Advancing Sexual and Reproductive Health and Rights
International Perspectives on Sexual and Reproductive Health
Volume 35, Number 2, June 2009

Prevalence of Adolescent Sexual Behaviors May Be Rising in Asian Cities

In Asia, rapid change in adolescent romantic and sexual behaviors has accompanied the transition from a traditional Confucian to a more modern society, according to a study of young people in metropolitan areas in Vietnam (Hanoi), China (Shanghai) and Taiwan (Taipei).1 As a result, levels of these behaviors were more common among a recent cohort of adolescents than they were among their counterparts from just a few years earlier. For example, the researchers found that 15–19-year-olds in their sample often were more likely to have dated or to have engaged in fondling or intercourse by age 18 than were 20–24-year-old respondents, who had matured at a slightly earlier stage of societal modernization. A similar relationship between adolescent behavior and modernization: Sexual behaviors were most prevalent in relatively modern Taipei, and least prevalent in less developed Hanoi.

The researchers interviewed a total of 17,016 young people aged 15–24 from the three major metropolitan areas. In Hanoi and Shanghai, participants were recruited using a multistage household sampling process, with statistically appropriate numbers recruited from factory housing and university dorms; participants in Taipei were recruited primarily from schools and universities, but an appropriate number of nonstudents were interviewed as well. About one-quarter of participants were recruited from rural sites, as all three metropolitan areas contained districts that were largely agricultural or were relatively sparsely populated.

Respondents were asked to report on the ages at which they had engaged in various behaviors (if at all), including romantic and sexual behaviors (dating, fondling and first intercourse) and nonsexual risk behaviors (smoking and drinking). They also provided information on their age at sexual maturation and household economic status (measured as the number of assets in the respondent's home at age 14). To explore short-term societal changes in sexual and risk behaviors, the investigators used survival analyses to examine differences between two cohorts of similar age (15–19 and 20–24). They also calculated descriptive statistics, and conducted Cox regression analyses to identify factors associated with specific behaviors; the models included gender, age cohort, residence (urban or rural), study site and in some cases number of household assets.

In general, the more modern the study site, the greater the proportion of participants who had dated or engaged in fondling or coitus by age 16 or 18. For instance, the proportion of 20–24-year-old males who had dated by age 18 was greater in Taipei (65%) than in Shanghai (33%) or Hanoi (36%); a similar pattern, albeit with smaller differences among sites, was apparent for 15–19-year-old males (69%, 62% and 56%, respectively). In both Shanghai and Hanoi, the prevalence of the three romantic and sexual behaviors was generally higher among younger than older participants; in Taipei, this was true for dating, but not for fondling or intercourse.

In all three societies, the proportion of males reporting each act was greater in both age-groups than the proportion of females. For example, while 2–5% of males in Hanoi reported having had sex, fewer than 1% of females did so; a similar pattern was evident in Shanghai (5–11% of males vs. 2–6% of females) and, to a lesser extent, Taipei (23–24% vs. 17–22%). The gap between the sexes tended to be smaller in the more modern cities: Among the younger cohort, for example, the difference between the proportions of males and females who had experienced fondling by age 18 was 15 percentage points in Hanoi, 10 in Shanghai and fewer than three in Taipei. Differences between urban youth and rural youth also tended to be less pronounced at the more modern sites. As with sexual and romantic behaviors, levels of smoking and drinking increased with the level of modernization both across sites and over time, and the gap between males and females appeared to shrink.

Surprisingly rapid changes in young people's age at sexual maturity accompanied modernization in all sites, perhaps because of improvements in diet. For male respondents, self-reported age at first ejaculation was lower among 15–19-year-olds than among 20–24-year-olds in Hanoi (15.5 vs. 15.9), Shanghai (15.6 vs. 16.2) and Taipei (14.7 vs. 15.3). Similarly, for females, age at menarche was lower in the younger age-group than in the older one in Hanoi (14.4 vs. 15.0) and Shanghai (14.2 vs. 14.4), although not in Taipei (13.3 in both age-groups).

A multivariate analysis that pooled data from the three sites revealed that members of the older cohorts were significantly less likely than those in the younger cohorts to have dated (hazard ratio, 0.6), fondled (0.6) or had sex (0.8) by age 18. Males were more likely than females to have dated (1.2), fondled (1.7) or had sex (1.5), they note, and urban respondents were more likely than their rural counterparts to have dated (1.2) or fondled (1.2). In addition, large differences were apparent by study site: Compared with young people in Hanoi, those in Shanghai and Taipei were more likely to have dated (1.1 and 2.4, respectively) or had sex (2.5 and 11.7) by age 18. The likelihood of fondling was higher in Taipei (1.8) than in Hanoi, although there was no difference between the latter and Shanghai.

When the analyses were further adjusted to take into account household assets—an indicator of modernization—the hazard ratios stayed the same for age-group and gender, but the differences between urban and rural residents disappeared. The greatest change was in study site: Although adolescents in Taipei remained significantly more likely than those in Hanoi to have dated (hazard ratio, 1.1) or to have had sex (7.6), the differences were much smaller than in the previous analysis, and they were now less (rather than more) likely to have engaged in fondling (0.9). Young people in Shanghai had a slightly lower likelihood of having dated or engaged in fondling (0.9 and 0.8, respectively) than those in Hanoi, but despite this, they were more likely to have experienced coitus (2.2).

The researchers posit that exchange with the international community is lessening traditional values in Asia regarding sexual and risk behaviors, thus leading to increased prevalence of such behaviors among adolescents; this trend, they note, appears to affect male and urban adolescents first. Moreover, the findings suggest that adolescent sexual activity may continue to increase as these societies have further contact with the outside world and become more modern—a transition that began relatively recently in Hanoi, is well underway in Shanghai and has occurred to an even greater degree and for a longer time in Taipei. The investigators caution that policies and programs for adolescents must be sensitive to the sometimes extensive differences between adjacent age-groups of young people, because "the rapid changes reported here demonstrate that treating adolescence and youth as a single period not only poses the usual hazard of minimizing individual developmental change but also carries the risk of overgeneralization."—H. Ball


1. Zabin LS et al., Levels of change in adolescent sexual behavior in three Asian cities, Studies in Family Planning, 2009, 40(1):1–12.