Program planners and administrators in developing countries must often budget resources and design interventions based on limited information. Their decisions determine whether family planning clinics will have adequate supplies of pills and condoms, whether clients at risk of HIV infection will receive information and counseling about how to protect themselves and whether programs designed to serve young people will provide the right kinds of services in the right way. The articles in this issue of International Family Planning Perspectives look at some of the questions involved in program decisions, and why the answers sometimes may not be clear.

• One such question involves the relationship between contraception and abortion. Some observers argue that adoption of family planning methods decreases reliance on abortion, and others contend that abortion incidence rises with contraceptive prevalence. In the lead article, Cicely Marston and John Cleland [see article] review seemingly contradictory evidence from countries with reliable information on both contraception and abortion. In one group of countries, abortion incidence declined as adoption of modern contraceptive methods increased and as the method mix shifted from traditional to modern methods. However, in a second group--including Cuba, Denmark, the Netherlands, the Republic of Korea, Singapore and the United States--rates of abortion and contraceptive use rose simultaneously during the period studied. A more detailed examination of the two groups of countries found one key difference--the stability of levels of fertility. Countries in which rising contraceptive use coincided with increases in abortion were experiencing sharp drops in fertility, whereas fertility levels in the other countries were stable. The authors conclude, therefore, that rates of abortion and contraception may rise simultaneously if the proportion of couples desiring smaller families increases faster than the proportion adopting a modern family planning method; when fertility is stable, however, a rise in contraceptive use or effectiveness inevitably leads to a decline in induced abortion.

• In countries with large national family planning programs, predicting the potential demand for contraceptive services can be a daunting task. As a proxy for potential demand, most program planners use unmet need, a measure derived from women's stated fertility intentions and their current contraceptive use. Few use women's intentions to practice contraception, a more direct and thus possibly more accurate measure. To test the relative usefulness of these two measures, T.K. Roy and colleagues [see article] analyze data on a sample of married women who participated in both the 1992-1993 National Family Health Survey (NFHS) and a 1999 follow-up survey. The proportion of women adopting a method by 1999 was highest (63%) among women who had said in the NFHS that they planned to practice contraception and to have no children and lowest (25%) among those who had said they intended not to use a method but did not want children. The authors therefore conclude that a combination of contraceptive and childbearing intentions would provide the best estimate of potential demand.

• Two articles in this issue provide complementary perspectives on young people in Ghana. Using data from a nationally representative survey of 3,739 unmarried 12-24-year-olds, Ali Mehryar Karim and colleagues [see article] look at the effects of a broad range of factors related to sexual risk-taking among youth overall. Regression analyses show strong effects on sexual behavior for demographic characteristics such as age and education, as well as for contextual factors such as school attendance, peer behaviors and community connectedness. Condom use, on the other hand, was influenced more strongly by young people's personal characteristics, such as gender role perceptions, feelings of self-efficacy in condom use and communication with partners about risks related to pregnancy and STIs.

• In contrast, Evam Kofi Glover and colleagues [see article] look at differences among social subgroups of Ghanaian youth. Drawing their sample from three Ghanaian regional capitals, the authors interviewed 12-24-year-olds who were either in school, in apprenticeship programs or unaffiliated (in neither of those groups). In-school youth were the least likely to have had sex; of youths in this group who were sexually experienced, both males and females were more likely than their peers in the other groups to have used condoms to protect against STIs, and females were less likely to have been pregnant. This group also had the least-traditional gender role attitudes and the least tolerance for violence against women. Apprenticed youth were the most likely to be sexually experienced and to have been sexually active in the previous month. Almost half of the sexually experienced females in this group had been pregnant. Unaffiliated youth had the most traditional gender role attitudes, were the most accepting of partner violence against women and were the least likely to have used condoms to protect themselves against STIs. On the basis of these data, the authors conclude that broad interventions are less likely to be effective than those designed to meet the needs of specific subgroups of youth.

• Sara L. Zellner [see article] examines factors that predict condom use among adults in Côte d'Ivoire. She finds that the variable of greatest interest, accuracy of knowledge about AIDS, had no effect among either men or women after the effects of social and demographic factors, education and source of knowledge were taken into account. On the other hand, condom use was higher among men and women who had a secondary or higher education. The author suggests that efforts to increase levels of general education in Côte d'Ivoire are likely to be more effective in promoting condom use than interventions that attempt to improve men's and women's knowledge about AIDS.

--The Editors