At any given time, roughly one in five homeless women in the United States are pregnant--a substantially higher proportion than is found among women in the general population or among low-income women who are not homeless. Nevertheless, key questions that may help in the design of reproductive health services for this hard-to-reach population have not been closely studied. For example, what factors do homeless women identify as standing between them and effective contraceptive use? And how do perceived obstacles vary among the large, diverse population of homeless women? Lillian Gelberg and colleagues address these questions, using data from a sample of homeless women in Los Angeles County, in this issue of Perspectives on Sexual and Reproductive Health [see article].

More than one in five women in this sample considered the side effects or potential health risks associated with contraceptives, their partner's dislike of contraception and the cost of a method major deterrents to use; other factors commonly identified as "big problems" were lack of a storage place and not knowing what method to use or how to use a method. Not surprisingly, women who perceived substantial deterrents to use were less likely than others to report consistent method use. Moreover, women's views of possible deterrents were related to a number of background and behavioral characteristics--notably, race or ethnicity and history of substance abuse.

As the authors comment, the broad range of perceived deterrents and the differences in the way various groups of homeless women view them suggest that "a uniform approach to increasing contraceptive use is unlikely to be effective for the entire homeless population." Rather, Gelberg and her coauthors argue for the need to develop culturally competent family planning services for homeless women, and to integrate these services into a comprehensive system of health care and social services.

Also in This Issue

• Whether promoting marriage among welfare-dependent women with children is an effective strategy for reducing poverty and welfare caseloads has been the focus of intense debate. Yet, as Deborah Roempke Graefe and Daniel T. Lichter point out [see article], little is known about the marriage patterns of unwed mothers. In analyses of data from the 1995 National Survey of Family Growth, Graefe and Lichter examine patterns of marriage and of divorce among women who have an out-of-wedlock first birth. They find that these women are less likely to marry before age 40 than are those who delay childbearing until they marry; although marriage rates vary among white, black and Hispanic women, they are reduced among unwed mothers in all three groups. Moreover, women who marry after giving birth are more likely than those whose first birth occurs within marriage to divorce; again, levels vary among racial and ethnic groups, but are associated with nonmarital childbearing for all women. The authors comment that the emphasis on marriage is "at once important and misplaced," and that "marriage promotion might best begin with expanded efforts to reduce nonmarital childbearing."

• Unintended pregnancies occur when couples at risk do not practice contraception, use a method inconsistently or incorrectly, or experience a method failure. Using data from a nationwide survey of U.S. women having abortions in 2000-2001, Rachel K. Jones, Jacqueline E. Darroch and Stanley K. Henshaw explore the contribution of each of these factors to the occurrence of unintended pregnancies ending in abortion [see article]. They find that nearly half of women having abortions were not using a contraceptive around the time they conceived, largely because they believed that their risk of pregnancy was low. Most users relied on male condoms or the pill, and the majority of these women attributed their pregnancy to inconsistent or incorrect use (but negligible proportions used emergency contraceptive pills for backup protection); fewer than 15% reported having used the method perfectly. Jones and her colleagues conclude that in addition to accessible and affordable contraceptive services, women and men need accurate information about the risk of pregnancy in the absence of correct, consistent contraceptive use.

•Relationships between young women and older men may be marked by dynamics that increase the woman's risk of unintended pregnancy and sexually transmitted disease. A research note by Christine E. Kaestle, Donald E. Morisky and Dorothy J. Wiley [see article] provides fresh insights into possible consequences of such relationships by demonstrating how romantic partners' age difference relates to the likelihood that the couple will engage in intercourse. Using data from the National Longitudinal Study of Adolescent Health, the investigators show that teenage women with an older romantic partner are more likely to have intercourse with their partner than are their peers with a partner who is their age. The effect is most dramatic among the youngest teenagers. According to the investigators, the findings highlight the need for interventions to address the behaviors of both young women and their older partners; to improve parents' ability to talk with teenagers about healthy romantic relationships; and to assist young women in identifying their own interests and in asserting their decisions.

•In a special report [see article], Jill L. Schwartz and Henry L. Gabelnick review the types of contraceptive methods that are currently in the research and development pipeline--chemical and mechanical barriers, hormonal methods, male methods, new sterilization techniques and immunocontraceptives. While no radically new types of method are on the horizon, many of the ones the authors describe could make contraceptive use simpler and more acceptable for men and women throughout the world.