The beginning of sexual intercourse is a key social development transition in adolescence with regard to physical maturation, cognitive development, increased awareness and appreciation of one’s body, strengthening of individual and gender identity, and formation of sexual relations [1-3]. The prevalence of sexual experience (defined as ever having had sex) is a common index measuring sexual activity at the population level. In the 20th century, most industrialized nations have experienced a decline in the age of sexual initiation among adolescents from cohorts born in the 1920s to those born in the 1970s [4]. However, the median age of sexual debut and related social norms vary by region and culture, and no universal trend has been found towards earlier sexual intercourse [5].

Patterns of sexual behaviors are converging across developed countries. In other words, variation in the timing of sexual initiation has decreased within and between countries. In the United States, although the prevalence of sexual experience among adolescents from 1991 to 1997 decreased by 11%, the age at first intercourse continues to decline [6]. Korean society has traditionally been based on Confucian norms, according to which premarital sexual intercourse and early initiation of sexual intercourse are regarded as taboo [7]. Adolescents in Europe also showed a reduction in age of sexual initiation starting in the 1950s. Early sexual intercourse is typically defined as sexual initiation before the ages of 13 or 14 [8]. Its frequency varies by gender and race/ethnicity, with higher rates among boys and black or Hispanic teens [9-11]. According to a recent report from the Korea Centers for Disease Control and Prevention (KCDC), the prevalence of having experienced sexual intercourse was 4.8% among 31,362 adolescents in 2016 [7]. Although this rate is lower than that of Western countries, adolescents’ experiences of sexual intercourse are still an important topic because numerous studies have reported associations between earlier sexual activity and negative physical and mental health outcomes such as the risk of sexually transmitted infections and depressive symptoms [12-15].

Life course theory is a prospective view that looks at individuals’ or cohorts’ life experiences, including analyses that span generations, to find clues to future patterns of health and disease, considering that both past and present experiences are shaped by the socioeconomic and cultural context [16]. In the framework of the life course theory, adolescents are in a transitional period before becoming adults. Therefore, our study is meaningful in that we provide basic data for predicting patterns of sexual behavior among adults in their 30s and 40s in the future by analyzing patterns of sexual behavior among adolescents. Although the prevalence of sexual experience among Korean adolescents is relatively low compared to that of adolescents in Western countries, younger generations are more prone to have early sexual experiences than older generations owing to the emergence of an information-oriented society in which adolescents are exposed to numerous sources of unfiltered content and media [17]. However, most studies on the prevalence of sexual experience among Korean adolescents were cross-sectional studies [18,19]. Such studies are unsuitable for analyzing the age-specific or birth cohort-specific prevalence of sexual experience [20,21].

The age-period-cohort (APC) model has been widely used to overcome the limitations of cross-sectional data [22-24]. Previous studies have used the APC model not only for analyzing long-term trends of mortality and the incidence of chronic disease [22,25,26], but also for analyzing lifestyle patterns such as smoking in association with death rates [27,28]. By analyzing the independent effects of age, period, and birth cohort on changing trends in the prevalence of sexual experience among Korean adolescents aged 12-17 years, our study aimed to provide evidence that could be used to develop intervention strategies for improving adolescents’ sexual health and to project the prevalence of sexual experience among adolescents into the future. Since the age at first sexual experience is becoming younger, we conducted an APC analysis to test our hypothesis that the prevalence of sex experience has increased among older age groups, in recent periods, and in younger cohorts.

Table 1 shows the sex-specific, age-specific, and period-specific prevalence of sexual experience among subjects. The prevalence of sexual experience generally increased by age in all periods in both boys and girls. In boys, the prevalence of sexual experience increased in recent periods, especially in the age group of 12-13 years, while the prevalence of sexual experience decreased in the age group of 16-17 years. In girls, the age group of 12-13 years showed an increased prevalence of sexual experience in recent periods. However, the prevalence showed a decreasing trend in the age group of 16-17 years. Figure 1A and 1B depicts the age-specific prevalence of sexual experience in boys and girls by period. The age-specific prevalence of sex experience in the age group of 12-13 was higher in 2012-2015 than in other periods. However, the age group of 16-17 years showed the highest prevalence in 2006-2007. Figure 1C and 1D shows the birth-cohort specific prevalence of sexual experience by age groups. In general, the prevalence of sexual experience increased with age in the older birth cohorts. In boys, the age group of 16-17 years showed a decrease in the prevalence of sexual experience in increasingly younger cohorts, while the prevalence of sexual experience increased in younger age groups. In girls, all age groups except the group of 12-13 years showed a trend for the prevalence of sexual experience to decline in increasingly younger cohorts. However, the rate of decline slowed in recent cohorts.

Table 2 shows the results of various models explaining the prevalence of sexual experience. Of these models, the APC and age-cohort models best explained the prevalence of sexual experience in both boys and girls. Figure 2 presents a schematic diagram of the APC model shown in Table 2. In boys, the prevalence of sexual experience rapidly escalated with age. The prevalence of sexual experience steadily increased from 2006 to 2011, and declined in the birth cohorts prior to 1996 and 1997, before steadily increasing in recent cohorts. The results for girls were similar to those for boys in general. The overall increase by period was somewhat more stable in girls than in boys. In the Joinpoint analysis, changes in linear trends were detected (Figure 3). The prevalence of sexual experience among Korean youth decreased from 2007 to 2016, but increased from 2016 to 2018 in girls.

Our study aimed to identify the independent effects of age, period, and cohort on trends in the prevalence of sexual experience in adolescents aged 12-17. The prevalence of sexual experience increased with age. Especially in a relatively recent cohort (1998- 1999), the prevalence of sexual experience was high in the youngest age group (12-13 years).

In our study, boys were found to be more likely to engage in sexual intercourse than girls. This finding is consistent with previous studies from other countries, including the Netherlands [31] and the United States [32-34]. However, this discrepancy may be explained by the social context emphasizing chastity for women, which may lead girls to under-report their sexual experience, whereas boys may tend to over-report their sexual experience since traditional norms regard sexual prowess as a crucial component of masculinity [35].

It has been previously found that boys’ sexual attitudes are influenced primarily by individual factors including curiosity, while females’ sexual attitudes are influenced more by family-related factors such as parental attitudes [36]. Therefore, from a social perspective, rapid exposure to massive amounts of sexually explicit media may be associated with the increased prevalence of sexual experience among boys, while the increasing prevalence of the nuclear family structure might be associated with the increased prevalence of sexual experience among girls [37].

From a demographic perspective, the increasing rate of later marriage in most countries has led to an increase in premarital sex. The prevalence of premarital sex is generally higher in developed countries than in developing countries. Furthermore, it is generally higher in men than in women [38-40]. The average age of first marriage in Korea was reported to be 32.94 years for men and 30.24 years for women [41]. Therefore, increased early exposure to sexual content and late marriage might be associated with the overall increasing prevalence of sexual experience in younger generations.

Since sexual activity is commonly initiated during adolescence, it is often accepted as a normative part of the transition to adulthood. However, sexual activity can lead to negative consequences such as sexually transmitted infections and unwanted pregnancies [42]. Sexual activity in adolescents is more dangerous than sexual activity in adults, not only due to adolescents’ relative physical, emotional, and cognitive immaturity, but also due to their tendency to engage more frequently in risky behaviors such as unprotected intercourse [43]. A study of data from a European population reported that in Western, Central, and Eastern Europe, 40% of adolescents who have had intercourse did not use condoms when they most recently engaged in sexual intercourse [42]. Links between early sex and other risky behaviors including smoking, alcohol drinking, and drug use have also been reported in previous studies [44,45].

Existing studies have suggested that the trend toward a younger age at sexual initiation in general is common. Thus, patterns of adolescents’ sexual behaviors appear to be becoming increasingly homogeneous in developed countries [46]. The results from our study using KYRBS are consistent with those of previous studies that showed that the prevalence of sexual experience at age 12 has generally increased in recent cohorts. Moreover, Lee [47] reported that 22.3% of adolescents had their first sexual experience in elementary school, 39.4% in middle school, and 38.2% in high school. This implies that the prevalence of sexual experience in Korean adolescents may continue to increase in the future. This might be due to urbanization, industrialization, and informatization in the past few decades in Korea. These changes might have enabled young generations to be exposed to massive amounts of information through the mass media [37]. Although the prevalence of sexual experience in Korean adolescents is lower than has been reported in data from Western countries, the growth in its prevalence is steep and steady for Korean society, which is strongly Confucian. Findings of previous studies have documented benefits of delaying intercourse, at least into the late teens and ideally into early adulthood [48,49].

Previous findings have suggested the importance of enhancing problem-solving skills, health-promotion behaviors, and appropriate education [50]. Jessor & Jessor [50] observed that adolescents with poor problem-solving skills are more likely to engage in behaviors such as precocious sex. Because problem-solving involves perceiving, processing, and using information regarding oneself, it plays an important role in behavioral health. In general, problem-solving comprises motivation, perceived self-efficacy, and the ability to regulate cognitive, behavioral, and emotional reactions when solving problems. Therefore, rather than merely regulating the sexual experience of adolescents through traditional education that emphasizes chastity, it is important to teach decision-making and problem-solving skills to prevent unwanted sexual behavior and to provide the necessary knowledge for safe sex.

In conclusion, although the prevalence of sexual experience among Korean adolescents from 2006 to 2017 slightly decreased, the prevalence of early sexual experience is increasing. Thus, it is necessary to strengthen sex education in schools to include not only contraception or chastity, but also life skills such as refusal and self-decision-making skills from an early age.

There are several strengths of this study. First, it used representative data from the KYRBS. Second, in terms of methodology, we employed an APC model and joinpoint analysis to analyze long-term trends in the prevalence of sexual experience at various ages. This method allows researchers to analyze age effects; period effects, defined as the social and environmental effects on everyone in a particular period; and birth cohort effects, which only specific birth cohorts experience, as separate risk factors. However, separating these 3 effects is challenged by the statistical collinearity among them. We solved this problem by using the IE method. Since our data on the prevalence of sexual experience were drawn from self-report measurements, the validity of the data has limitations. Thus, the results of this study require cautious interpretation.