
No	7 (2.4)
Number of pregnancies; range 0–11	3.3 ± 1.5
Categorical measure of number of pregnancies
None	7 (2.4)
1–2	85 (29.6)
3–4	137 (47.7)
5+	58 (20.0)
Ever had an unplanned pregnancy
Yes	220 (76.7)
No	27 (9.4)
Never pregnant	7 (2.4)
Not reported	33 (11.5)
Percent of pregnancies that were unplanned (range 0%–100%; N = 247)	66.2 ± 35.8
Categorical measure of percent of unplanned pregnancies
0.0	27 (10.9)
14.3–50.0	62 (25.1)
51–83.3	52 (21.0)
100.0	106 (42.9)
Number of abortions; range 0–2	0.27 ± 0.55
None, never pregnant	7 (2.4)
None, ever pregnant	192 (66.9)
1	41 (14.3)
2	13 (4.5)
Not reported	6 (2.0)
Number of births; range: 0–10	2.5 ± 1.6
None, never pregnant	7 (2.4)
None	45 (15.7)
1–2	126 (43.9)
3–4	81 (28.2)
5+	9 (3.0)
Not reported	19 (6.6)
Ever became pregnant while using birth control
Yes	90 (31.4)
No (including never pregnant)	167 (58.2)
Not reported	30 (10.5)
Most cited failed method (N = 90; 1–4 methods selected)
Oral contraceptives	58 (64.4)
Condoms	24 (26.7)
Depo-Provera	23 (25.6)
Withdrawal	12 (13.3)
Not reported	2 (2.2)

BTL, bilateral tubal ligation; IUD, intrauterine device.

Table 4.

Reproductive Knowledge and Attitudes

	n	Mean ± SD
Important to avoid pregnancy (range 1–5)	283	4.0 ± 1.4
Perceived risk of NAS²⁸
Methadone use during pregnancy can affect the fetus	284	4.1 ± 1.1
Taking prescription painkillers during pregnancy can affect the fetus	285	4.6 ± 0.7
Pregnant women should not use painkillers	285	4.2 ± 1.1
Community members are concerned about women using painkillers during pregnancy	284	4.2 ± 1.0
Percent correct on knowledge deficit scale²⁸ (27 items; range 4.6%–95.5%)	280	61.1 ± 15.3
Percent correct on knowledge of IUDs²² (6 items; range 0%–100%)	280	42.3 ± 3 1.4
Social support (4 items; range 1–5)	286	4.3 ± 1.0
The government makes certain that birth control is safe before they come onto the market²⁸ (range 1–5)	281	3.3 ± 1.1
Drug companies do not care if birth control is safe, they just want people to use it so they can make money²⁸ (range 1–5)	284	2.4 ± 1.1
Will get pregnant regardless of BC²⁸ (range 1–5)	284	2.4 ± 1.4
You cannot stop to use BC when aroused²⁹ (range 1–5)	285	2.1 ± 1.2
Using BC is morally wrong²⁹ (range 1–5)	285	1.6 ± 1.0
Men are more responsible with BC than women²⁹ (range 1–5)	285	2.0 ± 1.0

BC, birth control; NAS, neonatal abstinence syndrome.

Sociodemographic characteristics

Age was calculated from the woman's birthdate and the date of survey completion. Women selected all races that applied to them and indicated if they identified as Hispanic or non-Hispanic. The nine categories of employment status were recoded to reflect employed, unemployed, and not in the labor force.¹⁹

Health status and drug use

Depressed mood is the mean value of four items (α = 0.68) adapted from the Center for Epidemiologic Studies Depression Scale.²⁰ Respondents reported how often in the past week they felt depressed, everything they did was an effort, hopeful about the future (reverse-coded), and happy (reverse-coded; 1 = “rarely or not at all” and 4 = “most of the time”). Higher mean scores reflected higher levels of depressed mood. Years since first opioid use were calculated from date of survey, birthdate, and age at first opioid use. Number of cigarettes smoked was assessed with the question, “How many cigarettes do you normally smoke in a day?” Alcohol use was assessed with the 4-item T-ACE (tolerance, annoyance, cut down, and eye opener) score²¹: (1) How many drinks does it take to make you feel buzzed? (0 = “less than 2”; 2 = “2 or more”), (2) Have people annoyed you by criticizing your drinking? (0 = “no,” 1 = “yes”), (3) Have you felt you ought to cut down on your drinking? (0 = “no,” 1 = “yes”), and (4) Have you ever had a drink first thing in the morning to steady your nerves or to get rid of a hangover (0 = “no,” 1 = “yes”). A summative score of 2 or higher indicated positive for risk drinking.

Reproductive health outcomes and behaviors

Contraceptive groups were created based on responses about current birth control use, birth control use in the past year, and reasons for missing or not scheduling appointments for birth control (e.g., BTL). Responses were intentionally prioritized to create three groups of contraception use: (1) BTL, (2) using contraception, and (3) not using contraception. Women identifying as pregnant were a distinct category. Women who indicated having a BTL and/or hysterectomy in response to any of the questions were categorized in the BTL group. Next, women who responded “no use” regarding current birth control use were categorized as not using contraception. Finally, women were identified as currently using contraception if they reported current use of condoms, oral contraceptives (pill), the patch, Depo, the NuVa ring, or “the ring,” IUD or Mirena, a diaphragm, an implant, and/or a partner with a vasectomy. Women were requested to indicate all contraception methods used.

Extra care was taken in calculating pregnancy measures due to inconsistencies in responses to separate questions about the number of pregnancies and pregnancy outcomes. A woman was considered ever pregnant if she indicated one or more pregnancies, abortions, miscarriages, or births. The total number of pregnancies was calculated as the greater of either the total number of pregnancies reported or the sum of miscarriages, abortions, and births. Total pregnancies minus planned pregnancies represented the number of unplanned pregnancies.

Reproductive knowledge and attitudes

Knowledge and attitudes about NAS risk were measured with four adapted survey questions related to fetal alcohol syndrome disorder,²⁸ answered on a 5-point scale (1 = “strongly disagree” to 5 = “strongly agree”). Contraception and pregnancy knowledge were measured as the percent of 27 true/false items answered correctly.³⁰ Due to the particular effectiveness of LARCs, we also separately examined knowledge about IUDs and implants as the percent correct of a subset of 6 of the 27 knowledge questions. Social support³¹ was a 4-item scale (α = 0.94) that assessed partner's and friends' social support related to contraception and the women's ability to discuss contraception with their doctor and partner.

Plan of analysis

The data contained a relatively large proportion of missing responses. Because there was no discernible pattern of missingness, complete-case analysis would have drastically reduced the robustness of the sample and underpowered the study. All 287 cases were analyzed. In the presentation of descriptive statistics, we included a category for “no response.”

The characteristics of the participating women were described using means and standard deviations for the continuous measures and frequencies and percentages for the categorical measures. Chi-square or one-way analysis of variance (ANOVA) was used to explore associations between current contraception status and other characteristics or experiences of the women, with significance determined at p < 0.05. Given the descriptive nature of the study and the sample size of the contraception groups, the level of significance for the bivariate models was not adjusted for multiple comparisons. An adjustment could have increased the risk of a type II error.^32,33 Post hoc Scheffe tests followed all significant ANOVA models. All analyses were conducted using SPSS version 24.³⁴

Results

Sociodemographic characteristics

The 287 study participants resided primarily in 16 East Tennessee counties (97%; n = 278). Almost all were white (92%; n = 264) and non-Hispanic (98%; n = 281), and the mean age was 33.9 years. Over a third of the women (38%; n = 108) were married and most had children (76%; n = 217). Thirteen percent had lost custody of a child at some point.

Socioeconomically, the women were struggling. Fewer than half (46%; n = 133) were employed and half reported an annual household income of $20,000 or less (50%; n = 142). The stressors of poverty accompanied low incomes: nearly two-thirds (61%; n = 175) had difficulties paying their rent, mortgage, or utilities in the past year, and a third (33%; n = 95) moved in with friends or family because they had no other choice. Regarding health insurance, 41% (n = 118) had no insurance and 42% (n = 121) had public insurance.

Health status and substance use

Nearly two-thirds of the women reported their overall health as good, very good, or excellent (64%; n = 182). Regarding depressed mood, 18% (n = 51) had an average score of three or greater—equivalent to “often” or “most of the time.” A third of the women (32%; n = 93) reported ever having nonconsensual sex. Reflecting on the past year, nearly 17% (n = 49) reported physical abuse and 6% (n = 18) reported physical abuse from a romantic partner.

The majority of the women first used opioids as a teenager (median age = 19) and initially accessed opioids through a personal prescription (42%; n = 121) or friend (35%; n = 101). Most (78%; n = 223) smoked cigarettes daily, and of these, almost half (44%; n = 99) smoked a pack or more each day. Nicotine uptake also occurred through smokeless tobacco (9%; n = 26) and e-cigarettes (40%; n = 114). Based on T-ACE scores, 49 women (17.1%) were positive for alcohol risk; however, this score was not calculated for most women (54%; n = 158), as they did not answer the annoyance question.

Reproductive health outcomes and behaviors

Over a third of the women 37% (n = 106) reported they had a BTL and/or hysterectomy, 35% (n = 100) reported currently using contraception, and 26% (n = 100) reported no current contraceptive use. The remaining 2% (n = 6) were pregnant. Condoms were the most common contraception used in the past 5 years and second most common form used in the past year. No insurance and/or expense was selected by 36 women (13%) as a reason for not using birth control in the past year. Few women reported not using contraception because of safety concerns (4.2%; n = 12) or partner opposition (1.7%; n = 5). The most cited reason for not using birth control in the past year was BTL and/or hysterectomy (33%; n = 94).

Over three-quarters of the women (77%; n = 220) reported at least one unplanned pregnancy, and all pregnancies were unplanned for 37% (n = 106). A third (31%; n = 90) reported getting pregnant while using birth control, most frequently the pill (64%; n = 58). Approximately, a fifth (19%; n = 54) reported having an abortion and 5% (n = 13) reported multiple abortions.

Reproductive knowledge and attitudes

Most of the 283 women who answered the question about avoiding pregnancy agreed or strongly agreed that it was important to avoid pregnancy now (71%; n = 201). On average, there was a high level of appreciation for the perinatal risk posed by opioid use (mean = 4.3). In contrast, there was great variability in contraceptive knowledge. The percent of true/false knowledge questions answered correctly ranged from 5% to 96%, with a mean of 61%. The women, on average, answered fewer than half of the six true/false knowledge questions about LARCs correctly. Most reported high levels of social support regarding accessing and talking about contraception (mean = 4.3).

Correlates of current contraception status

Table 5 presents the bivariate associations between the three categories of contraceptive use—BTL (n _max = 106), using contraception (n _max = 100), and not using contraception (n _max = 75)—for the 281 women who were not currently pregnant. None of the sociodemographic characteristics differed between contraception groups. Group differences in health insurance status approached statistical significance (p < 0.10), with 46% (n = 34) of women who did not use contraception being insured compared to ∼62% of those who used contraception (n = 62) or had a BTL (n = 64).

Table 5.

Chi-Squared or Analysis of Variance with Scheffe Post Hoc Test by Current Contraception Status Maximum n = 281

	BTL, n _max = 106	Using contraception, n _max = 100	Not using contraception, n _max = 75
Survey measure	Frequency (%) or mean ± SD			n	Chi-square or F	p	Sig mean contrasts
Sociodemographic characteristics
Age	34.2 ± 6.2	34.0 ± 5.7	34.0 ± 6.4	255	0.0	0.974
Married/cohabitating	41 (40.2)	38 (40.4)	27 (38.0)	267	0.1	0.945
Child ever in Department of Children's Services' custody	12 (14.0)	14 (17.9)	11 (90.5)	214	0.6	0.721
More than high school	35 (34.3)	38 (40.4)	28 (39.4)	267	0.9	0.643
Employed fulltime	34 (42.5)	23 (31.1)	26 (45.6)	211	3.4	0.184
Insured	64 (61.5)	62 (62.6)	34 (45.9)	277	5.8	0.055
Challenge paying rent	58 (56.3)	66 (66.7)	48 (64.9)	276	2.6	0.275
Health status and substance use
General health	2.9 ± 0.9	2.8 ± 1.0	2.8 ± 0.9	271	0.4	0.666
Depressed mood	2.1 ± 0.6	2.2 ± 0.7	2.4 ± 0.7	273	4.2	0.016	3 > 1
Ever had nonconsensual sex³⁵	38 (36.5)	30 (30.6)	25 (34.2)	275	0.8	0.670
Ever physically abused³⁵	18 (17.6)	17 (17.3)	14 (19.2)	273	0.1	0.949
Physically abused by romantic partner³⁵	9 (8.7)	8 (8.2)	1 (1.4)	274	4.3	0.117
Age of first opioid use	21.9 ± 5.6	19.7 ± 5.0	18.8 ± 5.4	262	7.8	0.001	1 > 2,3
Years since first opioid use	12.7 ± 6.6	14.4 ± 6.1	16.5 ± 7.2	230	5.9	0.003	3 > 1
First time in medication maintenance therapy	63 (63.0)	66 (71.7)	52 (73.2)	263	2.6	0.274
T-ACE alcohol risk score	1.0 ± 1.4	1.2 ± 1.4	0.8 ± 1.2	128	0.8	0.435
Cigarette use	86 (82.7)	80 (81.6)	52 (77.6)	269	0.7	0.698
Reproductive health outcomes and behaviors
Number of pregnancies	3.6 ± 1.3	3.4 ± 1.6	3.0 ± 1.7	278	3.7	0.025	1 > 3
Percent unplanned pregnancies	62.2 ± 36.7	70.4 ± 31.7	75.5 ± 32.9	241	3.0	0.054
Percent of pregnancies while on BC	17.9 ± 24.9	12.2 ± 21.0	8.0 ± 15.2	250	4.2	0.016	1 > 3
Never missed birth control appointment³⁶	4 (3.8)	20 (20.0)	10 (13.3)	281	12.9	0.002
Reproductive knowledge and attitudes
Important to avoid pregnancy	4.1 ± 1.3	4.3 ± 1.1	3.6 ± 1.5	277	7.0	0.001	1,2 > 3
Methadone use during pregnancy can affect the fetus	4.2 ± 1.0	4.1 ± 1.1	4.0 ± 1.1	278	0.8	0.445
Taking prescription painkillers during pregnancy can affect the fetus	4.7 ± 0.7	4.5 ± 0.7	4.4 ± 0.8	279	2.3	0.107
Pregnant women should not use painkillers	4.4 ± 0.9	4.2 ± 1.1	3.9 ± 1.2	279	6.2	0.002	1 > 3
Community members are concerned about women using painkillers during pregnancy	4.1 ± 1.0	4.3 ± 0.9	4.2 ± 1.0	278	0.8	0.461
Percent correct on contraception knowledge	59.2 ± 16.3	64.2 ± 13.0	59.8 ± 16.5	274	3.0	0.051
Percent correct on knowledge of IUDs	35.9 ± 31.2	47.9 ± 29.1	44.2 ± 32.9	274	4.0	0.020	2 > 1
Social support	4.4 ± 1.0	4.2 ± 1.1	4.2 ± 0.9	280	1.7	0.396
Government makes certain BC safe	3.2 ± 1.1	3.5 ± 1.0	3.2 ± 1.1	275	1.6	0.200
Drug companies do not care if BC is safe	2.5 ± 1.2	2.4 ± 1.0	2.4 ± 1.1	278	0.5	0.603
Will get pregnant regardless of birth control	2.3 ± 1.5	2.2 ± 1.3	2.6 ± 1.3	278	1.5	0.233
You cannot stop to use BC when aroused	2.2 ± 1.3	2.0 ± 1.2	1.9 ± 1.1	279	1.3	0.283
Using BC is morally wrong	1.5 ± 0.9	1.5 ± 0.9	1.7 ± 1.1	279	1.0	0.360
Men are more responsible with BC than women	2.0 ± 1.1	2.0 ± 1.0	1.9 ± 0.9	279	0.2	0.829

Women who did not use contraception had a higher mean level of depressed mood (mean = 2.4) than women in the BTL group (mean = 2.1). The BTL group initiated opioid use later than the other two groups. Women not using contraception reported a greater number of years since first opioid use than the BTL group. No differences emerged for the other seven health measures.

For reproductive health, the BTL group reported more pregnancies and that a higher proportion of pregnancies occurred while on birth control than did the women not using contraception. The BTL and using contraception groups were more likely to agree with the statement that it was important to avoid pregnancy compared to the women not using contraception. The BTL group had a significantly higher level of agreement with the statement that pregnant women should not use painkillers than women not using contraception. Finally, the women using contraception had higher mean levels of LARC knowledge than women not using contraception. No significant differences by current contraception status emerged for the other 10 knowledge and attitude measures.

Discussion

This study provides a comprehensive description of women aged 18–45 enrolled in MAT in East Tennessee in 2015. The co-occurrence of OUD with multiple socioeconomic, physical, and emotional challenges calls for access to high quality, multidisciplinary healthcare that applies a woman-centered approach.³⁷ For example, the women's high rate of nonconsensual sex (32%) could make gynecological visits physically and emotionally challenging.³⁸ To reduce the likelihood of retraumatization, medical practitioners are encouraged to complete training in treating women with histories of sexual trauma.^39
–41 Recommended training content includes screening all women for sexual abuse before examination, examining with sensitivity, providing empowering messages, and being prepared to refer patients to a therapist.^42
–44

More than a third of the women in our study reported having had surgical sterilization (i.e., BTL and/or hysterectomy). Sterilization was not related to the women's age and far exceeded the estimate for the general U.S. population (15.5% of women aged 15–44 in 2011–2013).⁴⁵ The prevalence of BTL in our study also exceeded that of other studies of women with OUD. A review of 24 such studies identified that the use of very effective contraception (i.e., IUDs, implants, or BTL) ranged from 2% to 40%.⁴⁶ In a recent Appalachian study in North Carolina, 22.5% of participating women receiving MAT for OUD had BTL.⁴ Our measures created a forced inclusion of hysterectomy in the BTL category, which may contribute to the higher prevalence in our study. Further ecological investigations of women with BTL who are enrolled in MAT are warranted, including the exploration of decision-making factors. Such research will help us more effectively support women who are considering BTL, so they are confident, well informed, and at peace with their eventual decision.

Our findings highlight three areas where policy changes might increase access to effective contraception for women receiving MAT: (1) decreasing financial barriers; (2) increasing women's contraception knowledge; and (3) reducing nonfinancial barriers to contraception.

Decreasing financial barriers

For over a third of the women in this study, all of their pregnancies were unintended, and socioeconomic factors were cited as primary contributors to lack of contraception use. Over 80% of the participating women were uninsured or had public insurance. To date, Tennessee and 21 other states have not adopted the ACA Medicaid expansion.⁴⁷ Tennessee also does not have a so-called Section 1115 waiver or a state plan amendment to offer Medicaid-funded expanded family planning services to low-income women who earn too much to qualify for regular Medicaid.⁴⁸ Hence, Tennessee's public insurance coverage is insufficient in reducing economic barriers to contraception.

Providers have also cited inadequate reimbursement/insurance as their greatest barrier to LARC insertion.⁴⁹ The October 2017 expansion of Tennessee Medicaid reimbursement for immediate postpartum LARC insertion (i.e., IUD or implant) may enhance LARC access for a small subset of women enrolled in MAT. In addition, a private foundation, A Step Ahead Foundation of East Tennessee, now offers free transportation and reproductive healthcare for women who desire a contraceptive shot, progestin implant, or IUD.⁵⁰

Increasing contraception knowledge

Of the participating women, 31% reported at least one pregnancy, while using birth control, most often the pill, condoms, or Depo-Provera. These failure rates and the low levels of LARC knowledge, particularly among women who do not use contraception, indicate an educational opportunity for this population. The Substance Abuse and Mental Health Service Administration's (SAMSHAs) newly revised Opioid Treatment Program (OTP) Guidelines⁴⁶ recommend that “OTPs should provide reproductive health education for all patients and, when needed, make appropriate referrals for contraceptive services” (p. 42). These recommendations stress the importance of reproductive health education and services for women enrolled in MAT. OTP collaborations with local public health departments or community health centers may assist in meeting SAMSHAs recommendation.

Expanding OTP services to include family planning could reduce both financial and knowledge barriers to accessing effective contraception. Current SAMSHA guidelines require OTPs to conduct a gynecological exam within 14 days of admission and test for sexually transmitted infections, human immunodeficiency virus, and hepatitis C virus. Recent qualitative research with women in both residential and out-patient opioid treatment in Baltimore found that women were supportive of receiving family planning education and services within their treatment programs.⁵¹ OTP collaborations with local public health departments or community health centers may assist in providing comprehensive family planning services, particularly same-day LARC insertions, thereby reducing cost and increasing access.^49,52

Nonfinancial barriers

Even when public insurance covers contraception, policies regulating when and how contraceptive services are provided can limit access. For example, a 1978 federal policy mandates Medicaid coverage for surgical sterilization immediately postpartum for women aged 21 or older, who consent at least 30 days (but no more than 180 days) before the procedure.^53,54 Due to Medicaid policy requirements, an estimated 24% of sterilization requests are unfulfilled.⁵⁵ Further complicating access to LARCs, Medicaid in Tennessee's coverage of IUD or implant removal is subject to certain medical conditions or “medical necessity.”⁵⁶ This restriction may deter women from using these methods, as the removal may be cost prohibitive. Finally Tennessee's Medicaid does not cover over-the-counter contraception, including spermicide, sponges, or condoms,⁵⁶ one of the most used contraception methods for the participating women. This restriction can reduce or eliminate rapid access to contraception.

Public Chapter 820 was also in effect during data collection. As previously mentioned, this legislation criminalized illegal narcotic use (including opioids) that resulted in harm of the infant (e.g., NAS).¹⁸ Amnesty International deems such legislation that targets marginalized, pregnant women as discriminatory and inhumane.^57,58 Such discrimination appears evident in Tennessee; Shelby County, where two-thirds of the residents are from communities of color, had some of the lowest NAS rates, but the highest arrest rates under Public Chapter 820.⁵⁹

Under the threat of criminal charges, postpartum women with OUD may have opted for BTL to reduce the possibility of criminal charges. Our respondents did not report the date of surgical sterilization, so we cannot explore this possibility. The effect of this legislation on MAT enrollment or pregnancy termination is also unknown, yet women with OUD did cite Public Chapter 820 as leading them to delay or eliminate prenatal care and/or deliver their child outside of the medical setting to avoid prosecution.^22,57 The state legislature allowed this law to sunset on July 1, 2016.²³

Limitations

This study provides unique and comprehensive insight into the characteristics and contraception use of women enrolled in MAT for OUD. Despite the study's contributions, results must be interpreted and used cautiously. First, even though the data were recently collected in 2015, findings from this study may not be generalized to women currently enrolled in MAT in East Tennessee or elsewhere. Data collection occurred during the enforcement of Tennessee's Public Chapter 820 and before the 2017 establishment of A Step Ahead Foundation of East Tennessee.⁵⁰

A second limitation is the inconsistency of the survey responses. Although a focus group was conducted with the target audience to test comprehension of the survey questions, the number of missing and inconsistent responses seemed particularly high, especially regarding pregnancy. For example, of the 36 women who reported no pregnancies, 29 (80.6%) reported one or more births. Response errors, nonreporting, or intentional misreporting may have occurred if respondents misunderstood the questions, could not accurately retrieve information from memory, or had difficulty mapping their experiences onto one of the response categories. The survey also asked about sensitive topics such as pregnancy intentions, pregnancy termination, and contraceptive use, which respondents may misreport to conform to social desirability bias or because they do not trust the confidentiality of their answers.²⁴

Conclusions

The increase of OUD and associated unintended pregnancies demands a better understanding of the contraception challenges of women enrolled in MAT. The identified barriers to contraception—particularly to LARC—were comparable to those identified by the CDC for women in general,²⁵ yet the multidimensional challenges faced by women enrolled in MAT create greater obstacles in overcoming these barriers. Women currently not using contraception were less likely to be insured, had greater depressed mood, and had used opiates for a longer period than their counterparts, even though they were not older. The women's out-of-pocket enrollment in MAT strongly indicates their intention to lead healthier lives. Empowering decision-making about fertility through knowledge and access to quality, affordable care would further support their pursuit of health.

Footnotes

Acknowledgment

This study was funded by contract #GG-15-41325-00, awarded by the Tennessee Department of Health to the Knox County Health Department.

Author Disclosure Statement

No competing financial interests exist.

References

Curet

, Hsi

. Drug abuse during pregnancy. Clin Obstet Gynecol, 2002; 45:73–88.

Howell

, Chasnoff

. Perinatal substance abuse treatment. Findings from focus groups with clients and providers. J Subst Abuse Treat, 1999; 17:139–148.

Goodman

, Onwumere

, Milam

, Peipert

. Reducing health disparities by removing cost, access, and knowledge barriers. Am J Obstet Gynecol, 2017; 216:382.e381–e382.e385.

Poulton

, Parlier

, Scott Ryan

, Fagan

, Galvin

. Contraceptive choices for reproductive age women at methadone clinics in Western North Carolina. MaAHEC Online J Res, 2015; 2:1–11.

Heil

, Jones

, Arria

, et al. Unintended pregnancy in opioid-abusing women. J Subst Abuse Treat, 2011; 40:199–202.

Cornford

, Close

, Bray

, Beere

, Mason

. Contraceptive use and pregnancy outcomes among opioid drug-using women: A retrospective cohort study. PLoS One, 2015; 10:e0116231.

Matusiewicz

, Melbostad

, Heil

. Knowledge of and concerns about long-acting reversible contraception among women in medication-assisted treatment for opioid use disorder. Contraception, 2017; 96:365–369.

Black

, Stephens

, Haber

, Lintzeris

. Unplanned pregnancy and contraceptive use in women attending drug treatment services. Aust N Z J Obstet Gynaecol, 2012; 52:146–150.

Finer

, Zolna

. Declines in unintended pregnancy in the United States, 2008–2011. N Engl J Med, 2016; 374:843–852.

10.

Kavanaugh

, Jerman

. Contraceptive method use in the United States: Trends and characteristics between 2008, 2012 and 2014. Contraception, 2018; 97:14–21.

11.

Winner

, Peipert

, Zhao

, et al. Effectiveness of long-acting reversible contraception. N Engl J Med, 2012; 366:1998–2007.

12.

Committee Opinion No. 711: Opioid use and opioid use disorder in pregnancy. Obstet Gynecol, 2017; 130:e81–e94.

13.

Ross

, Graham

, Money

, Stanwood

. Developmental consequences of fetal exposure to drugs: What we know and what we still must learn. Neuropsychopharmacology, 2015; 40:61–87.

14.

Miller

, Warren

. Neonatal abstinence syndrome surveillance annual report 2015. Nashville, TN: Tennessee Department of Health, 2016.

15.

Tennessee Department of Health. Neonatal abstinence syndrome surveillance summary, week 50: December 10–December 16, 2017. Nashville, TN: Tennessee Department of Health, 2017.

16.

Dunn

, Behringer

, Bowers

. Substance use. In: Ludke

, Obermiller

, eds. Appalachia health and well-being. Lexington, KY: The University Press of Kentucky, 2012:251–274.

17.

Castle

. Abortion in the United States' Bible Belt: Organizing for power and empowerment. Reprod Health, 2011; 8:1.

18.

Tennessee Department of Health. FAQ regarding Public Chapter 820 (PC 820) and other requirements related to neonatal abstinence syndrome (NAS) in Tennessee. Nashville, TN: Tennessee Department of Health, 2014.

19.

United States Department of Labor.. How the government measures unemployment. United States Bureau of Labor statistics,. 2014. Available at: https://www.bls.gov/cps/cps_htgm.pdf. Accessed October 31, 2017 .

20.

Radloff

. The CES-D scale: A self-report depression scale for research in the general population. Appl Psychol Meas, 1977; 1:385–401.

21.

Sokol

, Martier

, Ager

. The T-ACE questions: Practical prenatal detection of risk-drinking. Am J Obstet Gynecol, 1989; 160:863–868; discussion 868–870.

22.

Goldensohn

, Levy

. The state where giving birth can be criminal. The Nation,. 2014. Available at: https://www.thenation.com/article/state-where-giving-birth-can-be-criminal/. Accessed April 6, 2018 .

23.

Tennessee's Fetal Assault Law Sunsets July 1, 2016 [press release]. 2016.

24.

Tourangeau

, Yan

. Sensitive questions in surveys. Psychol Bull, 2007; 133:859–883.

25.

Boulet

, D'Angelo

, Morrow

, et al. Contraceptive use among nonpregnant and postpartum women at risk for unintended pregnancy, and female high school students, in the context of Zika preparedness—United States, 2011–2013 and 2015. MMWR Morb Mortal Wkly Rep, 2016; 65:780–787.

26.

Institute for Social

Research

, University of Michigan. National Survey of America's Families (NSAF) Series,. 2017. Available at: https://www.icpsr.umich.edu/icpsrweb/ICPSR/series/00216 Accessed May 12, 2017 .

27.

Achenbach

, Becker

, Döpfner

, et al. Multicultural assessment of child and adolescent psychopathology with ASEBA and SDQ instruments: Research findings, applications, and future directions. J Child Psychol Psychiatry, 2008; 49:251–275.

28.

Peadon

, Payne

, Henley

, et al. Women's knowledge and attitudes regarding alcohol consumption in pregnancy: A national survey. BMC Public Health, 2010; 10:510.

29.

Centers for Disease Control and Prevention. Measuring healthy days: Population assessment of health-related quality of life. Atlanta: GA, Centers for Disease Control and Prevention, 2000.

30.

Kaye

, Suellentrop

, Soup

. The fog zone: How misperceptions, magical thinking, and ambivalence put young adults at risk for unplanned pregnancy. Washington, DC: The National Campaign to Prevent Teen and Unplanned Pregnancy, 2009.

31.

Ong

, Temple-Smith

, Wong

, McNamee

, Fairley

. Contraception matters: Indicators of poor usage of contraception in sexually active women attending family planning clinics in Victoria, Australia. BMC Public Health, 2012; 12:1108.

32.

Rothman

. No adjustments are needed for multiple comparisons. Epidemiology, 1990; 1:43–46.

33.

Feise

. Do multiple outcome measures require p-value adjustment?. BMC Med Res Methodol, 2002; 2:8.

34.

IBM SPSS Statistics for Windows, Version 24.0 [computer program]. Version 24.0. Armonk, NY: IBM Corp., 2015.

35.

Basile

, Hertz

, Back

. Intimate partner violence and sexual violence victimization assessment instruments for use in healthcare settings. Version 1.0. 2007. Available at: http://www.cdc.gov/violenceprevention/pdf/ipv/ipvandsvscreening.pdf. Accessed May 25, 2015 .

36.

Neal

, Hussain-Gambles

, Allgar

, Lawlor

, Dempsey

. Reasons for and consequences of missed appointments in general practice in the UK: Questionnaire survey and prospective review of medical records. BMC Fam Pract, 2005; 6:47.

37.

Morse

, Ramesh

, Jackson

. Reassessing unintended pregnancy: Toward a patient-centered approach to family planning. Obstet Gynecol Clin North Am, 2017; 44:27–40.

38.

Coles

, Jones

. “Universal Precautions”: Perinatal touch and examination after childhood sexual abuse. Birth, 2009; 36:230–236.

39.

Seng

, Sparbel

, Low

, Killion

. Abuse-related posttraumatic stress and desired maternity care practices: Women's perspectives. J Midwifery Womens Health, 2002; 47:360–370.

40.

Farley

, Golding

, Minkoff

. Is a history of trauma associated with a reduced likelihood of cervical cancer screening?. J Fam Pract, 2002; 51:827–831.

41.

WHO Guidelines Approved by the Guidelines Review Committee. Responding to intimate partner violence and sexual violence against women: WHO Clinical and Policy Guidelines. Geneva: World Health Organization, 2013.

42.

Committee Opinion No. 498: Adult manifestations of childhood sexual abuse. Obstet Gynecol, 2011; 118(Pt 1):392–395.

43.

Beck

. Posttraumatic stress disorder after birth: A metaphor analysis. MCN Am J Matern Child Nurs, 2016; 41:76–83; e75–e76.

44.

LoGiudice

. A systematic literature review of the childbearing cycle as experienced by survivors of sexual abuse. Nurs Womens Health, 2017; 20:582–594.

45.

Daniels

, Daugherty

, Jones

. Current contraceptive status among women aged 15–44: United States, 2011–2013. NCHS Data Brief, 2014; 1–8.

46.

Terplan

, Hand

, Hutchinson

, Salisbury-Afshar

, Heil

. Contraceptive use and method choice among women with opioid and other substance use disorders: A systematic review. Prev Med, 2015; 80:23–31.

47.

The Henry

. Kaiser Family Foundation. State requirements for insurance coverage of contraceptives, 2017. Available at: https://www.kff.org/other/state-indicator/state-requirements-for-insurance-coverage-of-contraceptives/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Coverage%20Required%3F__Prescription%20Contraception%22,%22sort%22:%22asc%22%7D Accessed October 9, 2017 .

48.

The Henry

. Kaiser Family Foundation. States that have expanded eligibility for coverage of family planning services under medicaid,. 2017. Available at: https://www.kff.org/medicaid/state-indicator/family-planning-services-waivers/?currentTimeframe=0&sortModel=%7B%22colId%22:%22Location%22,%22sort%22:%22asc%22%7D Accessed March 5, 2018 , 2018.

49.

Luchowski

, Anderson

, Power

, Raglan

, Espey

, Schulkin

. Obstetrician-gynecologists and contraception: Long-acting reversible contraception practices and education. Contraception, 2014; 89:578–583.

50.

A Step Ahead Foundation of East Tennessee.. About A Step Ahead Foundation of East Tennessee,. 2018. Available at: www.astepaheadeasttn.org/about Accessed January 31, 2018 , 2018.

51.

Robinowitz

, Muqueeth

, Scheibler

, Salisbury-Afshar

, Terplan

. Family planning in substance use disorder treatment centers: Opportunities and challenges. Subst Use Misuse, 2016; 51:1477–1483.

52.

Tocce

, Sheeder

, Python

, Teal

. Long acting reversible contraception in postpartum adolescents: Early initiation of etonogestrel implant is superior to IUDs in the outpatient setting. J Pediatr Adolesc Gynecol, 2012; 25:59–63.

53.

Borrero

, Zite

, Potter

, Trussell

. Medicaid policy on sterilization—Anachronistic or still relevant?. N Engl J Med, 2014; 370:102–104.

54.

Brown

, Chor

. Adding injury to injury: Ethical implications of the Medicaid sterilization consent regulations. Obstet Gynecol, 2014; 123:1348–1351.

55.

Borrero

, Zite

, Potter

, Trussell

, Smith

. Potential unintended pregnancies averted and cost savings associated with a revised Medicaid sterilization policy. Contraception, 2013; 88:691–696.

56.

The Henry

. Kaiser Family Foundation. Medicaid coverage of family planning benefits: Results from a state survey. 2016. Available at: https://www.kff.org/womens-health-policy/report/medicaid-coverage-of-family-planning-benefits-results-from-a-state-survey/. Accessed February 21, 2018 .

57.

Amnesty International.. USA: Tennessee “fetal assault” law a threat to women's health and human rights. Amnesty International,. 2016. Available at: https://www.amnesty.org/download/Documents/AMR5136232016ENGLISH.pdf. Accessed January 9, 2018 .

58.

Amnesty International. Criminalizing pregnancy: Policing pregnant women who use drugs in the USA. London, United Kingdom: Amnesty International, 2017.

59.

United States Census Bureau.. QuickFacts: Shelby County, Tennessee,. 2016. Available at: https://www.census.gov/quickfacts/fact/table/shelbycountytennessee/RHI225216#viewtop. Accessed October 31, 2017 .