“It just happens”: A qualitative study exploring low-income women’s perspectives on pregnancy intention and planning

Abstract

Objective

Unintended pregnancy is common and disproportionately occurs among low-income women. We conducted a qualitative study with low-income women to better typologize pregnancy intention, understand the relationship between pregnancy intention and contraceptive use, and identify the contextual factors that shape pregnancy intention and contraceptive behavior.

Study design

Semi-structured interviews were conducted with low-income, African-American and white women aged 18–45 recruited from reproductive health clinics in Pittsburgh, Pennsylvania to explore factors that influence women’s pregnancy-related behaviors. Narratives were analyzed using content analysis and the constant comparison method.

Results

Among the 66 participants (36 African-American and 30 white), we identified several factors that may impede our public health goal of increasing the proportion of pregnancies that are consciously desired and planned. First, women do not always perceive that they have reproductive control and therefore do not necessarily formulate clear pregnancy intentions. Second, the benefits of a planned pregnancy may not be evident. Third, because preconception intention and planning do not necessarily occur, decisions about the acceptability of a pregnancy are often determined after the pregnancy has already occurred. Finally, even when women express a desire to avoid pregnancy, their contraceptive behaviors are not necessarily congruent with their desires. We also identified several clinically relevant and potentially modifiable factors that help to explain this intention-behavior discrepancy, including women’s perceptions of low fecundity and their experiences with male partner contraceptive sabotage.

Conclusion

Our findings suggest that the current conceptual framework that views pregnancy-related behaviors from a strict planned behavior perspective may be limited, particularly among low-income populations.

1. Introduction

Over the past several decades, the proportion of pregnancies in the United States (US) that are unintended has remained stubbornly high at approximately 50%.¹ Furthermore, unintended pregnancy continues to be disproportionately experienced by low-income populations and racial minorities.¹ As unintended pregnancy confers significant adverse social and health consequences for women and their families, disparities in unintended pregnancy can contribute to the cycle of disadvantage experienced by vulnerable populations.^2–5

The proximate cause of unintended pregnancy is sexual activity in the absence of effective contraception. Thus, efforts to reduce unintended pregnancy have primarily focused on improving education and knowledge about methods or on increasing access to contraceptive services and methods.^6,7 These efforts, however, have not to date made a substantial dent in the national rate of unintended pregnancy. Although hopes remain that widespread access to no-cost contraception under the Affordable Care Act may yet change the national landscape, there is a clear need to also consider the larger socio-cultural contexts in which pregnancy and contraceptive decision making occur.

Most approaches to understanding pregnancy decision making are grounded in a planned behavior framework.⁸ In this framework, pregnancy intention has been posited as the most immediate determinant of fertility-related behaviors including contraceptive use.^9–12 Women are viewed as formulating pregnancy intention (either anti-pregnancy or pro-pregnancy intention) and then acting, to the extent possible, in accordance with this intention. However, as women’s thoughts about pregnancy are often complicated and may even be contradictory, some researchers have called for a more nuanced characterization of intention to include various categories of sub- or ambiguous intention in order to better inform pregnancy prevention efforts.^13–15 Therefore, we conducted a qualitative study with low-income African-American (AA) and white women in Pittsburgh, PA to better typologize pregnancy intention, understand the relationship between pregnancy intention and contraceptive use, and identify the contextual factors that shape pregnancy intention and contraceptive behavior in a population at high risk of unintended pregnancy.

2. Methods

2.1 Recruitment

Flyers advertising the study were posted in 7 reproductive health clinics that serve low-income populations in Western Pennsylvania. Women responding to advertisements were screened for eligibility over the phone and were considered eligible if they were between the ages of 18–45; self-identified as either AA or white; and were either currently pregnant, had an abortion within the prior 2 weeks, or were not pregnant but had been sexually active with a man in the previous 12 months. We excluded women who were not fluent in English and who had a household income above 200% of the federal poverty level.

In qualitative studies, sample size is driven by thematic saturation, and many researchers suggest that thematic saturation will be reached by 12–15 interviews per group.¹⁶ Therefore, we conducted interviews with at least 15 women from each racial group (AA and white) and from each pregnancy category (pregnant and non-pregnant). We also used a sampling matrix to ensure that we heard the perspectives of participants from each race who varied with respect to age, parity, and among pregnant women, whether they planned to continue or terminate their pregnancy.

2.2 Interview procedures

Semi-structured interviews were conducted between June 2010 and January 2013 by a skilled interviewer (C.N.) with extensive experience collecting qualitative data on sensitive topics such as sexuality, pregnancy, and contraception in diverse populations. Using the Theory of Planned Behavior as framework,⁸ we developed an interview guide to explore factors that might influence women’s conceptualization around pregnancy intention including: 1) thoughts about pregnancy and motherhood, 2) contraceptive use, including perceived barriers to and facilitators of contraceptive use; 3) nature of relationship with partner and partner influence on contraceptive behavior and pregnancy decisions; and 4) attitudes toward and perceived social norms regarding sexuality, pregnancy, contraception, abortion, and partnerships. For pregnant or recently pregnant women, we emphasized that we were interested in hearing about their preconception thoughts and behaviors, although we also explored how these may have evolved throughout the pregnancy.

All sessions were audio-recorded and transcribed verbatim except that participants’ names were omitted for confidentiality. At the end of the interview, each participant was asked to complete a brief paper-based socio-demographic questionnaire. All participants received $50 as compensation for her time. This study was approved by the University of Pittsburgh Institutional Review Board.

2.3 Data analysis

Study transcripts were analyzed using content analysis. This method involves the breakdown of interview text into “units” which are formulated into thematic categories. These categories represent both an exploration of pre-defined areas of study inquiry, as well as new themes that emerged during participant interviews.¹⁷ A codebook, reflecting primary categories and subcategories, was developed and refined as new themes emerged. Two coders independently coded 50% of the transcripts using Atlas.ti qualitative coding software (GmbH, Germany) and compared their coding to determine whether there were any inconsistencies, which were typically resolved through discussion. The principal investigator (S.B.) was available to adjudicate any differences in interpretation between the coders and to review the coding scheme. The primary coder (C.N.) then coded the remaining half of the transcripts. As codes were classified into larger themes, we also searched for meaningful patterns by race and pregnancy status using the constant comparison method, a central analytical approach in which codes are compared across participant types thus leading to relational discovery.¹⁸

3. Results

The final study sample included a total of 66 women (36 non-Hispanic AA women and 30 non-Hispanic white women). Forty-three participants (65%) were between the ages of 18–24 and 23 (35%) were 25–45 years of age. Of the 35 women who were pregnant or recently pregnant, 17 planned to continue their pregnancies and 18 (51%) had a recent abortion or were planning to terminate their pregnancy. Thirty-one women were not pregnant at the time of interview. Additional study sample characteristics are shown in Table 1.

Table 1.

Sample demographic characteristics, (%)

Characteristic	Total (n=66)	African American Women (n=36)	White Women (n=30)
Pregnancy status
Pregnant and continuing pregnancy	25.7	25.0	26.7
Recent abortion or planning abortion	27.2	30.5	23.3
Non-pregnant	46.9	44.4	50.0
Age
18–24	65.2	63.9	66.7
25–45	34.8	36.1	33.3
Education
<High School diploma	10.6	11.1	10.0
High School diploma/GED	43.9	61.1	23.3
Trade/technical school	3.0	2.8	3.3
Some college	21.2	13.8	30.0
College degree	21.2	11.1	33.3
Income
$0 – $9,999	45.5	47.2	43.3
$10,000 – $19,999	30.3	33.3	26.7
$20,000 – $29,999	10.6	5.6	16.7
$30,000 – $49,999	13.6	13.9	13.3
Marital Status
Single	46.9	58.3	33.3
Single, living with male partner	40.9	30.6	53.3
Married	6.1	5.6	6.7
Divorced/separated	4.5	2.8	6.7
Widowed	1.5	2.8	0.0
Parity
0	50.0	41.7	60.0
1	24.2	19.4	30.0
2	13.6	22.2	3.3
3	6.1	8.3	3.3
4+	6.1	8.3	3.3
Insurance
Private	16.7	5.6	30.0
Public	65.2	86.1	40.0
None	18.2	8.3	30.0
Religion
None	42.4	47.2	36.7
Protestant	4.5	5.6	3.3
Catholic	21.2	2.8	43.3
Other Christian	24.2	41.7	3.3
Other	7.6	2.8	13.4

NOTE: One participant took part in both the pregnant and not pregnant cohort interviews. Given the time lapse between interviews, she was no longer pregnant at the time of the second interview.

Four overarching themes emerged from women’s narratives about pregnancy and contraceptive decision making. We found that these themes were similar across race and pregnancy status, thus, the results presented below reflect perspectives from all women. However, a few subthemes differed across cohorts of women, which we highlight in the text.

Theme 1: Women do not always formulate pregnancy intentions

Women generally fell into one of the following four categories of pregnancy intention: wanting to avoid pregnancy, desiring pregnancy, ambivalent about their pregnancy desires, and finally, lacking intention. While the majority of women expressed that they did not desire pregnancy when explicitly asked about their current or preconception thoughts about pregnancy, only 2 women in the entire cohort expressed desire for pregnancy and had actively taken steps to help ensure conception, and 1 woman described conflicting emotions and thoughts (ambivalence) regarding potential pregnancy. Many women, however, had not formulated any thoughts about their pregnancy desires or intentions.

Scenarios in which women described lack of intention included spontaneous (unplanned) sex and being under the influence of alcohol or drugs. However, more commonly, lack of intention stemmed from perceptions of low reproductive control, as many of these women expressed that they did not feel that they necessarily had any agency over their reproductive outcomes. A few women invoked religious explanations, “this is the will of God,” but others simply felt that pregnancy was not something that could or should be prevented. One woman explained, “like you never had it in your mind that you wanted to have a baby, it just happened.” Another participant said, “Nobody can really plan for a pregnancy, like, you could try but a lot of people that wanna get pregnant don’t get pregnant…then there’s a lot of people that don’t want to get pregnant and it just happens.” “It just happened” was a commonly used phrase to describe participants’ experiences with pregnancy. Another woman felt, “If you are meant to have a kid, you are meant to have a kid. Why take something to prevent it?” Thus, an unintended pregnancy did not necessarily reflect clear anti-pregnancy intentions preconception, but often a lack of intention.

Theme 2: Pregnancy planning was described as an unattainable ideal by many women

Pregnancy planning, distinct from pregnancy intention, was described by most women as a very deliberate act in which both partners discuss and reach consensus about the timing of pregnancy, and then take steps to prepare for a potential pregnancy, including “getting your finances in order.” Because nearly all of the women in our study had strong feelings about the ideal circumstances (specifically, being in a committed relationship and financially stable) in which one should plan a pregnancy, yet few, if any, women actually achieved either relationship or financial stability, pregnancy planning seemed irrelevant and rarely occurred among the women in our sample. One woman actually criticized her friend who planned a pregnancy within what she considered non-ideal circumstances:

Participant: “they’re engaged but they decided to have this baby before they were going to get married. Like they were striving, like she planned this baby. She got off all her meds, she started going to the doctor’s and taking prenatal pills before she got pregnant. Like she planned to have this baby. And I didn’t know that part cause I would have had a issue with that.”

Interviewer: “Why?”

Participant: “Because you’re not married”

Even though most of the pregnancies in the participant’s social network occur within the context of non-marital unions (including her own pregnancy), the act of planning a pregnancy under such circumstances was judged as inappropriate. Less value was placed on the inherent benefits of planning a pregnancy than the context in which it should occur.

Several women agreed that planning pregnancy was, in theory, beneficial with regard to optimizing timing in terms of educational goals and career opportunities. However, nearly all women reported that their current pregnancies were not only unplanned but occurred at a non-optimal time. Only 2 woman (both white) in our sample made a conscious decision to get pregnant, took steps to ensure conception at a time that they perceived was optimal for them, and adopted healthier behaviors (took prenatal vitamins and/or ate more healthily). Other than these 2 women, no one discussed the advantages of planning a pregnancy with respect to the opportunity to engage in healthier prenatal behaviors.

Theme 3: Pregnancy intendedness, happiness about pregnancy, and acceptability of pregnancy are distinct constructs

Many women expressed happiness with (the prospect of) a pregnancy, regardless of their intention. For example, one 22 year-old woman with 3 children was pregnant with her 4^th child after an IUD expulsion. She expressed being happy and excited about the pregnancy:

“I’m more excited than … scared. I was scared in the beginning ‘cause it’s like 4 kids, like, I’m 22, like wow, what is that. But after a while I got used to, like, it’s another baby. I mean, I been doin’ good as a mother with 3 kids at 22. It’s kinda hard to bring another one but this is, this is the hands I was dealt so.” (AA participant, age 22)

Conversely, 2 women described scenarios in which they terminated a “wanted pregnancy” because of their current unsuitable financial or relationship situation. Furthermore, circumstances which dictate the acceptability of pregnancy in this sample of women are constantly in flux. For example, one participant who had an abortion 4 months previously, was 8 weeks pregnant at the time of her interview. She decided to continue with this pregnancy, however, because “all my money is starting to fall in line and school, I’ll be finishing it up…when I’m due.” Thus, both cognitive factors (intention) and affect (happiness) are important, and potentially conflicting, in determining the acceptability and outcome of a pregnancy. Rather than pre-conception intention and planning, most women in our sample made post-conception assessments about the timing, their readiness and happiness, and thus, the acceptability of an established pregnancy.

Theme 4: The relationship between desire to avoid pregnancy and contraceptive behavior was often unclear

Although most women voiced that they did not desire or intend pregnancy at this time or prior to their current pregnancy, this did not necessarily mean they were actively trying to avoid pregnancy by engaging in effective contraceptive use. Given that almost none of the women in our sample were seeking pregnancy yet contraceptive behaviors ranged widely (Table 2), there was no obvious relationship between intention and contraceptive behavior. However, as many of these women were engaging in unprotected intercourse, they realized that the potential for pregnancy existed, and described ways in which they either approached this risk or tried to explain the apparent contradiction in their behavior. These included simple willingness to take the risk (“playing roulette”) and “carelessness” about getting hormonal methods refilled or having condoms on hand. Several (n=5) of the currently/recently pregnant women reported method failure (IUD expulsions, condom break, vasectomy failure, pill failure). Almost no one cited cost or lack of access as reasons for contraceptive non-use, even when explicitly asked. We also identified two potentially modifiable factors that contributed to contraceptive use or inconsistent use, both of which appeared to be more common among AA participants:

Table 2.

Women’s reported use of contraception, (n)

Type of contraceptive methoda	Contraceptive method used prior to conception among women who were currently or recently pregnant at the time of interview (n = 35)	Current or most recent contraceptive method used among women who were not pregnant at the time of interview (n= 31)
Sterilization (male or female)	1	1
IUD	2	5
Injection	0	4
Pill	5	7
Condoms	10	8
None b	17	6

^aOnly those methods used by study participants are shown in the table. If women reported using more than 1 method, only the most effective method is included.

^bOnly 2 women reported actively trying to get pregnant (1 woman in the pregnant cohort and 1 woman in the non-pregnant cohort)

1. Perceived low susceptibility to pregnancy. About a quarter (23%) of the total sample indicated that they believed that they were subfertile or infertile. This explanation was more common in the currently/recently pregnant cohort where 43% of women reported believing that they could not get pregnant and was reported by both women who had had prior pregnancies as well as those for whom this was the first pregnancy. Additionally, in the pregnant cohort, more AA women than white women reported perceptions of subfertility (45% vs 33%, respectively). As a result, “shocked” was a word commonly used by women to describe their initial reaction when learning about a pregnancy. Many women reported that previous unprotected intercourse without pregnancy led to their assumption of subfertility and subsequent contraceptive non-use or inconsistent use:

   Participant: “I personally thought I couldn’t, not that I couldn’t conceive but just I wasn’t fertile.”

   Interviewer: “Why? What gave you that thought?”

   Participant: “Cause I had sex unprotected before, and just the fact that I’ve never been pregnant.” (AA woman who is not using contraception, age 19)

2. Male partner reproductive coercion: Twenty-one (32%) of our participants reported one or more personal experiences with male partner reproductive coercion ranging from verbal and emotional pressure to get pregnant to overt birth control sabotage. Reports of reproductive coercion were more common among AA participants compared to white participants (44% vs 17%, respectively). Furthermore, accounts provided by white participants did not describe the same degree of overt contraceptive sabotage and pregnancy pressure that the AA women in our sample described. More AA women than white woman (n=8 and 1, respectively) reported their current or a past pregnancy resulted directly from birth control sabotage and/or pregnancy pressure by a male partner. One woman described her experience:

   I had condoms, he threw them away. I had contraceptive stuff, the foam stuff, he threw it away…And I had a whole bag of stuff, the day after pills, he just threw the whole bag away…[Regarding birth control pills] I had ‘em hidden for a minute…I told him they were vitamins and… I guess he researched on ‘em and then I came home one day and [he said], ‘these are not vitamins.’ (AA woman, age 19)

4. Discussion

In this qualitative study exploring reproductive decision making in low-income AA and white women in Pittsburgh, PA, we identified several factors that may serve as roadblocks to achieving our public health goal of increasing the proportion of pregnancies that are consciously desired and planned. First, women do not always perceive that they have reproductive control and therefore do not necessarily formulate clear pregnancy intentions. Second, the benefits of a planned pregnancy may not be evident. Third, because preconception intention and planning do not necessarily occur, decisions about the acceptability of a pregnancy are often determined after the pregnancy has already occurred. Finally, even when women express desire to avoid pregnancy, their contraceptive behaviors are not necessarily congruent with their desires. We identified two clinically relevant and potentially modifiable factors that help to explain this intention-behavior discrepancy: women’s perceptions of low fecundity and their experiences with male partner contraceptive sabotage. Our findings suggest that the current conceptual framework that views pregnancy-related behaviors from a strict planned behavior perspective may be limited, particularly among low-income populations.

Researchers have long pointed out that traditional constructs which dichotomize pregnancy intention leave little room to capture women’s complex and even contradictory thoughts and feelings about pregnancy.^13–15 Consistent with a recent study by Aiken and Potter, we found that women sometimes had incongruent intentions and feelings about pregnancy, and that this incongruence did not seem to reflect ambivalence.¹⁹ Ambivalence has received much attention, especially in adolescent populations, as a relatively common phenomenon that may undermine consistent contraceptive use.^20–25 We did not find that many women in our sample, all of whom were over age 18 and many of whom had children already, expressed ambivalence, although it has been suggested that contraceptive risk-taking is a manifestation of subconscious ambivalence.²⁶ However, over 50% of our pregnant cohort terminated or were planning to terminate their pregnancies, suggesting that most of the pregnancies that occurred in contexts of contraceptive nonuse or inconsistent use were actually unacceptable. Instead, contraceptive-risk taking in our sample seemed to be driven more by perceptions of low reproductive control. Similar sentiments, sometimes called fatalism, have been described among socially disadvantaged women in other studies.^27–30 For many women in our cohort, pregnancies “happened” and women subsequently determined whether or not they were acceptable based on an internal assessment of their individual and social capital at that particular time.

Planning for pregnancy was a related but distinct concept from pregnancy intention or desire. Even beyond factors stemming from low reproductive control, planning was not a particularly salient concept often because the context in which women felt planning should take place (marital relationship and stable finances) was elusive. Although childbearing largely occurred outside of wedlock in participants’ social networks, women still placed tremendous value on marriage - a phenomenon that has also been described by sociologist Kathryn Edin in her seminal work on family formation in low-income communities.³¹ Furthermore, as the theoretical advantages ascribed to timing a pregnancy were based on educational and/or job opportunities, which were also limited, the benefits of planning pregnancy timing were not accessible. As women did not acknowledge the health benefits to either mother or infant of a planned pregnancy, the inherent value of planning and preparing for a pregnancy was seemingly not evident. How to best engender planning salience within women’s particular psycho-social context deserves some attention to help women access the benefits of optimizing their health behaviors prior to conception. Perhaps abandoning the term “planning” and instead helping women to “best prepare for whatever might happen” may be one strategy.

Even when women intended to avoid pregnancy, their behaviors were not necessarily congruent with these intentions. We identified several clinically-relevant mechanisms that helped to explain these incongruences, including perceptions of low susceptibility to pregnancy and male partner reproductive coercion. Furthermore, these specific pathways to unintended pregnancy emerged more commonly among AA women. Additional population-level research is warranted to determine whether these factors contribute to observed disparities in unintended pregnancy.

Over 40% of women who were pregnant or recently pregnant reported believing they could not get pregnant. Our findings are consistent with other recent data which have identified perceptions of low fecundity as a common reason for contraceptive non-use.^32–34 As in this study, misperceptions about personal pregnancy risk seem to arise from having had previous unprotected intercourse without conception and thus assuming sub-fecundity rather than from beliefs that unprotected sex is a low-risk activity. Thus, counseling strategies that query and address women’s misperceptions about subfertility may represent an opportunity to help women reduce their personal risk of unintended pregnancy.

We also found that a substantial number of women in our study reported experiences with reproductive coercion. This finding is consistent with recent reports documenting the prevalence of this phenomenon and its link with unintended pregnancy.^35–39 As providers have primarily focused on women’s behaviors, these findings highlight the need to probe about male partner’s reproductive intentions and consider the possibility that reproductive coercion may be undermining women’s contraceptive efforts. The American College of Obstetrics and Gynecology (ACOG) recently released a report recommending that providers routinely screen women for reproductive coercion and counsel on harm reduction strategies, including use of “hidden” methods of contraception that cannot be detected by male partners such as an intrauterine device, contraceptive injection, or subdermal implant. ³⁸

There are important limitations to consider in interpreting our findings. First, qualitative research is used to explore concepts from participants’ perspectives in significant depth. Given the extensive length of the transcripts and the complexity of the analysis, sample sizes are generally small, thereby limiting the ability to capture prevalence on a population level. However, such narratives are critical to understanding the psycho-social contexts that influence women’s pregnancy-related behaviors and help to ensure that providers can meet women “where they are.” Second, confronting a pregnancy may have shaped some women’s retrospective accounts of their preconception behaviors. For example, we found more women in the pregnant (or recently pregnant) sub-cohort reported beliefs of subfertility. It is unclear whether pregnancy, however, provided more clarity about preconception attitudes or whether women were trying to rationalize contraceptive non-use. However, a strength of this study is that we included perspectives from women at different reproductive stages (pregnant, recently pregnant, and sexually active but not pregnant).

How can we hope to achieve our public health goal of increasing the proportion of pregnancies that are consciously desired and planned for when these concepts do not necessarily resonate with many of our patients? Recognizing that poverty and social inequality can undermine women’s perceptions of self-determination and upward social mobility and thus pregnancy planning and contraceptive use is a start. In the era of patient-centered outcomes research, our findings also raise questions about the appropriateness of continuing to focus on unintended pregnancy as a primary outcome of interest. If there are populations of women whose social realities do not support the idea of a consciously desired and planned pregnancy and who are often happy when confronted with what public health practitioners would label an unintended pregnancy, do we need to consider realigning our goals with those that women find most relevant and meaningful? Perhaps more work is needed to develop measures that can more accurately identify those pregnancies that will pose an unacceptable risk to women’s quality of life or health.

Implications.

This study identified several cognitive and interpersonal pathways to unintended pregnancy among low-income women in Pittsburgh, PA including perceptions of low reproductive control, perceptions of low fecundity, and male partner reproductive coercion.