Women's Health Issues 24-1 (2014) e131–e138

www.whijournal.com

Original article

The Role of Intimate Partners in Women’s Reasons for SeekingAbortion

Karuna S. Chibber, DrPH a,*, M. Antonia Biggs, PhD a,b, Sarah C.M. Roberts, DrPH a,Diana Greene Foster, PhD a

aBixby Center for Global Reproductive Health, Advancing New Standards in Reproductive Health (ANSIRH), University of California San Francisco,Oakland, CaliforniabBixby Center for Global Reproductive Health, Philip R. Lee Institute for Health Policy Studies, University of California San Francisco,San Francisco, California

Article history: Received 21 December 2012; Received in revised form 20 October 2013; Accepted 21 October 2013

a b s t r a c t

Purpose: The literature on partners and abortion focuses on intimate pa

rtner violence (IPV) and risk for abortion, andpartners’ control of women’s abortion decisions. This paper examines how partners figure in women’s abortion de-cisions, and identifies factors associated with identifying partner as a reason (PAR) for abortion.Methods: Baseline data were used from the Turnaway Study, a longitudinal study among women (n ¼ 954) seekingabortion at 30 U.S. facilities between 2008 and 2010. Mixed methods were used. Data were analyzed using thematiccoding and logistic regression.Findings: Nearly one third of women reported PAR for abortion. Three most common partner-related reasons were poorrelationships, partners unable/unwilling to support a baby, and partner characteristics that made them undesirable tohave a baby with. Eight percent who mentioned PAR identified having abusive partners as a reason for abortion. Onewoman in this subgroup reported being pressured by her partner to seek abortion, whereas others in this subgroupsought abortion to end abusive relationships or to avoid bringing children into abusive relationships. Factors associatedwith identifying PAR for seeking abortion included race, education, partner’s pregnancy intentions, relationship withman involved in the pregnancy, and experience of IPV.Conclusion: Women make decisions to terminate pregnancies considering the quality of the relationship with andpotential support they will receive from the man involved. Even women who report IPV, who may be vulnerable tocoercion, report their motivation for the abortion is to end an abusive relationship, rather than coercion into abortion.

Copyright � 2014 by the Jacobs Institute of Women’s Health. Published by Elsevier Inc.

There is increasing recognition in the public health commu-nity of men’s influence inwomen’s reproductive health decisionsand outcomes. Male partners may control women’s contracep-tive use (Miller et al., 2010; Moore, Frohwirth, & Miller, 2010;Thiel de Bocanegra, Rostovtseva, Khera, & Godhwani, 2010) andcoerce them to become pregnant (Miller et al., 2010). Maleperpetuation of intimate partner violence (IPV) may increasewomen’s risk for adverse reproductive health outcomes(Campbell, Woods, Chouaf, & Parker, 2000), including the like-lihood of experiencing an unintended pregnancy (Uscher-Pines

* Correspondence to: Karuna S. Chibber, DrPH, Public Health Social Scientist,University of California San Francisco, Bixby Center for Global ReproductiveHealth, 1330 Broadway, Suite 1100, Oakland, CA 94612. Phone: 510-986-8963;fax: 510-986-8960.

E-mail address: karuna.chibber@gmail.com (K.S. Chibber).

1049-3867/$ - see front matter Copyright � 2014 by the Jacobs Institute of Women’http://dx.doi.org/10.1016/j.whi.2013.10.007

& Nelson, 2010). A growing body of literature also shows thatmale partners can support women’s effective contraceptive useand childbearing decision making (Kraft et al., 2010; Schwartz,Brindis, Ralph, & Biggs, 2011).

One area that remains less explored is partners’ influence inwomen’s abortion decision-making. Studies in the United Statesexamining women’s reasons for seeking abortion have touchedon the issue of partner influence (Faria, Barrett, & Goodman,1985; Finer, Frohwirth, Dauphinee, Singh, & Moore, 2005;Glander, Moore, Michielutte, & Parsons, 1998; Kirkman, Rowe,Hardiman, Mallett, & Rosenthal, 2009; Santelli, Speizer, Avery,& Kendall, 2006; Torres & Forrest, 1988). Torres and Forrest(1988) found that half of abortion patients surveyed across 30U.S. facilities identified relationship problems or not wanting tobe single parents as their reason for seeking abortion. In a studywith patients from 11 abortion facilities across the United States,

s Health. Published by Elsevier Inc.

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138e132

Finer and associates (2005) also found that relationship prob-lems or wanting to avoid single motherhoodwere ranked amongwomen’s top reasons for seeking abortion and reported by nearlyhalf the sample. These two large, quantitative studies listed someof the open-ended responses that women gave about theirpartners, for example, having a partner who did not want to getmarried, being afraid of breaking up with a partner, or beingmistreated or abused by partners.

Much of the literature relating to partners and abortion fo-cuses on the association between IPV and risk of abortion, andmale control of women’s abortion-related decisions. Forexample, research shows that women experiencing IPV are morelikely to have an induced and/or repeat abortion compared withwomen not experiencing IPV (Coker, 2007). Evidence supportingthe idea that men control women’s abortion decisions or coercethem to have abortions comes from three qualitative studieswith women with histories of IPV (Coggins & Bullock, 2003;Hathaway, Willis, Zimmer, & Silverman, 2005; Moore et al.,2010). In two of these studies, the authors also described howsome study participants explicitly talked about IPV; for example,they described being attacked by their partners in efforts tocause spontaneous abortions or threatened with violence if theydid not terminate the pregnancy (Coggins & Bullock, 2003;Moore et al., 2010). The few quantitative studies on this topicfind that coercion is rare in the general abortion patient popu-lation (Finer et al., 2005; Foster, Gould, Taylor, & Weitz, 2012).

The political climate regarding abortion in the United States,coupled with the economic recession might be affectingwomen’s reasons for seeking abortion, warrant a new investi-gation into women’s reasons for seeking abortion. Data from theTurnaway Study, a longitudinal study among women seekingabortion at 30 U.S. facilities, have been analyzed for women’sreasons for seeking abortion more broadly (Biggs, Gould, &Foster, 2013), and how women describe alcohol, tobacco, and/or drug use as reasons for seeking abortion (Roberts, Avalos,Sinkford, & Foster, 2012). In this paper, we focus exclusively onhow women describe their partners’ influence in their abortiondecision making. By using quantitative data and data from twoopen-ended questions rather than a checklist of researcher-generated reasons, we aim to provide a more in-depth andrecent examination of how women themselves describe theirpartners’ influence in their abortion decision making, subse-quently termed partner as a reason (PAR) for seeking abortion.We present the characteristics of women mentioning PARcompared with women mentioning other reasons for seekingabortion, and identify factors associatedwithmentioning PAR forseeking abortion.

Methods

Study Setting and Procedures

This mixed methods study uses baseline data from theongoing Turnaway Study, a prospective longitudinal study thataims to understand the effect of abortion on women’s lives.Detailed information about study sites and recruitment pro-cedures have been published elsewhere (Foster, Kimport, Gould,Roberts, & Weitz, 2013; Gould, Perrucci, Barar, Sinkford, & Foster,2012). Briefly, between January 2008 and December 2010,women were recruited from 30 U.S. abortion facilities where nofacility nearby offered care at a later gestational age of preg-nancy. At each facility, staff were trained by Turnaway Studyresearchers to conduct recruitment.

Study eligibility included being pregnant, English or Spanishspeaking, 15 years or older, not having any known fetal anoma-lies or demise, and belonging to one of the following threegroups: 1) Women just over the facility’s gestational age limitand denied an abortion (n ¼ 231), 2) women just under the limitandwho received an abortion (n¼ 452), and 3) women receivinga first trimester procedure (n ¼ 273). After assessing eligibility,facility staff informed potential participants about the study andconnected themwith Turnaway Study researchers by telephone.The researchers explained the study in greater detail, screenedfor eligibility, and obtained informed consent. After verballyagreeing to participate, each woman signed a written consentform faxed by facility staff to a, private dedicated fax machine atthe Turnaway project director’s office. Of all eligible participantsapproached, 37.5% (n ¼ 1,132) agreed to participate in semi-annual telephone interviews for 5 years. Of those who con-sented, 85% (n ¼ 956) completed the baseline telephoneinterview, which took place approximately 1 week after womensought the abortion and lasted an average of 40 minutes. Studyinterviewers were female, fluent in English and/or Spanish, andexperienced in reproductive health research and interviewingtechniques. They received extensive training on data collectionprotocols, handling sensitive issues, how to encourage accuratereporting, and record keeping. Data quality assurance checkstook place routinely. The structured interview guide includedquestions on participant sociodemographics, experiencesbecoming pregnant, and abortion decision-making process. Theinterview guide and all study protocols were pilot tested among64 women receiving or being denied abortions at a local abortionfacility. All study protocols have been approved by the Universityof California, San Francisco, Committee for Human Research.

This paper uses quantitative and qualitative data from thebaseline interview, and combines the three study groups foranalysis. The total sample includes 954 women because 2 par-ticipants did not provide reasons for seeking abortion and wereexcluded from the analysis.

Measures

OutcomeWomen were asked the following open-ended question:

“What are some of the reasons you decided to have an abortion?”Women providing more than one reasonwere asked, “What wasthe main reason you decided to have an abortion?” Responses toboth questions were examined because participants were oftenunable to isolate a main reason, provided more than one mainreason, or provided new reasons in response to the secondquestion. Women who referred to their partners whendescribing their reasons for seeking abortion were classified asreporting PAR for seeking abortion.

PredictorsSociodemographic factors included age (15–19, 20–24,

and �25 years), race/ethnicity (White, Black, Hispanic/Latina,and other), education (less than high school versus more thanhigh school), and poverty level (household income below 100%federal poverty level [FPL], 100–200% FPL, >200% FPL, andpoverty missing). Relationship status with the man involved inthe pregnancy 1 week after seeking abortion was classified as norelationship, friend/acquaintance, boyfriend/partner, and fianc�e/husband. The first category, no relationship, included womenreported no current relationship with the man involved inthe pregnancy, described him as an ex-boyfriend, ex-husband/

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138 e133

separated, baby daddy, or father of her child/children. Experi-encing IPV from the man involved in the pregnancy was definedas having experienced physical violence (being “pushed, hit,slapped, kicked, chocked, or physically hurt in any way”), orpsychological violence (being “frightened for safety as a result ofanger or threats made”) or rape (reporting that pregnancy “was aresult of rape” [yes/no]). Other variables included parity (0, 1,or �2 children), partner’s pregnancy intentions (partner did notwant pregnancy, did not know about pregnancy, was not sureand left it up to woman to decide, and wanted pregnancy),women’s feelings regarding pregnancy (indifferent, happy, orvery happy versus somewhat unhappy or very unhappy), diffi-culty deciding to have an abortion (neither difficult nor easy,somewhat easy, or very easy versus somewhat or very difficult),and gestational age discovered pregnancy in weeks. Note weincluded gestational age pregnancy discovered instead ofgestational age at the time of recruitment because the latter wasnot significantly associated with mentioning PAR as for seekingabortion (Biggs et al., 2013).

Analysis

Qualitative datawere analyzed by two authors, onewith a PhDin psychology and one with a doctorate in public health. Bothauthors independently reviewed the first 100 responses, con-ducting a line-by-line analysis of each response reflecting on thefollowing questions: “What is going on here?” “What is thisabout?” and “How is it similar or different to preceding re-sponses?” (Strauss & Corbin, 1998). Using this approach, the twoauthors independently developed a non-hierarchical list ofthemes emerging directly from the data. An analytical meetingwas then held to discuss these themes and their meanings toensure consistent and accurate interpretation, and then jointlydevelop a list of final themes pertaining to howwomen describedtheir partner’s influence on their abortion decision making. Bothauthors then independently coded all responses using the finalset of themes. Differences of opinion between authors regardingthemes and their applicationwere reconciled bygoing back to thedata and through ongoing discussion until reaching consensus onall responses. All coding was done in Microsoft Excel.

Mixed effects logistic regression was used to analyze quanti-tative data. First, mixed effects logistic regression was conductedfor each predictor variable to understand the characteristics ofwomenmentioning PAR for seeking abortion.Multivariablemixedeffects logistic regression followedto identify significantpredictorsofmentioningPARfor seekingabortion.All covariates significant inthe unadjusted mixed effects logistic regression analysis (p < .05)were included in the multivariable analysis. All quantitative ana-lyses account for clustering by recruitment site and were con-ducted using Stata 12 (StataCorp, Inc., College Station, TX).

Results

Women’s Description of Their Partner-Related Reasons forAbortion

Just under one third (31%) of the 954 women in the TurnawayStudy gave partner-related reasons for seeking abortion. Wefound six distinct partner related themes. Most women citingPAR gave responses that fell under one theme; 22% articulatedmore than one of the six themes. The denominator of the per-centages below is the total number of women mentioning apartner related reason (n ¼ 298).

1. No, poor, or new relationships with partner:Of the 298womenwho identified PAR for seeking abortion, 31% describedrelationship problems: They no longer had a relationshipwith the partner or their relationship was poor or new.

I was no longer with the person so I thought it was best to justdo that [get an abortion]. (19 year old, no relationship withman involved in the pregnancy).

Others, although currently cohabiting, described being un-happy in their relationships as the reason for seeking abor-tion. For example, one respondent explained that she was stilltrying to work out the differences with her partner, andhaving children would not “make the situation better.” A fewwomen said that they were in new relationships, and concernabout the sustainability of their relationships led them to seekabortions.

Me and my boyfriend have not been together long enough forthat [having a baby]. (34 year old, man involved in thepregnancy is a boyfriend/partner).

2. Partner will not or cannot support the woman to have a baby:Twenty-six percent of women said that their partner wasunableorunwilling to support them inhaving ababy. Reasonsfor feeling this way included the following: Partner was notphysically around to support the woman and care for a child,including situationswherepartnerswere in jail or hadmovedaway for professional/educational reasons; partner was notfinancially able to provide for a child; partner was “not readyto become a father” or not “responsible enough” to raise achild; and partner did not support thewoman in raising theirexisting children and was therefore unlikely to support herwith a new baby. Thirty-nine percent of women identifyinglackof partner support inhaving ababyalso emphasized theirown financial inability to raise a child at the time.

Financially I wasn’t ready and neither was he. (24 year old,man involved in the pregnancy is a boyfriend/partner).

I have not finished with school. I plan on going back. I don’tbelieve I make enough money to raise a child and the personthat I had the baby with wouldn’t have been able to help me.(25 year old, man involved in the pregnancy is a boyfriend/partner).

3. Partner is the “wrong guy” to have a baby with: Twenty-onepercent of women identified specific characteristics of thepartner that made him the “wrong guy” to have a baby with.These included describing the partner as not the kind ofperson with whom the participant envisioned a future orwith whom she wanted a baby, married to someone else,already a father from another relationship, someone otherthan the participant’s husband, someone other than the fa-ther of the participant’s other children, using drugs, or in jail.

Because I didn’t see myself with him [in the long term].(25 year old, man involved in the pregnancy is a friend/acquaintance).

He was a felon and I didn’t want to breed a criminal. (28 yearold, no relationship with man involved in the pregnancy).

Some participants emphasized that they did not want to havea baby with the “wrong guy” because they would have tocontinue their association with these men.

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138e134

I didn’t want to be tied to the father for the next 18 years.(40 year old, no relationship with man involved in thepregnancy).

4. Partner does not want baby: Ten percent said that theirpartner did not want the baby. Only one respondent usedlanguage explicitly suggesting overt pressure from herpartner to get an abortion.

The father was pressuring me. (30 year old, no relationshipwith man involved in the pregnancy).

Most, however, said their partners did not want the baby orthat they were not interested in having a baby, not alluding tocoercion.

The guy that I was with was acting like he didn’t care. (29 yearold, no relationship with man involved in the pregnancy).

I didn’t want to have another child without my husband’sblessing. (37 year old, man involved in the pregnancy is herhusband).

Many respondents who said that their partner did not wantthe baby also described personal constraints.

I had an abortion because the guy I got pregnant by didn’twant the baby plus I couldn’t afford another child.(24 year old, man involved in pregnancy is a friend/acquaintance).

Partner I am involvedwith does notwant another child.Wrongtime in life [for me] to be reliable, or to raise a baby. (31 yearold, man involved in the pregnancy is a boyfriend/partner).

A few participants indicated that they had mutually decidedwith their partners to seek an abortion.

Have a daughter who is 14 years old. We only wanted onechild. (35 year old, man involved in the pregnancy is aboyfriend/partner).

5. Partner is abusive: Eight percent of women said that theysought abortions because they had abusive partners. Abuseincluded being hit, threatened, mentally harassed, or raped.One participant said she had been forced into becomingpregnant. Some participants explained that they were con-cerned the abuse would persist even after childbirth, andthey did not want to raise children in abusive environments.

Our relationship became violent and I couldn’t see bringinganother kid into a life that was going to be surrounded byviolence. (40 year old, no relationship with man involved inthe pregnancy).

Others explained that having a baby would be a deterrent toending an abusive relationship.

I was trying to leave an abusive relationship and I didn’t wanthim to have any ties. (38 year old, no relationship with maninvolved in the pregnancy).

6. Respondent wants to be married before having a child or doesnot want to be single mom”: Twenty-seven percent wanted tobe married before having children or wanted to raise chil-dren in two-parent households. Five percent specificallymentioned wanting to be married before having a baby.

I just wouldn’t have a baby without being married. (33 yearold, man involved in the pregnancy is a friend/acquaintance).

Twenty-two percent did not to want to be single parents.The majority in this subsample explained that, because oftheir current partner, they would have to raise the child ontheir own.

He didn’t have a job or a driver’s license and wasn’t reallysticking around. Just felt like it would be another kid to raiseon my own. (30 year old, man involved in the pregnancy is afriend/acquaintance).

Some women also expressed fear and anxiety at the prospectof single-handedly raising a child.

Because I was scared to be alone with three kids.just scared.(28 year old, no relationship with man involved in thepregnancy).

I can’t fathom raising two babies on my own. (24 year old,man involved in pregnancy is a boyfriend/partner).

Others also emphasized not wanting to take on the burden ofsingle parenting.

With my husband out of the country, I am doing everythingon my own. All of it is just too overwhelming. (28 year old,man involved in pregnancy is her husband)

A few women articulated how being a single parent was nottheir preference about how to raise children.

Kids turn out a whole lot better when they are raised in ahouseholdwith amother and a father. They don’t just becomeanother statistic. In my heart it is real important that thebiological father is there. (25 year old, no relationship withman involved in the pregnancy).

Characteristics and Predictors of Women Identifying PAR forSeeking Abortion

Table 1 presents characteristics of womenmentioning PAR forseeking abortion (also see Table 1 for total sample distribution bypredictors). More women age 25 years and older identified PARfor seeking abortion than 15 to 19 year olds (37% vs. 24%). Morewomen reporting a household income above 200% FPL identifiedPAR for abortion thanwomenwith incomes below 100% FPL (41%vs. 30%). Fewer women in relationships with theman involved inthe pregnancy identified PAR for seeking abortion than womenwho were not in a relationship (22% vs. 51%). Women who wereexperiencing IPV from the man involved in the pregnancy weremore likely to report PAR thanwomenwithout IPV (62% vs. 28%).Of those reporting IPV, 23% said that having an abusive partnerwas their reason for seeking abortion, and 40% described otherpartner-related reasons for seeking abortion (data not shown).Fewer women whose partners were unsure about keeping thepregnancy mentioned PAR for abortion compared with womenwhose partners did not want the pregnancy (23% vs. 36%).

Table 2 presents results from the multivariable analysis. Ageand poverty were no longer associated with identifying PAR forabortion after controlling for other covariates. The odds ofidentifying PAR for seeking abortion was higher among moreeducated compared with less educated women (odds ratio [OR],

Table 1Characteristics of Women Mentioning Partner as a Reason (PAR) for Seeking Abortion

Variable Total Sample(n ¼ 954)

PAR for Seeking Abortion(n ¼ 298)

Other Reasons for Seeking Abortion(n ¼ 656)

p Value

Age (y)15–19* (reference) 18% 24% 76%20–24 36% 28% 72%�25 46% 37% 63% z

EducationLess than high school (reference) 53% 26% 74%More than high school 47% 37% 63% x

Race/ethnicityWhite (reference) 37% 38% 62%Black 29% 24% 76% xHispanic/Latina 21% 30% 70% kOther races 13% 31% 69%

Poverty (%FPL)<100 (reference) 34% 30% 70%100–200 22% 35% 65%>200 13% 41% 59% yMissing 32% 26% 74%

Relationship with man involved in the pregnancyNo relationship (reference) 27% 51% 49%Friend 14% 37% 63% zPartner/boyfriend 45% 21% 79% xFianc�e/husband 14% 22% 78% x

Experienced IPV from man involved in the pregnancy in the past yearNo (reference) 91% 28% 72%Yes 9% 62% 38% x

Partner pregnancy intentionsDid not want pregnancy (reference) 25% 36% 64%Did not know about pregnancy 17% 42% 58%Not sure and left it up to woman to decide 36% 23% 77% zWanted pregnancy 22% 31% 69%

Parity0 children (reference) 38% 33% 67%1 child 29% 35% 65%�2 children 34% 27% 73% kGestational age pregnancy discovered (mean weeks) 9 9 10 k

Difficulty deciding whether to get abortionNot difficult nor easy, somewhat easy, or very easy (reference) 44% 28% 72%Somewhat difficult or very difficult 56% 34% 66% k

Feelings regarding pregnancyIndifferent, happy, or very happy (reference) 45% 31% 69%Somewhat unhappy or very unhappy 55% 30% 70%

Abbreviations: FPL, federal poverty level; IPV, intimate partner violence.p Values were derived frommixed effects logistic regression analysis conducted for each predictor variable to understand the characteristics of womenmentioning PARfor seeking abortion.

* This age category includes one participant aged 14 who was recruited early in the study before the minimum enrollment age was changed to 15.y p < .05.z p < .01.x p < .001.k p < .10.

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138 e135

1.55; 95% confidence interval [CI], 1.12–2.15). African-Americanwomen were less likely to identify PAR for seeking abortioncompared with White women (OR, 0.53; 95% CI, 0.36–0.78).Women in a relationshipwith theman involved in the pregnancywere less likely to mention PAR for abortion compared withwomen not in a relationship (man involved in the pregnancy wasa friend/acquaintance [OR, 0.63; 95% CI, 0.40–0.99]; boyfriend/partner [OR, 0.27; 95% CI, 0.19–0.39]; fianc�e/husband [OR, 0.24;95% CI, 0.14–0.40]). Women reporting IPV from the man involvedin the pregnancy were more than three times as likely to identifyPAR for abortion thanwomen not reporting IPV (OR, 3.53; 95% CI,2.12–5.88). In relation to partners’ pregnancy intentions, womenwhose partners were unsure about keeping the pregnancy andleft it up to the woman to decide had a lower odds of identifyingPAR for seeking abortion compared with those whose partnersdid not want the pregnancy (OR, 0.61; 95% CI, 0.41–0.90).

Discussion

This study focused on women’s reports of their partner’s in-fluence in their abortion decision making. About one third of thesample identified partner-related reasons for seeking abortion.The three most commonly reported reasons were no, poor, ornew relationships; partners who were unable/unwilling to sup-port women in having a baby; and partner characteristics thatmade them undesirable to have a baby with. Factors associatedwith identifying PAR for seeking abortion included race, educa-tion, partner’s pregnancy intentions, relationship with maninvolved in the pregnancy, and experience of IPV.

Study findings confirm and build on previous research find-ings about the importance of intimate relationships in women’sabortion decision making (Faria et al., 1985; Finer et al., 2005;Santelli et al., 2006; Torres & Forrest, 1988). Findings suggest

Table 2Factors Associated with Identifying PAR for Seeking Abortion (n ¼ 954)

Variable Odds Ratio 95% CI

Age (y)15–19 1.0020–24 0.91 0.57–1.48�25 1.31 0.81–2.12

Race/ethnicityWhite 1.00Black 0.53* 0.36–0.78Hispanic/Latina 0.78 0.52–1.19Other 0.71 0.43–1.17

EducationLess than high school 1.00More than high school 1.55* 1.12–2.15

Poverty (%FPL)<100 1.00100–200 1.26 0.84–1.90>200 1.54 0.94–2.53Missing 0.83 0.55–1.24

Relationship with man involved in the pregnancyNo relationship 1.00Friend/acquaintance 0.63* 0.40–0.99Partner/boyfriend 0.27* 0.19–0.39Fianc�e/husband 0.24* 0.14–0.40

Experienced any IPV (physical/psychological/rape) fromman involved in thepregnancyNo 1.00Yes 3.53* 2.12–5.88

Partner pregnancy intentionsDid not want pregnancy 1.00Did not know about pregnancy 0.89 0.56–1.41Not sure and left it up to woman to decide 0.61* 0.41–0.90Wanted pregnancy 0.86 0.55–1.34

Abbreviations: FPL, federal poverty level; IPV, intimate partner violence.p Values were derived from multivariable mixed effects logistic regressionanalysis.

* p < .05.

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138e136

that, when women mention relationship problems as a reasonfor seeking abortion, they are mostly reflecting on whether theirrelationship is one in which they can foresee having a child atthis time. The odds of identifying PAR for seeking abortion weregreater among women reporting no relationship with the maninvolved in the pregnancy. Further, women described how hav-ing no relationship; being in a poor quality, unstable, or newrelationship, where long-term commitment was questionable;and being in a relationship where financial and logistical supportfrom their partners in childrearing was unlikely, as importantreasons for seeking abortion. They also highlighted partnercharacteristics that ruled out long-term relationships, such aspartners who were married to someone else or partners whowere in jail.

Also consistent with previous literature (Finer et al., 2005)was the finding that only few women (about 2.5% of all studyparticipants and 8% of participants mentioning PAR for seekingabortion) identified having abusive partners as a reason forseeking abortion. Importantly, those identifying abusive partnersas a reason for seeking abortion did not describe their partner asthreatening or physically hurting them as a way of forcing themto have an abortion. Instead, their descriptions suggest that thewomen independently decided to seek abortions, perceiving thisas their best option to end abusive relationships. Womenexpressed concerns that the abuse would continue even afterchildbirth, and that they would be unable to leave abusive re-lationships if there was a child involved. As has been previouslydocumented in the literature, women also emphasized the

importance of not raising children in abusive environments(Silverman et al., 2010; Williams & Brackley, 2009).

Although few women reported having abusive partners as areason for seeking abortion, IPV was not uncommon among thissample, with nearly 10% reporting IPV in the past year. This isconsistent with previous studies (Jones, Moore, & Frohwirth,2011; Woo, Fine, & Goetzl, 2005). Among those reporting expe-riencing IPV, fewer than half mentioned partner-related reasonsfor seeking abortion and fewer than one quarter mentionedabusive partners in particular. One explanation for why not allwomen experiencing IPV mention abusive partners as a reasonfor abortion could be that when asked about reasons for abortionwomen reflect more broadly on their relationships and life cir-cumstances rather than singling out specific traumatic experi-ences. For example, they may think about what they want fromthe father of their child, or when they might like to have child.

Another finding consistent with previous research (Fineret al., 2005; Torres & Forrest, 1988), was that not wanting to bea single parent was an important reason for seeking abortion.Importantly, women’s own descriptions of their partner’s influ-ence in their abortion decision making suggests that this senti-ment was closely related to the man involved in the pregnancy,with women not seeing this specific individual as the appro-priate person to have a child with, and their not wanting to takeon parenting responsibilities alone.

Finally, it is worth noting that partners’ influence onwomen’sabortion decision making was not necessarily adversarial.Although only one participant mentioned being pressured by herpartner to have an abortion, others described joint decisionmaking with their partners. Further, in many cases even whenpartners were mentioned as a reason for seeking abortion,women also described personal and financial constraints inhaving a child at the time. The finding from the multivariableanalysis that women whose partners’ were unsure about theirpregnancy intentions were less likely to mention PAR furthersubstantiates the fact that partners’ influence was not necessaryadversarial.

These findings should be viewed in light of study limitations.First, fewer than 40% of women who were eligible andapproached agreed to participate. Our sample is therefore notrepresentative of all women seeking abortion. Prospective cohortstudies which are lengthy, burdensome, and offer no directbenefit to participation, such as the Turnaway Study, which re-quires biannual interviews for 5 years, often suffer from lowerrecruitment rates and higher loss to follow-up than other studydesigns (Ejiogu et al., 2011; Rothman & Greenland, 2008).Moreover, this study’s response rate of 37.5% is comparable withother, similar studies. As noted in a 2006 review of 355 publishedobservational studies, 63% of prospective cohort studies did notreport on study recruitment and among those that did theparticipation rate ranged from as low as 20% to as high as 100%(Morton, Cahill, & Hartge, 2006). Additionally, this study’s sam-ple demographics, with the exception of overrepresentation ofwomen beyond the first trimester, closely mirror the de-mographics of women seeking abortion in the United States(Jones & Finer, 2012; Jones & Kavanaugh, 2011). Oversamplingwomen seeking abortions at later gestational ages allows us tostudy and make inferences about a previously understudiedgroup. Second, this paper relies on responses to two open-endedquestions on women’s reasons for abortion. Without the op-portunity to probe, as with in-depth qualitative interviews, itwas difficult to determine if the partner was the root cause ofwhy women sought abortions, or whether there were other,

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138 e137

equally important considerations, all of which together influ-enced women’s abortion seeking. A related issue is that manywomen mentioned multiple reasons for seeking abortion, and itwas difficult to isolate the reasons that pertained to partnersalone. Partner-related problems could have triggered otherchallenges and women may have emphasized the latter whiledescribing their reasons for abortion. For example, women whoreported financial difficulties as their reason for abortion and thiscould stem from the absence of a supportive partner but wouldnot be captured as such in this paper. Finally, IPV tends to beunderreported in research (Evins & Chescheir, 1996; Garcia-Moreno, Jansen, Ellsberg, Heise, & Watts, 2005). We tookseveral measures to create a safe environment to encourage ac-curate IPV reporting. The IPV rates reported in this study arecomparable with previous studies conducted among womenseeking abortions (Jones & Kavanaugh, 2011; Woo et al., 2005).

Implications for Practice

Partners play a role in women’s abortion decision making.Our study sample, however, shows no evidence that mostwomen lack control in their abortion decision making. Instead,when deciding to get an abortion, women in our study seem tobe reflecting more broadly on the complexities of their intimaterelationships, and their lives, values, and vulnerabilities as awhole. Even some women in abusive relationships seem to seekabortion as a way of ending abusive relationships rather thanbeing motivated by fear of partner’s threats or violence if theykeep the pregnancy. Future qualitative research with the op-portunity to engage in deeper and more meaningful discussionswith women seeking abortion is needed. From a programperspective, health care providers need to be aware of the pos-sibility of IPV among women seeking abortion and the fact thatsome women may be seeking abortion as a way to escapeabusive relationships. Providers should be prepared to discusswomen’s experiences of IPV. Integrating IPV identification withroutine abortion and family planning services may enable pro-viders to offer women seeking abortion and reporting IPV withinformation about safety and referrals to counseling and othersupport services.

Acknowledgments

The authors thank Rana Barar, Heather Gould, and SandyStonesifer for study coordination and management; Janine Car-penter, Undine Darney, Ivette Gomez, Selena Phipps, ClaireSchreiber, and Danielle Sinkford for conducting interviews;Michaela Ferrari and ElisetteWeiss for project support; Jay Fraserand John Neuhaus for statistical and database assistance; and allthe participating providers for their assistance with recruitment.

This study was supported by research and institutional grantsfrom the Wallace Alexander Gerbode Foundation, the David andLucile Packard Foundation, the William and Flora HewlettFoundation and an anonymous foundation.

References

Biggs, M. A., Gould, H., & Foster, D. G. (2013). Understanding why women seekabortion in the US. BMC Women’s Health, 13, 29.

Campbell, J. C., Woods, A. B., Chouaf, K. L., & Parker, B. (2000). Reproductivehealth consequences of intimate partner violence. A nursing research re-view. Clinical Nursing Research, 9(3), 217–237.

Coggins, M., & Bullock, L. F. (2003). The wavering line in the sand: The effects ofdomestic violence and sexual coercion. Issues in Mental Health Nursing,24(6-7), 723–738.

Coker, A. L. (2007). Does physical intimate partner violence affect sexual health?A systematic review. Trauma, Violence & Abuse, 8(2), 149–177.

Ejiogu, N., Norbeck, J. H., Mason, M. A., Cromwell, B. C., Zonderman, A. B., &Evans, M. K. (2011). Recruitment and retention strategies for minority orpoor clinical research participants: Lessons from the healthy aging inneighborhoods of diversity across the life span study. Gerontologist,51(Suppl. 1), S33–S45.

Evins, G., & Chescheir, N. (1996). Prevalence of domestic violence among womenseeking abortion services. Women’s Health Issues, 6(4), 204–210.

Faria, G., Barrett, E., & Goodman, L. M. (1985). Women and abortion: Attitudes,social networks, decision-making. Social Work in Health Care, 11(1), 85–99.

Finer, L. B., Frohwirth, L. F., Dauphinee, L. A., Singh, S., & Moore, A. M. (2005).Reasons U.S. women have abortions: Quantitative and qualitative perspec-tives. Perspectives on Sexual and Reproductive Health, 37(3), 110–118.

Foster, D. G., Gould, H., Taylor, J., & Weitz, T. A. (2012). Attitudes and decisionmaking among women seeking abortions at one U.S. clinic. Perspectives onSexual and Reproductive Health, 44(2), 117–124.

Foster, D. G., Kimport, K., Gould, H., Roberts, S. C., & Weitz, T. A. (2013). Effect ofabortion protesters on women’s emotional response to abortion. Contra-ception, 87, 81–87.

Garcia-Moreno, C., Jansen, H. A. F. M., Ellsberg, M., Heise, L., & Watts, C. (2005).WHO multi-country study on women’s health and domestic violence againstwomen: Initial results on prevalence, health outcomes and women’s responses.Geneva: World Health Organization.

Glander, S. S., Moore, M. L., Michielutte, R., & Parsons, L. H. (1998). The preva-lence of domestic violence among women seeking abortion. Obstetrics andGynecology, 91(6), 1002–1006.

Gould, H., Perrucci, A., Barar, R., Sinkford, D., & Foster, D. G. (2012). Patient ed-ucation and emotional support practices in abortion care facilities in theUnited States. Women’s Health Issues, 22(4), 359–364.

Hathaway, J. E., Willis, G., Zimmer, B., & Silverman, J. G. (2005). Impact of partnerabuse on women’s reproductive lives. Journal of the American MedicalWomen’s Association, 60(1), 42–45.

Jones, R. K., & Finer, L. B. (2012). Who has second-trimester abortions in theUnited States? Contraception, 85(6), 544–551.

Jones, R. K., & Kavanaugh, M. L. (2011). Changes in abortion rates between 2000and 2008 and lifetime incidence of abortion. Obstetrics and Gynecology,117(6), 1358–1366.

Jones, R. K., Moore, A. M., & Frohwirth, L. F. (2011). Perceptions of male knowl-edge and support among U.S. women obtaining abortions. Womens HealthIssues, 21(2), 117–123.

Kirkman, M., Rowe, H., Hardiman, A., Mallett, S., & Rosenthal, D. (2009). Reasonswomen give for abortion: A review of the literature. Archives of Women’sMental Health, 12(6), 365–378.

Kraft, J. M., Harvey, S. M., Hatfield-Timajchy, K., Beckman, L., Farr, S. L.,Jamieson, D. J., et al. (2010). Pregnancy motivations and contraceptive use:Hers, his, or theirs? Womens Health Issues, 20(4), 234–241.

Miller, E., Decker, M. R., McCauley, H. L., Tancredi, D. J., Levenson, R. R.,Waldman, J., et al. (2010). Pregnancy coercion, intimate partner violence andunintended pregnancy. Contraception, 81(4), 316–322.

Moore, A. M., Frohwirth, L., & Miller, E. (2010). Male reproductive control ofwomen who have experienced intimate partner violence in the UnitedStates. Social Science and Medicine, 70(11), 1737–1744.

Morton, L. M., Cahill, J., & Hartge, P. (2006). Reporting participation in epide-miologic studies: A survey of practice. American Journal of Epidemiology,163(3), 197–203.

Roberts, S., Avalos, L., Sinkford, D., & Foster, D. (2012). Alcohol, tobacco and druguse as reasons for abortion. Alcohol and Alcoholism, 47(6), 640–648.

Rothman, K., & Greenland. (2008).Modern Epidemiology. Philadelphia: LippincottWilliams & Wilkins.

Santelli, J. S., Speizer, I. S., Avery, A., & Kendall, C. (2006). An exploration of thedimensions of pregnancy intentions among women choosing to terminatepregnancy or to initiate prenatal care in New Orleans, Louisiana. AmericanJournal of Public Health, 96(11), 2009–2015.

Schwartz, S. L., Brindis, C. D., Ralph, L. J., & Biggs, M. A. (2011). Latina adoles-cents’ perceptions of their male partners’ influences on childbearing:Findings from a qualitative study in California. Culture Health & Sexuality,13(8), 873–886.

Silverman, J. G., Decker, M. R., McCauley, H. L., Gupta, J., Miller, E., Raj, A., et al.(2010). Male perpetration of intimate partner violence and involvement inabortions and abortion-related conflict. American Journal of Public Health,100(8), 1415–1417.

Strauss, A. L., & Corbin, J. M. (1998). Basics of Qualitative Research: Techniques andProcedures for Developing Grounded Theory (2nd ed.). ThousandOaks, CA: Sage.

Thiel de Bocanegra, H., Rostovtseva, D. P., Khera, S., & Godhwani, N. (2010). Birthcontrol sabotage and forced sex: Experiences reported by women in do-mestic violence shelters. Violence Against Women, 16(5), 601–612.

Torres, A., & Forrest, J. D. (1988). Why do women have abortions? Family Plan-ning Perspectives, 20(4), 169–176.

K.S. Chibber et al. / Women's Health Issues 24-1 (2014) e131–e138e138

Uscher-Pines, L., & Nelson, D. B. (2010). Neighborhood and individual-levelviolence and unintended pregnancy. Journal of Urban Health, 87(4), 677–687.

Williams, G. B., & Brackley, M. H. (2009). Intimate partner violence, pregnancyand the decision for abortion. Issues in Mental Health Nursing, 30(4), 272–278.

Woo, J., Fine, P., & Goetzl, L. (2005). Abortion disclosure and the association withdomestic violence. Obstetrics and Gynecology, 105(6), 1329–1334.

Author Descriptions

Karuna S. Chibber, DrPH, is a public health social scientist with Advancing NewStandards in Reproductive Health (ANSIRH) at the UCSF Bixby Center for GlobalReproductive Health. Her research focuses on the intersection of intimate partnerviolence and women’s health.

M. Antonia Biggs, PhD, is a senior researcher at the UCSF Bixby Center for GlobalReproductive Health. Her research focuses on the evaluation of reproductivehealth programs, access to family planning services, abortion, and unintendedpregnancy.

Sarah C.M. Roberts, DrPH, is a public health social scientist at ANSIRH. She studiespolicy and social determinants of women’s health, with a focus on alcohol and druguse in the context of pregnancy, parenting, and reproductive health.

Diana Greene Foster, PhD, a demographer, is the principal investigator of theTurnaway Study, Director of Research at ANSIRH, and an associate professor in theDepartment of Obstetrics, Gynecology and Reproductive Sciences at UCSF.