Abstract
Background
We studied the effect of contraceptive social networking on postabortion intrauterine device (IUD) uptake. This study explores whether women who have heard personal stories of IUD use are more likely to use an IUD for postabortion contraception.
Study Design
We surveyed 299 women undergoing induced abortion at San Francisco General Hospital's Women's Options Center before and after contraceptive counseling. Both English- and Spanish-speaking women, aged 15 years and older, were surveyed.
Results
Fifty percent of women surveyed chose to use an IUD for postabortion contraception. Women choosing IUDs were more likely than women choosing other contraceptives or no contraceptives to be multiparous, Latina and interested in IUDs prior to contraceptive counseling. Disclosure of personal IUD use by a clinic staff member was independently associated with the decision to use an IUD (odds ratio 8.1, 95% confidence interval 3.8–17.2).
Conclusions
Women undergoing abortion in an urban clinic have knowledge and high acceptance of IUDs, and sharing of contraceptive experiences is common among women of all demographics. Controlling for demographics and prior knowledge of IUDs, sharing of personal IUD experiences by providers is significantly associated with IUD use.
Keywords
1. Introduction
Nearly half (49%) of pregnancies in the United States are unintended, and 4 in 10 of these pregnancies result in termination [
[1]
]. Having an abortion is a significant risk factor for having another abortion in the future; 45% of women presenting for abortion have had an abortion in the past [[2]
]. Women having abortions in the United States are often young (56% in their 20s) and economically disadvantaged (57%), and have already had one or more children (61%) [3
, 4
]. Although abortion rates are decreasing overall, unintended pregnancy rates are increasing among women who are economically disadvantaged [5
, 6
].Women have higher rates of contraceptive continuation and lower rates of repeat unintended pregnancy when they use a method of contraception that does not require frequent or pericoital attention [
[2]
]. Long-term, reversible methods of contraception available in the United States are the levonorgestrel intrauterine device (IUD), the copper IUD and the etonorgestrel implant, approved for 5, 10 and 3 years of use, respectively.Intrauterine devices are extremely safe and effective, yet only 5% of women using contraception in the United States choose IUDs [
[7]
]. This is due in part to a history tarnished by the Dalkon Shield, an IUD in the 1970s that was associated with increased rates of pelvic inflammatory disease and infertility [[8]
]. Even now, the stigma associated with IUD use among both patients and providers prevents many women from choosing this highly effective method, despite the fact that IUDs have been proven to be safe and effective in all women, including adolescent and nulliparous women [[9]
]. Additionally, IUDs have been shown to be just as safe and effective to insert immediately postabortion, making the postabortion population an excellent target for increasing IUD use [7
, 10
].A significant barrier to increasing IUD use in the United States is a lack of knowledge among both providers and patients. A study of young pregnant women showed that only half of the women had heard of an IUD, and only 25% of women knew someone using an IUD [
[2]
]. This study seeks to explore women's knowledge of IUDs and influences that affect a woman's decision to select an IUD, particularly the impact of shared personal IUD experiences. The primary aim of this study is to determine the association, if any, that exists between hearing another woman's personal experience with an IUD and making the decision to have an IUD placed for postabortion contraception. If we can better understand the factors that may play a role in a woman's decision to choose an IUD, we will form better strategies for increasing acceptability of IUDs in the future.2. Materials and methods
2.1 Design
This cross-sectional study was approved by the University of San Francisco, California (UCSF) Institutional Review Board (IRB). Patients presenting to the Women's Options Center at San Francisco General Hospital were consecutively recruited from February to June 2009. Each woman received two questionnaires; the first questionnaire was given when the patient first arrived at the Women's Options Center, and the second was given after the abortion procedure. Corresponding surveys were coded in order to avoid a need for patient identifiers. When the first questionnaire was handed out, the corresponding second questionnaire was immediately put in the patient chart to be distributed after the procedure. Surveys were completed anonymously and sealed in envelopes provided with each survey. The study, as approved by the UCSF IRB, did not require written consent from the patients; information sheets describing the study and risks/benefits were distributed with each survey.
2.2 Study participants
Surveys were distributed consecutively to all women aged 15 years and older seeking an abortion for unintended pregnancy, with gestational age less than 23 weeks and 1 day. Surveys were available in both English and Spanish, and women were excluded if they could not read or write either English or Spanish. Women for whom the pregnancy was desired, i.e., women having an abortion because of fetal anomalies or fetal demise, were excluded.
The required sample was determined based on a two-sided α=0.05, β=0.2, effect size 20% and 30% exposure to primary predictors. The calculated sample size of 263 was inflated by 10% to account for the effect of covariate adjustment, resulting in a sample size of 289.
2.3 Measurements
The primary predictors were disclosure of a positive, personal IUD experience by a friend or family member and disclosure of personal IUD use by a counselor or other staff member at the Women's Options Center. The primary outcome was the choice of IUD vs. a different method or no method of contraception. Additional predictors were demographic characteristics (age, race/ethnicity, language, country of origin, years in the United States, education level, income level), sexual/reproductive history (gravidity, parity, past abortions, number of sexual partners, frequency of intercourse), contraceptive plans (past methods used, initial method desired, who helped make contraceptive decisions, comfort level talking to partner about contraception) and an IUD knowledge (previously heard of IUDs, friends/family with IUDs, exposure and reaction to IUD TV commercial). Patients were also asked whether these exposures had a positive effect on their decision-making regarding IUDs.
2.4 Analysis
Bivariate logistic regression was used to determine unadjusted odds ratios (ORs) for predictors of IUD choice. Parity was dichotomized to women who were nulliparous vs. those who were not. This is more clinically relevant as there is a lot of stigma against IUD use for women who have never given birth, despite current literature showing that IUDs are just as safe and effective for nulliparous women and do not increase risk of future infertility [
[9]
]. Several of the variables, such as education level and income, had more categories but were collapsed into dichotomous and trichotomous variables due to low counts in some categories. As for gestational age, many women either reported no gestational age or only reported weeks and not days. However, the data on whether the surveys belonged to women having 1-day procedures (first-trimester abortion) or 2-day procedures (second-trimester abortion) made it possible to create a dichotomous variable for gestational age.Multivariate logistic regression was employed for predictors of interest based on the bivariate analyses. The multivariate model was constructed from the primary predictors and from predictors that had a significant association with IUD choice in the bivariate analyses. Missing data were handled both by case-wise deletion and by a conservative estimate considering missing answers to questions regarding knowledge of and exposure to IUDs. The latter approach is presented here, as the two models had the same significant predictors and similar ORs.
In the case of variables that had significant collinearity, the decision was made to use the more clinically relevant predictor. This included the variables for reaction to an IUD television (TV) commercial and having the impression that the IUD would be a good method choice based on the TV commercial. The latter variable was chosen to represent the influence of the TV commercials, as it was a more significant predictor of IUD choice and was more clinically relevant. A composite variable was created for predictors that had significant interaction, such as hearing personal IUD experiences from a counselor and hearing personal experiences from another individual at the clinic.
An additional analysis was conducted comparing women who had chosen an IUD after stating an initial IUD interest in the first survey to women who chose an IUD but had not previously stated an interest in getting an IUD. This secondary analysis was also conducted using bivariate and multivariate logistic regression techniques. All analyses were performed using Stata version 10.1 (Stata Corporation, College Station, TX, USA).
3. Results
Out of 415 survey pairs distributed, 393 women (95%) completed at least one half and 299 (72%) completed both halves of the survey. Fifty percent (50%) of the women surveyed chose to use an IUD for postabortion contraception. Demographic characteristics of the study population are presented in Table 1. Race/ethnicity, parity, and gestational age were significant predictors of IUD choice (Table 2). Additionally, 8% of women reported previous IUD use, compared to 28% with previous use of contraceptive injections and 72% who reported previous use of a combined (estrogen and progestin) hormonal method, such as the pill, the patch or the ring.
Table 1Demographic and reproductive characteristics of women in an urban abortion clinic (n=299)
| Characteristic | n (%) |
|---|---|
| Demographic characteristics | |
| Age in years (mean ± SD) | 24.8 ± 6.6 |
| Race/ethnicity | |
| Black | 88 (31) |
| Latina | 77 (27) |
| White | 50 (17) |
| Other | 73 (25) |
| English language at home | 218 (76) |
| Country of origin | |
| USA | 234 (81) |
| Latin America | 30 (10) |
| Other | 24 (8) |
| Years living in USA | |
| Less than 10 years | 25 (9) |
| 10 or more years | 262 (91) |
| Education | |
| HS graduate or less | 155 (54) |
| Education beyond HS | 133 (46) |
| Income level | |
| Less than 25K per year | 142 (74) |
| 25K or more per year | 51 (26) |
| Reproductive history | |
| Multigravid | 211 (73) |
| 1 or more children | 171 (59) |
| Prior abortion | 157 (55) |
| Gestation ≥15 weeks | 127 (44) |
| IUD baseline knowledge | |
| Previously heard of IUD | 205 (75) |
| Knew someone with an IUD | 133 (50) |
| Saw TV ad for IUD | 144 (51) |
| Wanted IUD before coming to clinic | 84 (28) |
HS=high school.
a Due to incomplete surveys, some proportions were calculated out of a number of women less than the 299 surveyed.
Table 2Bivariate association between demographic characteristics and IUD choice in an urban abortion clinic
| Predictors of IUD choice | % who chose IUD | Unadjusted OR (95% CI) | p Value |
|---|---|---|---|
| Age (years) | 1.01 (0.98–1.05) | .57 | |
| Race/ethnicity | |||
| White | 31/50=62% | Reference category | |
| Black | 38/88=43% | 0.47 (0.23–0.95) | .035 |
| Latina | 47/77=61% | 0.96 (0.46–2.00) | .91 |
| Language at home | |||
| English | 108/218=50% | Reference category | |
| Spanish | 12/22=55% | 1.22 (0.51–2.95) | .66 |
| English and Spanish | 16/30=53% | 1.16 (0.54–2.50) | .70 |
| Other | 6/13=33% | 0.51 (0.18–1.41) | .19 |
| Country of origin | |||
| United States | 118/234=50% | Reference category | |
| Latin America | 17/30=57% | 1.29 (0.60–2.77) | .52 |
| Other | 7/24=29% | 0.40 (0.16–1.01) | .053 |
| Education level | |||
| HS graduate or less | 78/155=50% | Reference category | |
| Beyond HS | 64/133=48% | 0.92 (0.58–1.46) | .71 |
| Annual income level | |||
| Less than 25K | 78/142=55% | Reference category | |
| 25K or more | 27/51=53% | 0.92 (0.49–1.75) | .81 |
| Gravidity | |||
| Primigravid | 32/77=42% | Reference category | |
| Multigravid | 111/211=53% | 1.56 (0.92–2.65) | .098 |
| Parity | |||
| Nulliparous | 42/118=36% | Reference category | |
| 1 or more children | 101/171=59% | 2.61 (1.61–4.24) | <.001 |
| Prior abortion | |||
| No prior abortions | 65/129=50% | Reference category | |
| 1 or more prior abortions | 76/157=48% | 0.92 (0.58–1.47) | .74 |
| Gestational age | |||
| Less than 15 weeks | 67/159=42% | Reference category | |
| 15 weeks or more | 75/127=59% | 1.98 (1.23–3.18) | .005 |
Three fourths (75%) of the women who were surveyed had previously heard of an IUD, while 50% reported knowing a friend or family member with an IUD (Table 1). Twenty-six percent (26%) of all women knew a friend or family member who had shared a positive experience using an IUD. Looking only at the women who chose an IUD, 83% had heard of IUDs previously and 60% knew someone with an IUD, while 33% knew someone with a positive, personal IUD experience.
Twenty-eight percent (28%) of women surveyed wanted an IUD before discussing birth control options with anyone at the clinic (Table 1). Women who were already interested in choosing an IUD for postabortion contraception accounted for 50% of women who ultimately chose an IUD, while only 7% of women who chose something other than an IUD stated an initial interest in using an IUD. Many patients reported disclosure of personal IUD use by their counselors (25%) or by anyone at the clinic (43%). A third (35%) of women who chose an IUD experienced disclosure from their counselors, and 57% from anyone at the clinic, including counselors, nurses, doctors, residents, medical students and other patients. These proportions were 15% and 28% in the group of women who did not choose an IUD for postabortion contraception.
Many variables regarding IUD knowledge and sharing of personal experiences with IUDs were significant predictors of IUD knowledge when looking at bivariate analyses (Table 3). Statistically significant predictors of IUD choice were having heard of an IUD, knowing someone with an IUD, knowing someone with a positive IUD experience, having a positive impression based on the IUD TV commercial (i.e., thinking it would be a good method to choose based on the commercial), experiencing disclosure of IUD use from a counselor or from someone else at the clinic and already wanting an IUD.
Table 3Bivariate predictors of IUD choice related to previous knowledge and shared experiences at an urban abortion clinic
| Predictors of IUD choice | % who chose IUD | Unadjusted OR (95% CI) | p Value |
|---|---|---|---|
| Previously heard of IUD | |||
| No | 24/69=35% | Reference category | |
| Yes | 114/205=56% | 2.35 (1.33–4.14) | .003 |
| Knew someone with an IUD | |||
| No | 52/132=39% | Reference category | |
| Yes | 78/133=59% | 2.18 (1.33–3.56) | .002 |
| Knew someone who shared positive IUD experience/impression | |||
| No | 100/221=45% | Reference category | |
| Yes | 49/78=63% | 2.04 (1.20–3.47) | .008 |
| Saw TV ad for IUD | |||
| No | 59/137=43% | Reference category | |
| Yes | 82/144=57% | 1.75 (1.09–2.80) | .020 |
| Had positive reaction to TV ad | |||
| No | 99/227=44% | Reference category | |
| Yes | 50/72=69% | 2.94 (1.67–5.17) | <.001 |
| Had positive impression based on TV ad (thought it would be good method for them) | |||
| No | 104/237=44% | Reference category | |
| Yes | 45/62=73% | 3.39 (1.83–6.26) | <.001 |
| Experienced disclosure of personal use by counselor | |||
| No | 97/224=43% | Reference category | |
| Yes | 52/75=69% | 2.96 (1.70–5.17) | <.001 |
| Experienced disclosure of personal use by another person at clinic | |||
| No | 95/227=42% | Reference category | |
| Yes | 54/72=75% | 4.17 (2.30–7.56) | <.001 |
| Experienced disclosure of personal use by anyone at clinic | |||
| No | 68/181=38% | Reference category | |
| Yes | 81/118=69% | 3.64 (2.22–5.95) | <.001 |
| Wanted IUD precounseling | |||
| No | 75/215=35% | Reference category | |
| Yes | 74/84=88% | 13.81 (6.74–28.31) | <.001 |
CI=confidence interval.
Stepwise regression was employed with the predictors found to be significant in the bivariate analyses to create a multivariate logistic regression model, with choice of IUD as the outcome (Table 4). The predictors that were significant in the multivariate model were wanting an IUD before coming to the clinic, experiencing disclosure of IUD use by a counselor or other clinic staff member, Latina ethnicity and multiparity. Gestational age was not a significant predictor of IUD choice in the final model. The sharing of personal IUD experiences by a counselor was a significant predictor of IUD choice even when controlling for whether or not each woman had a counselor who used an IUD personally (p=.003).
Table 4Multivariate predictors of IUD choice at an urban abortion clinic (n=231)
| Predictors of IUD choice | Adjusted OR | 95% CI |
|---|---|---|
| Wanted an IUD before coming to clinic | 25.5 | 8.2–79.4 |
| Counselor or other clinic worker shared personal IUD experience | 8.1 | 3.8–17.2 |
| Latina ethnicity | 2.7 | 1.2–5.8 |
| Multiparity | 2.2 | 1.0–4.5 |
CI=confidence interval.
a Adjusted for race/ethnicity, second-trimester gestation, having heard of IUDs, knowing someone with an IUD, having a positive impression based on an IUD television ad and knowing someone with a positive IUD experience.
As a secondary analysis, the stage at which women decided on an IUD was also investigated. Demographic characteristics were not significantly different between women who wanted an IUD coming into the clinic compared to women who changed their minds about a different method or no method and decided to get an IUD while at the clinic. Additionally, data were analyzed for women who completed only half of the survey. The demographic characteristics of women who only completed the first half of the survey, including age, race/ethnicity, language, education level and income, were not significantly different than the characteristics of women who completed both parts of the survey.
Of the 149 women who chose an IUD for postabortion contraception, 74 stated a desire to get an IUD prior to contraceptive counseling, while 75 initially wanted a different method of birth control or no method of birth control. There were no statistically significant differences in demographic characteristics between these two groups. The women who initially wanted IUDs were more likely than women who decided on an IUD while at the clinic to have heard of IUDs previously (96% vs. 70%, p<.001), know someone with an IUD (69% vs. 51%, p=.032), know someone with a positive IUD experience (49% vs. 17%, p<.001), have seen the Mirena TV commercial (67% vs. 49%, p=.032) or have a positive impression based on the TV commercial (50% vs. 11%, p<.001). There were no significant differences in sharing of personal experiences by counselors or other clinic workers between the two groups. Thus women who changed their minds to choose an IUD were not more likely to experience counselor disclosure of personal IUD use than women who initially desired an IUD.
4. Discussion
The most important predictors for choice of IUD at the Women's Options Center at San Francisco General Hospital were coming into the clinic already wanting an IUD, hearing from a counselor or other clinic staff member who shared a personal IUD experience, Latina ethnicity and multiparity. This study was unique in that we found that, even while controlling for all factors correlated with IUD choice (including wanting an IUD before coming to the clinic), sharing of a personal IUD experience by someone at the clinic increased the likelihood of someone choosing an IUD. Given that this is a group of women at high risk for repeat unintended pregnancy, exposing them to the option of a long-term method of contraception can have a very positive effect on their future. Almost half (43%) of the women heard another woman's personal IUD experience while at the clinic, and this played an important role in many women's decisions to choose an IUD. Counselors and other clinic staff members who have used IUDs themselves have the ability to reassure women about contraceptive methods with which they may not already be familiar.
The IUD is a very common choice for postabortion contraception at San Francisco General Hospital's Women's Options Center. Overall, 41% of women at this clinic choose an IUD for contraception, with 39% choosing the levonorgestrel IUD and 2% choosing the copper IUD. The study population had a slightly higher uptake of IUDs, with 50% of women choosing IUDs, due primarily to the inclusion and exclusion criteria of the study. Considering that the rate of IUD use in the United States is only 5% among women using contraception, the proportion of women who decide to use an IUD for postabortion contraception at this clinic is very impressive [
[7]
]. A study conducted several years ago in another population of young pregnant women demonstrated that 50% of these women had heard of IUDs and only 25% of women knew someone with an IUD [[2]
]. Considering the high uptake of IUDs in the Women's Options Center population, it was not surprising that previous knowledge of IUDs was also much higher, with 75% reporting previous knowledge of IUDs and 49% reporting a friend or family member with an IUD.The authors propose several reasons why IUD knowledge may be higher in this population than in previous studies. First, knowledge and uptake of IUDs have increased in the United States [
[7]
]. Stigma against IUD use in young, nulliparous and nonmonogamous women has decreased, and many organizations support its safety in these populations [11
, 12
, 13
]. In addition, the IUD is now known to be safe in the postabortion setting [12
, 13
, 14
]. Third, many participants had been to the Women's Options Center previously for abortion care and already received comprehensive contraceptive counseling that included a discussion about IUDs; others have been exposed to television commercials and other advertising for IUDs (Table 2, Table 3). Fourth, women at our clinic are able to access IUD insertion at the time of abortion because they live in a state that understands the importance of funding effective contraception for low-income women through the Medi-Cal and Family Planning, Access, Care and Treatment (Family PACT) programs. Finally, women are discussing their birth control experiences within their communities. More than one quarter of the women surveyed reported talking to a friend or family member who not only had an IUD but also shared a positive story about their experience using an IUD for contraception. These data suggest that sharing of contraceptive experiences is common among women of all demographics.The high uptake of IUDs in the Women's Options Center may indicate that open discussion about IUDs allows women to feel less apprehensive of IUDs as a safe method and might increase their method satisfaction. However, a staff member's positive experience with an IUD does not necessarily translate to a similar experience for the patient, and in such cases, women may feel disappointed and less satisfied with the IUD. In general, high-quality research about contraceptive counseling has been lacking, and our findings call for additional research to be completed regarding the effect of the content of this counseling [
[15]
]. This includes an examination of how staff opinions and experiences impact counseling and whether patients influenced by disclosure are satisfied with their methods. It does not appear that our subjects were overly influenced by disclosure, as half of women choosing an IUD wanted one before they arrived at the clinic. However, the fact that sharing of personal IUD experiences by providers was independently associated with its use highlights the potential impact of sharing contraceptive stories on women's contraceptive decision-making. Our results call for further improvements in contraceptive counseling, including an examination of how clinic staff impressions of methods impact counseling and whether staff members' contraceptive method choice should even be disclosed at all. It will also be important to examine whether those patients influenced by staff method disclosure are as satisfied with their IUDs as those who were not exposed to this information. Overall, these results are reassuring in that half of all women choosing an IUD wanted an IUD before they even came to the clinic. However, controlling for demographics and prior knowledge of IUDs, sharing of personal IUD experiences by providers is significantly associated with IUD use. This highlights the impact that providers can have on contraceptive decision-making and the power that sharing of personal contraceptive stories can afford.Acknowledgments
This project was supported by a grant from the Doris Duke Charitable Foundation. This project was supported by NIH/NCRR/OD UCSF-CTSI Grant Number TL1 RR024129. Its contents are solely the responsibility of the authors and do not necessarily represent the official views of the NIH.
References
- Disparities in rates of unintended pregnancy in the United States, 1994 and 2001.Perspect Sex Reprod Health. 2006; 38: 90-96
- Young pregnant women's knowledge of modern intrauterine devices.Obstet Gynecol. 2006; 108: 1417-1422
- Contraceptive care for adolescents.Clin Obstet Gynecol. 2008; 51: 268-280
- Facts on induced abortion in the United States, in brief. Guttmacher Institute.(Available from:)
- Patterns in the socioeconomic characteristics of women obtaining abortions in 2000–2001.Perspect Sex Reprod Health. 2002; 34: 226-235
- Abortion in the United States: incidence and access to services, 2005.Perspect Sex Reprod Health. 2008; 40: 6-16
- Use of contraception in the United States: 1982–2008.Vital Health Stat. 2010; 29: 1-44
- Use of contraception and use of family planning services in the United States: 1982–2002.Adv Data. 2004; 350: 1-36
- The levonorgestrel intrauterine system in nulliparous women.Contraception. 2007; 75: S12-S15
- Immediate postabortal insertion of intrauterine devices.Cochrane Database Syst Rev. 2004; 4: CD001777
- ACOG practice bulletin no. 59: intrauterine device.Obstet Gynecol. 2005; 105: 223-232
- Medical eligibility criteria for contraceptive use.3rd ed. WHO, Geneva2004
- U.S. medical eligibility criteria for contraceptive use, 2010: adapted from the World Health Organization medical eligibility criteria for contraceptive use, 4th edition..MMWR Recomm Rep. 2010; 59: 1-86
- Insertion of intrauterine contraceptives immediately following first- and second-trimester abortion.Contraception. 2009; 79: 397-402
- Counseling in the clinical setting to prevent unintended pregnancy: an evidence-based research agenda.Contraception. 2003; 67: 115-132
Article info
Publication history
Published online: August 17, 2011
Accepted:
May 27,
2011
Received in revised form:
May 27,
2011
Received:
January 3,
2010
Identification
Copyright
© 2012 Elsevier Inc. Published by Elsevier Inc.
User license
Creative Commons Attribution - NonCommercial - NoDerivs | How you can reuse 
Elsevier's open access license policy
Creative Commons Attribution - NonCommercial - NoDerivs
Permitted
For non-commercial purposes:
- Read, print & download
- Redistribute or republish the final article
Not Permitted
- Text & data mine
- Translate the article
- Reuse portions or extracts from the article in other works (except for the author)
- Sell or re-use for commercial purposes
Elsevier's open access license policy
