As age at marriage has risen in many developing countries, concern has grown about early sexual activity and the risks of unplanned pregnancy and HIV infection. Research has generally focused on identifying factors associated with early, premarital intercourse among young people and interventions that may be successful in preventing or delaying sexual initiation. Two articles in this issue of International Family Planning Perspectives take another look at the situation.

In the lead article, Ushma Upadhyay, Michelle Hindin and Socorro Gultiano explore patterns in the early emotional relationships and physical behaviors that adolescents engage in before they begin to have intercourse [page 110]. Overall, according to data from the Cebu Longitudinal Health and Nutrition Survey, adolescents move through a standard sequence: About two years after having their first crush (at a median age of 14 for males and 13 for females), they experience their first courtship, then their first romantic relationship and finally their first date. Although this sequence of relationships is the same for males and females, males progress through the sequence more quickly (a median of two years vs. four years). First intercourse tends to follow the first date by more than a year (the median age at first sex could not be calculated because only 31% of males and 20% of females aged 17–19 had had sex). The authors suggest that the period following an adolescent's first date provides an opportunity to ensure that the adolescent has access to the information and services needed to make informed choices about sexual behavior.

Alice Liu and colleagues look at factors related to sexual initiation in Thailand, where norms in premarital sexual behavior appear to have changed over the past two decades [see article]. According to data from a survey of more than 1,700 vocational school students aged 15–21 in Chiang Rai, the median age at first sex was 17 for males and 18 for females. At any given age, both males and females were more likely to have had sex if they had used alcohol or methamphetamine. For males, having had sex was also associated with, among other factors, having parents who did not live together, having a friend as a confidant, smoking tobacco, a high perceived risk of HIV infection and a high level of STI knowledge. For females, additional factors were living away from family, lacking a family member as a confidant, ever having smoked marijuana, having a high perceived risk of STIs, and a younger age at interview. According to the authors, their data confirm previous evidence indicating that young women are starting to have sex at younger ages. They conclude that programs aiming to reduce the adverse consequences of early sexual activity should target youth who become sexually active at an early age and those engaged in patterns of generalized risk-taking, and should take into account the influence of family and peers.

Also in This Issue

•Using birth histories from India's 1998–1999 National Family Health Survey, Saseendran Pallikadavath and R. William Stones seek to determine whether abortion is associated at the national level with maternal and social factors, including the sex of a woman's last child [see article]. The strongest predictor of abortion was education: The odds of abortion among women with at least a primary education were 1.9–6.7 times as high as the odds among women with no education. Women living in rural areas had a reduced risk of abortion. The sex of the previous child was not associated with the odds of abortion. While they found no indication that son preference is an important factor in abortion at the national level, the authors note that such influences may exist at the state or regional level, given recent evidence of highly skewed sex ratios in states such as Haryana.

•Although abortion is highly restricted in Guatemala, an estimated 65,000 abortions are performed in the country every year, according to calculations by Susheela Singh, Elena Prada and Edgar Kestler [see article]. Using the estimated number of abortions and the annual number of unplanned births, the authors estimate that one in three pregnancies in Guatemala are unintended. Noting that the gap between desired and actual family size remains high (despite declines in both), and that almost half of Guatemalan women do not know of any method of contraception (a proportion that is higher among indigenous women and in rural areas), the authors argue that reducing the level of induced abortion will require providing widespread access to family planning information and services, with special attention to poor, rural and indigenous women and couples.

•Migration to urban areas of Guatemala from rural areas—where levels of contraceptive knowledge and practice are much lower—may be creating a need for reproductive health programs focused on new urban residents. According to research by David P. Lindstrom and Coralia Herrera Hernández, migrants' contraceptive knowledge increases with the number of years spent in an urban environment [see article]. The level of contraceptive use among rural-to-urban migrants eventually comes close to that among urban natives, with the level of contraceptive knowledge being an important factor associated with use of modern methods. Because Mayan migrants do not accumulate this information at the same rate as Ladino migrants, perhaps because of cultural and linguistic barriers, the authors argue that programs should target recent arrivals—particularly indigenous Mayans—in urban areas.

—The Editors