Abstract
Background To reduce unintended pregnancy, it is necessary to understand why women have unprotected intercourse when they do not desire pregnancy.
Methods We devised a survey of 42 potential reasons why women have unprotected intercourse based on the responses of a focus group we had previously convened. We administered the survey to women between the ages of 18 and 39 years who were visiting primary care clinics and were not trying to get pregnant.
Results Of the 151 respondents, 84 (56%) were having unprotected intercourse. Women gave an average of 9 reasons for having unprotected intercourse. The most common reasons fell into 3 categories: lack of thought/preparation (87% of respondents), being in a long-term or strong relationship (70%), and concerns about side effects of contraception (80%). Eighty-three of the 84 women (99%) chose at least 1 of these categories.
Conclusion Basing survey questions on focus group responses provided important insights into the reasons women risk unintended pregnancy. A deeper understanding of this issue is critical to reducing unintended pregnancy.
What are the reasons women ordinarily give for unintended pregnancy? The results of our study show that some of the more common ones are not included on standard risk-assessment surveys. If we hope to offer patients a meaningful course of intervention, it would help to understand the issues these women contend with.
Despite the availability of effective contraception, many women have unprotected intercourse that puts them at risk for unintended pregnancy. Among women in the United States who are age 18 or older, slightly more than 40% of live births result from unintended conception.1 The reasons women have unprotected intercourse have not been clear. Of the few studies that have addressed this issue,2 some have restricted their investigation to a few potential reasons2-4 or have limited exploration to the reasons associated with a single episode of intercourse.5 The latter type of investigation is too narrow. A more comprehensive approach is needed because risk-taking is likely to be a complex phenomenon, with reasons changing as the context changes or as women try different forms of contraception.
We conducted focus groups with women who were risking unintended pregnancy.6 With results from the focus groups, we developed a survey to determine the relative prevalence of reasons given, and thereby direct future interventions at those that are most common.
Methods
We recruited participants from local primary care clinics serving financially disadvantaged populations. Flyers describing the study were posted, and interested women approached a research assistant stationed in the clinic. We explained the survey and reviewed eligibility criteria with those who inquired. Women who wished to participate gave verbal consent and were taken to a private area, where a research assistant administered the survey. The study was approved by the local institutional review board. We waived written consent because the survey was anonymous and we collected no identifiers.
Eligibility required that a woman be between the ages of 18 and 39 years, unmarried, and not be pregnant or trying to get pregnant. Women who reported having had a hysterectomy or tubal ligation or being menopausal were ineligible. We defined unprotected intercourse as vaginal intercourse with a fertile male without using a condom, hormonal method, diaphragm, intrauterine device (IUD), vaginal ring, Lea’s shield (a vaginal barrier contraceptive), emergency contraception, vaginal sponge, or cervical cap. These eligibility requirements were identical to those of the focus groups that had provided input for our survey questions.
Women who reported having unprotected sex in the past year were asked to choose from 42 possible reasons (foils) adapted from responses offered in the focus groups.6 When possible, we used the exact words uttered by focus group participants (eg, “I just went with the flow”). We asked women to select all the reasons that applied to them over the past year. The survey also included questions about previous pregnancies, use of home pregnancy test kits, and medical conditions that could affect an unintended pregnancy or fetus (preconceptual health status).
Analysis
We performed univariate analysis using the chi square test in the Statistical Analysis System package (SAS version 8.0, SAS Institute, Inc., Cary, NC). Age was evaluated as a dichotomous variable compared to the median.
Results
Demographics and health
The 151 respondents had a median age of 24 years, and a median household income of <$20,000 per year. Eighty-four women (56%) had unprotected intercourse in the past year. Of the 151 respondents, 56% were white and 41% were black. Twenty-two percent had not graduated from high school. Ten percent had recently been homeless, 9% had recently been jailed, 7% had a recent sexually transmitted disease, and 4% had traded sex for gain.