Contraceptive Practices, Preferences, and Barriers Among Abortion Clients in North Carolina

Abstract

Objective

Abortion clinics provide an ideal setting for women to receive contraceptive care because the women served may not have other contacts with the health system and are at risk for unintended pregnancies. The objective of this study was to understand practices, preferences, and barriers to use of contraception for women obtaining abortions at clinics in North Carolina.

Methods

We conducted a cross-sectional survey of abortion clients and facilities at 10 abortion clinics in North Carolina. We collected data on contraceptive availability at each clinic. We collected individual responses on women’s experiences obtaining contraception before the current pregnancy and their intentions for future use of contraception.

Results

From October 2015 to February 2016, 376 client surveys were completed at 9 clinics, and 10 clinic surveys were completed. Almost one-third of women (29%) reported that they had wanted to use contraception in the last year but were unable. Approximately three-fourths of respondents (76%) stated that they intend to use contraception after this pregnancy. Approximately one-fifth of women stated that would like to use long-acting reversible contraception (LARC) after this abortion. Only the clinics that accepted insurance for abortion and other services provided LARC at the time of the abortion (40%).

Conclusions

This study provides a unique, statewide view into the contraceptive barriers for women seeking abortion in North Carolina. Addressing the relatively high demand for LARC after abortion could help significantly reduce unintended pregnancy and recourse to abortion in North Carolina.

Women presenting for abortion are at risk for a subsequent unintended pregnancy and abortion.^1–3 Fertility can return as quickly as 1 to 2 weeks postabortion,^4,5 so timely initiation of contraception after abortion is important. Women who receive a contraceptive method after an abortion are less likely to have a repeat pregnancy postabortion,^2,6–8 and long-acting reversible contraceptive (LARC) methods in particular have been shown to reduce rates of repeat abortion in women who receive them immediately postabortion.^6,9–11 Many women state that they would like to use contraception after an abortion,^12–14 and providing highly effective contraception at the time of abortion offers a logical way to prevent another unwanted or mistimed pregnancy.

This study provides a unique snapshot of the contraceptive experiences of women seeking abortion at 10 abortion clinics in North Carolina, a state that embodies many of the challenges to access to contraception and abortion in the country. We examined the barriers and challenges to obtaining contraceptive methods before the current pregnancy and postabortion. We explored women’s willingness to use LARCs after the abortion. We also sought to understand clinic-level contraceptive availability at the time of abortion for women pursuing abortions in study clinics in North Carolina. This information is useful for planning programs to prevent unintended pregnancies, repeat unintended pregnancies, and abortions.

Methods

We conducted a cross-sectional survey of abortion clients and facilities in North Carolina between October 2015 and the end of February 2016. We first identified all of the abortion clinics in North Carolina via an Internet search and provider lists and then contacted every clinic to ask whether they would be willing to participate in the study. Facility surveys were administered in person by a member of the study team. For the client surveys, clinics were asked to hand clients a clipboard containing the survey at a point during the appointment when the client would be waiting and able to fill out the survey. Client surveys were self-administered and limited to women aged 18 years and older. Upon completion, the client enclosed the survey in a sealed envelope and placed it into a locked box to ensure confidentiality. No personal identifying information was obtained as part of the survey. Institutional review board approval for all of the study procedures was received from the University of North Carolina at Chapel Hill.

Questions for the facilities answered by the clinic managers included the number of abortions performed in the last 3 months, credentials of staff members, and capacity of clinics to provide services other than abortion. Client surveys included questions regarding basic demographics, gestational age, experience obtaining an abortion, and clients’ past use and current plans for contraception and access to contraception before the current pregnancy (copies of the survey tools available upon request from the first author).

We performed descriptive analyses of data using STATA 14.5 (StataCorp, College Station, TX). We used χ² and Fisher exact tests for bivariate analyses. To further explore the associations, we stratified the analyses by education level (less than or equal to high school education and greater than high school education).

Results

We approached 16 clinics in the state of North Carolina. Of these, 10 agreed to participate in both the facility and the client surveys. From October 1, 2015 through February 29, 2016, 376 client surveys were completed at 9 clinics. One additional high-volume clinic contributed only 10 client surveys because of logistical challenges in implementing the survey. Because of the small number of participants, client surveys from that clinic were excluded. The majority of clients surveyed (64%) were between 18 and 29 years of age. One-fourth of abortion clients were married or previously married and 69% were single or never married. More than half of the respondents (53%) had a high school education or less, and 44% had associate’s degrees, technical degrees, or were college educated; this sample had higher educational levels than the state or national level education levels for women seeking abortions. More than half were employed full-time (54%). Other demographics mirror general patterns in demographics for abortion clients in the United States¹⁵ and demographic characteristics by education status are as expected (ie, more educated abortion clients are more likely to be employed and have a higher family income; Table 1).

Table 1.

Characteristics of women obtaining abortions in North Carolina study sites^a

Characteristic	N	%	≤HS	%	>HS	%
Age, y
18–29	240	63.8	148	74.0	90	53.6
≥30	126	33.6	49	24.5	77	45.8
Don’t know/missing/refused	10	2.6	3	1.5	1	0.6%
Race/ethnicity
White	135	35.9	59	29.5	75	44.6
Black/African American	180	47.9	109	54.5	67	39.9
Asian, Hispanic, other or missing	61	16.2	32	16.0	26	15.5
Education
≤HS	200	53.2	—	—	—	—
>HS	168	44.7	—	—	—	—
Don’t know/refused	8	2.1	—	—	—	—
Current marital status
Married/previously married	104	27.7	44	22.0	60	35.7
Single/never married	261	69.4	151	75.5	106	63.1
Don’t know/missing/refused	11	2.9	5	2.5	2	1.2
Living arrangement
Living with partner	130	34.6	60	30.0	69	41.1
Living with parents	79	21.0	58	29.0	18	10.7
Work status
Employed full-time (≥30 h)	203	54.0	87	43.5	114	67.9
Employed part-time	74	19.7	51	25.5	23	13.6
Not employed	99	26.3	62	31.0	31	18.5
Currently attending high school or college	113	30.1	60	30.0	49	29.2
Family income, per year
<$25,000	149	39.6	101	50.5	46	27.4
$25,000–$49,000	108	28.7	45	22.5	62	36.9
>$50,000	61	16.3	14	7.0	46	27.4
Don’t know/missing/refused	58	15.4	40	20.0	14	8.3
Family receiving government assistance	152	40.4	101	50.5	49	29.2
Total	376	100	200	100	168	100

HS, high school.

^aResults by education exclude 8 cases missing on education.

We found that more than half of women (52%) had used contraception at some point in the year before the abortion (Table 2). The main methods used in the last year were the oral contraceptive pill, condoms, and withdrawal. There were no differences in the use of LARC methods among women with higher and lower educational attainment; however, women with lower educational attainment were more likely to use injectables, whereas women with higher educational attainment were more likely to use a vaginal ring, emergency contraception, or natural family planning methods.

Table 2.

Contraceptive experience in the last year among women obtaining an abortion at North Carolina study sites

Characteristic	No. (N = 376)	%	≤HS (n = 200), %	>HS (n = 168), %	P
Used any birth control in the last year	197	52.4	45.5	61.3	**
Method used in the last year among those who used a method (n = 197)a,b
Oral contraceptive pill	117	59.4	63.7	56.3
Injection	21	10.7	14.3	5.8	**
Implant	5	2.5	1.1	3.9
Diaphragm	1	0.5	0.0	1.0
Foam	1	0.5	0.0	1.0
Patch	4	2.0	2.2	1.9
Ring	13	6.6	3.3	9.7	*
IUD	12	6.1	6.6	5.8
Condom	87	44.2	42.9	46.6
Withdrawal	73	37.1	31.9	41.8
Natural family planning methods	21	10.7	2.2	18.5	***
Emergency contraception	28	14.2	8.8	18.5	*
Spermicide	2	1.0	0.0	1.9
“Was there ever a time in the last year that you wanted to use birth control but were not able to?”	106	28.8	31.5	25.6
“What prevented you from using birth control?” among women who were not able to use it in the last year (n = 106)b
Cost	60	56.6	57.1	55.8
Could not drive to pick it up	11	10.4	11.1	9.3
Did not know where to get it	11	10.4	9.5	11.7
Did not know about types of birth controla	9	8.3	7.9	7.0
Concerns about adverse effects	31	28.4	27.0	30.2
Did not want to see a doctor for prescription	10	9.2	6.4	14.0	**
My partner did not want to use birth control	4	3.7	0.0	9.3
Could not talk to parents about birth control	8	7.6	6.4	9.3
Other	20	18.4	17.5	20.9

HS, high school; IUD, intrauterine device.

^aResults by education exclude 8 cases missing on education.

^bMultiple responses possible.

^*P < 0.1,

^**P < 0.05,

^***P < 0.001.

At the time that the current pregnancy was conceived, more than half of the respondents were not using any type of contraception (54%; Table 3). A larger percentage of the women with greater than a high school education reported using a contraceptive method at the time of conception (52.4%) as compared with women with lower educational attainment (41%). Notably, of the 170 women who reported using a method at the time of the pregnancy, 29.4% reported that their method was withdrawal, followed by 27.7% who used condoms, and 26.5% who used contraceptive pills. Approximately 7% of all abortion clients surveyed reported that they used emergency contraception to prevent this pregnancy. When asked how much they wanted to become pregnant the month before the pregnancy, the majority of respondents (75%) reported “not at all.” There were no differences in pregnancy intentions among women with higher and lower educational attainment (Table 3).

Table 3.

Pregnancy intention and contraceptive use when this pregnancy occurred

Characteristic	No. (N = 376)	%	≤HS (n = 200), %	>HS (n = 168), %	P
Type of birth control used when this pregnancy occurred
No method	198	53.8	59.0	47.6	**
Some method	170	46.2	41.0	52.4	**
Method used at time of pregnancy among those using some method (n = 170)a,b	45	26.5	20.7	31.8
Oral contraceptive pill	4	2.4	3.7	1.1
Injection	2	1.2	1.2	1.4
Implant	0	0	0	0
Diaphragm/foam/patch	2	1.2	1.2	1.1
Vaginal ring	1	0.6	0	1.1
IUD	47	27.7	31.7	23.9
Condom	50	29.4	19.5	38.6	**
Withdrawal	8	4.7	0	9.1	**
Natural family planning methods	7	4.1	3.7	4.6
Other	—	—	—	—
Emergency contraception used to prevent this pregnancy (% yes among all women surveyed)	26	6.9	5.5	8.3
How much did you want to become pregnant in the month before this pregnancy?
Not at all	281	74.7	76.5	72.6
A little	22	5.9	5.5	6.6
Neutral	15	4.0	2.5	6.0
Somewhat	10	2.7	2.0	3.6
Very much	10	2.7	1.5	4.2
Don’t know	7	1.9	2.5	1.2
Missing/refused	31	8.2	9.5	6.0

HS, high school; IUD, intrauterine device.

^aResults by education exclude 8 cases missing on education.

^bMultiple responses possible.

^*P < 0.1,

^**P < 0.05,

^***P < 0.001.

Almost one-third of women (29%) reported that they had wanted to use contraception in the last year but were not able to do so (Table 2). The most common reasons women cited for not being able to use contraception in the year before the abortion were cost (56.6%) and concerns about adverse effects (28.4%). Other barriers included not knowing where to obtain birth control (10.4%), not having transportation (10.4%), and not wanting to visit a doctor for a prescription (9.2%; Table 2).

Approximately three-fourths of respondents (76%) stated that they intended to use contraception after this pregnancy, with the most common method desired being the oral contraceptive pill (52%; Table 4). A significantly greater percentage of the more educated women intended to use contraception compared with women with less education. Almost 11% of women stated that would like to use a contraceptive implant, and 10.5% stated they would like to use an intrauterine device (IUD). There were no differences in intention to use an LARC method between women with higher and lower educational attainment.

Table 4.

Intention to use contraception after abortion

Characteristic	No. (N = 376)	%	≤HS (n = 200), %	>HS (n = 168), %	P
Intend to use birth control after abortion	286	76.1	71.0	83.3	**
Preferred method to use after abortion among those who intend to use a method (n = 286)a,b
Oral contraceptive pill	148	51.8	45.8	57.9	**
Injection	37	12.9	15.5	10.0
Implant	31	10.8	12.0	9.3
Diaphragm	0	0	0	0
Foam	2	0.7	0	1.4
Patch	8	2.8	4.2	1.4
Vaginal ring	17	5.9	7.0	5.0
IUD	30	10.5	9.9	11.4
Condom	54	18.9	14.8	23.6	*
Withdrawal	13	4.6	3.5	5.7
Natural family planning methods	11	3.9	0	7.9	***
Emergency contraception	9	3.2	3.5	2.9
Other	1	0.4	0.7	0

HS, high school; IUD, intrauterine device.

^aResults by education exclude 8 cases missing on education.

^bMultiple responses possible.

^*P < 0.1,

^**P < 0.05,

^***P < 0.001.

We were unable to ascertain whether the abortion clients left with a prescription or a pack of pills at the end of the visit because respondents completed the survey before finishing their clinic visit. All of the clinics studied offered contraceptive counseling and prescriptions for birth control pills at the time of abortion (Table 5), however. Only the clinics that accepted insurance for abortion and other services provided LARC at the time of the abortion (4 of 10; 40%). The remaining clinics offered only short-term pill prescriptions (3–6 months) or a 1-month sample pack of pills. Two of the clinics that did not accept insurance did provide injections of depot medroxyprogesterone acetate on the same day as the abortion.

Table 5.

Clinic services offered at North Carolina study clinics

Characteristic	N	%
Gestational age limit for abortion at clinic
First trimester only	1	10
First and second trimester abortions	9	90
Insurance accepted at facility, yes	4	40
Other services offered at clinic (multiple responses possible)
Contraceptive counseling	10	100
Prescriptions for contraceptives	10	100
Contraceptive injections	7	70
IUD insertions	5	50
Implant insertions	4	40
Women’s health services (Pap tests, annual examinations)	5	50

IUD, intrauterine device.

^aMultiple responses possible.

Discussion

This study provides a unique, statewide view into the contraceptive situation for women seeking abortion in North Carolina. This is an area that has not been studied extensively. Most women surveyed wanted to use contraception or had used contraception in the year before their abortion, and as expected, only a small proportion of women were using an LARC method before this pregnancy. In contrast, of the women who intended to use a contraception method after the abortion, almost one-fifth were planning to use an LARC method. A substantial proportion of women had difficulty accessing contraception in the past year, mainly because of cost and fear of adverse effects, but also notably because of a lack of knowledge about methods, lack of transportation, and not wanting to visit a doctor for a prescription. There were no major differences in desire to use LARC or prior LARC use among women with higher and lower educational attainment.

Although most women would like to start some form of contraception after their abortion, the abortion clinics included in this study in North Carolina are not equipped to provide comprehensive contraceptive services such as LARC methods. All clinics are equipped to provide counseling and prescriptions, but methods typically paid for with insurance, such as LARC methods, are not as easily accessible in the study sites. Because not all contraception is equally effective, using the most effective methods has the greatest potential to decrease unintended pregnancy. With increasing use of LARC methods in the last several years, overall rates of pregnancy and abortion have decreased significantly.¹⁶ In North Carolina, approximately 8.3% of women use an LARC method of contraception, which is lower than the national average of 12%.⁷ Although there are no guidelines for an optimal percentage of LARC use in a population,¹⁷ increasing voluntary uptake of LARC methods can have a significant impact on rates of unintended pregnancy and abortion at a population level.⁷ A study in which LARC methods were provided free of cost to eligible women receiving an abortion found that 65% of women chose to initiate LARC methods, and 90% were still using them at 1-year follow-up.¹⁸

Abortion clinics provide an ideal setting for women to receive contraceptive care because the women served may not have other contacts with the health system and are at risk for unintended pregnancies. Unfortunately, multiple barriers to receiving contraception at the time of abortion exist.^19,20 In states where Medicaid does not pay for abortion, for instance, many women pay out of pocket for abortion. Furthermore, many clinics do not accept insurance for abortion and may not accept insurance or Medicaid for other services. Out of concern for their privacy, many women do not use their insurance for contraception, even if it would pay for contraception at the time of the abortion.²¹ In addition, availability of contraception varies greatly across abortion clinics and often is dependent on state policies.

Unfortunately, barriers to abortion care also may make it difficult for clinics to comprehensively address the issue of contraception at the time of abortion. Unlike most medical procedures, abortion carries a high degree of stigma.²² This stigma, which extends to state-sanctioned refusal to include abortion coverage in Medicaid, Tricare, and state-funded insurance in North Carolina, makes it difficult for many clinics to accept any insurance, even those that cover abortion. The upfront costs for implants and IUDs are high. Clinics not accepting insurance generally cannot afford to stock these expensive devices because women will not be able to pay out of pocket for them.¹⁹ This may lead to women being unable to obtain their desired contraception methods at the time of abortion. These women may be at risk of a repeat, unintended pregnancy, and possibly a repeat abortion.

Women in our study who wanted to use a contraceptive method most often chose contraceptive pills, which is consistent with other studies.^12,13,23 In studies in which women are offered more effective methods free of charge, they are more likely to choose these methods after abortion than less effective methods such as pills.^18,24 Identifying strategies to make the long-acting methods available consistently postabortion could go a long way to reduce unintended pregnancies, especially repeat unintended pregnancies for women who often lack access to contraceptive services because of cost, transportation, and lack of consistent healthcare access.

This study has several strengths. The majority of abortion clinics in the state participated. Although six declined, we were able to obtain a wide geographic distribution of clinics throughout the state. The anonymous nature of the study likely decreased social desirability bias for the participating women, although having clients self-administer the study during their visit may have reduced the numbers that participated. To our knowledge, this is the first study to look in detail at the contraceptive experiences of women seeking abortion across a single state.

The limitations of the study include a relatively small, convenience sample. Women who completed the survey may be different from women who attended the clinics during the same time frame but did not complete the surveys. Also, there were several large clinics in the state that did not participate. Missing these clinics makes these data less representative of the situation in the entire state. The number of questions asked was limited to facilitate high completion rates; it is possible that some barriers to contraceptive use were missed. We had planned to conduct regression analyses to determine predictors of contraceptive use after abortion, but our cell sizes were too small.

Addressing women’s need for contraception after abortion is imperative, and a comprehensive approach requires action on multiple levels. Abortion visits may offer an important opportunity to address the barriers to contraception that women cite. Women seeking abortion in North Carolina (and across the United States) disproportionately earn low incomes. Public health campaigns that promote accurate information on contraceptives and how to obtain them at low or no cost are crucial. In states such as North Carolina, where Medicaid cannot be used to cover abortion, wider campaigns to improve the utilization of Title X clinics, Family Planning Medicaid, and the contraceptive mandate of the Patient Protection and Affordable Care Act (PL 111-148) also are needed. Understanding women’s concerns about adverse effects and addressing them proactively can help women feel more confident about choosing the right method and using it effectively.^25–27 Given the high reliance among abortion clients on methods such as withdrawal and condoms, which have few adverse effects but low effectiveness, public health campaigns also should work to provide factual information and emphasize correct and consistent use for these methods. Campaigns also should support and promote improved understanding of more effective long-acting methods such as implants and IUDs. Working on global efforts to decrease the stigma around abortion could lead to an easing of restrictions on insurance and Medicaid for abortion coverage, which would make it easier for clinics to accept insurance and procure more expensive and effective contraceptive options. Abortion clinics provide an ideal setting for disseminating information about contraceptive methods and availability. A comprehensive effort to remove barriers to contraceptive nonuse and facilitate access to contraception could help women reduce their risk of repeat unintended pregnancy and the concomitant need for abortion.

Conclusions

Despite its limitations this study shows that women in North Carolina having an abortion appear not to have their contraception needs fully addressed at the time of abortion. For some women, particularly those with insurance and another healthcare provider, this may not be a problem. For women for whom contact with an abortion provider represents their only point of contact with the healthcare system, an important opportunity is missed, however. Addressing the relatively high demand for LARC methods after abortion could help significantly reduce unintended pregnancy and recourse to abortion in North Carolina.

Key Points.

• Women seeking abortions in North Carolina experience barriers to obtaining contraception.

• A substantial proportion of women seeking abortion in North Carolina would like to use a long-acting reversible contraceptive method after the abortion.

• Addressing the relatively high demand for long-acting reversible contraceptives after abortion could help significantly reduce unintended pregnancy and recourse to abortion in North Carolina.