Policies and programs intended to improve reproductive health are based on available information and designed with care and commitment. Why, then, do they often either fail to achieve their intended results or have only a limited effect? Sometimes programs work well in one set of circumstances, one setting or one population, but not in others. Further, even well-designed and well-conducted research may not be able to identify all the factors that can affect program outcomes. The articles in this issue report on situations in which conventional wisdom or conventional approaches to analysis needed to be reexamined.
In the lead article, "Unmet Need and Unintended Fertility: Longitudinal Evidence from Upper Egypt," John Casterline and colleagues take a new look at an issue basic to all family planning programs--identifying the group of women most at risk of unintended pregnancy [see article]. Unlike a 1999 study, in which the greatest share of unintended pregnancies in Peru occurred among users and among women who had recently discontinued use, this study finds that in Egypt, unintended pregnancies are concentrated in women with unmet need. On which group, then, should programs concentrate their resources? Casterline and colleagues conclude that in countries such as Peru, where unmet need is low, contraceptive prevalence is high and many women rely on traditional methods, contraceptive users are likely to account for a large share of unintended pregnancies. In these settings, it may be more cost effective to focus on providing users with the information and services they need to continue practicing contraception for as long as they wish to avoid pregnancy. However, in countries such as Egypt, where unmet need is high, contraceptive use is low and most use is accounted for by modern methods, women with unmet need are an appropriate target group.
Family planning programs using community-based distribution (CBD) have been used in the developing world for more than 30 years, with varying degrees of success. The quasi-experimental studies used to evaluate their effectiveness, however, have not generally given program planners information about the factors that contribute to their success or failure. Because CBD programs rely on the ability of agents living in the community to promote positive attitudes toward and adoption of contraceptives, Kirsten Stoebenau and Thomas W. Valente use network analysis to identify which individuals received information from and were influenced by two agents in a village in the highlands of Madagascar [see article]. By identifying the people with whom study respondents discussed family planning and determining whether the respondents had direct, indirect or no links to the CBD agents, the investigators find that the agents were central to the village's family planning network and influenced knowledge and use of contraceptives by direct and indirect communication with people in the community. These findings suggest that characteristics of the CBD agents themselves may be vital to the success of these programs.
Policies and programs designed to reduce risky sexual behavior among young people have often been based on the assumption that the same approach will work for both males and females. In "The Psychosocial Context of Young Adult Sexual Behavior in Nicaragua: Looking Through the Gender Lens," Manju Rani and colleagues examine the very different messages young men and women in that country receive from family and society about premarital sex and pregnancy and the implications of this double standard [see article]. For example, young men indicated higher agreement than young women with the statement that their friends would make fun of them if they did not have premarital sex, whereas young women indicated greater agreement with the statement that they would feel guilty if they did have sex. The researchers argue that young women feel pressure to have sex to preserve their relationship with their partner, but fear the disapproval of their family and society. These conflicting pressures, they say, place young women at high risk of unprotected sex and unplanned pregnancy, adding that programs must address this situation if they are to be successful.
In almost all countries where abortion is legal, few women resort to illegal procedures and subsequently suffer complications. In India, however, it is estimated that fewer than 10% of abortions are performed by registered providers in government-licensed facilities, and that about 20,000 deaths a year occur as a result of abortion complications. Nevertheless, in the state of Uttar Pradesh, few women come to referral-level facilities for treatment of complications. Heidi Johnston and colleagues report on research in two rural communities that explored where women seek care and what kind of treatment they receive [see article]. They find that women tend to seek care first from village-level providers, many of whom are untrained or inappropriately trained. As a result, their complications may be exacerbated, appropriate care delayed and the cost of treatment increased. The researchers recommend that links between village-level providers and the formal health system be strengthened, and that the social barriers that discourage women from seeking appropriate care for abortion complications be addressed.
--The Editors