In Côte d'Ivoire, the HIV epidemic is generalized. The country has the highest HIV prevalence in West Africa, with an estimated stable adult population prevalence of 7.1% [1]. In the country, the disease is mainly spread through heterosexual sex and now ranks second among the leading causes of death after malaria. The 2005 Côte d'Ivoire AIDS Indicator Survey (2005 AIS-Côte d'Ivoire) also revealed that premarital sexual intercourse is prevalent among youth in Côte d'Ivoire [2]. About 58% of female youths and 56% of male youths had premarital sex during the 12 months prior to the survey thereby increasing the risks associated with youth sexual activity such as sexually transmitted infections (including HIV/AIDS) and adolescent pregnancy [3–5]. Hence, the HIV prevalence rate is high among the youngest generation with females experiencing greater risks than males and rises with age reaching a peak of 10.4% for those ages 30 to 34 [2]. Moreover, because AIDS has a long latency period from infection-the average time between HIV infection and AIDS diagnosis is 7 to 9.8 years [6]-we assume people in this age group with AIDS were infected by HIV in adolescence (e.g., between the ages 15–24).

Understandably, efforts have been made to curb the spread of the infection. HIV/AIDS prevention programs promote sexual abstinence and delayed initiation of sex among never-married youth [7, 8].

Unfortunately, sexual abstinence has not been clearly defined or fully examined in the literature [9, 10]. Further, there is no consensus about whether sexual abstinence is a health protective behavior or something more inclusive [11]. Some researchers take a public health approach, defining abstinence as refraining from specific types of sexual contact [10]. Others are more inclusive, incorporating attitudes, moral and religious beliefs, and lifestyle choices into their definition. In one program evaluation, adults defined abstinence in behavioral terms (no vaginal, anal or oral intercourse), and youth also listed use of alcohol, cigarettes, drugs and pornography as incongruent with an abstinent lifestyle [12]. Moreover, there is no consensus on the specific sexual behaviors that define abstinence. Many adolescents and young adults do not define oral sex and other non-coital behaviors as "sex" [13, 14].

Finally, contextual and developmental issues influencing abstinence are poorly understood. Research shows that adolescents differ in beliefs, attitudes, and sexual behaviors by age, gender, race/ethnicity, educational status and sexual experience [15–23]. Family and peer factors are also particularly influential in predicting adolescent sexual behavior [24–26].

Studies of interventions to delay sexual onset find differences in outcomes between males and females, and between sexually experienced and inexperienced participants [27–29]. However, though documented epidemiologically, little is known about why and how these differences arise. Besides, the reasons why some adolescents abstain from sexual intercourse and others do not are still unclear. Therefore, research need to more clearly define sexual abstinence, as well as those factors that may promote or discourage these early activities.

This research sought to extend previous research by examining the theoretical and empirical evidence of the roles of specific socio-demographic, and other contextual factors on abstinence behavior among never-married individuals aged 15 to 24 in Côte d'Ivoire. Due to the paucity of information available on the sexual behavior of this group, particularly in the Sub-Saharan Africa region, findings of this research increase the understanding of youths' sexual abstinence behavior and can enhance preventative measures.

More male youths significantly reported that they knew someone who had AIDS or had died of AIDS. Less than one-tenths of youths reported that most young men and women they knew were abstaining from sex until marriage. Furthermore, youths had the perception that faithfulness was more common among married women than among married men (47.3%, compared to 18.6%).

Interestingly, HIV/AIDS-related knowledge was quite high among the sample. Regarding the questions about knowledge of preventive methods (including questions about abstinence, faithfulness and condom use) 80.1% of females and 78.2% of males knew that abstinence can reduce the risk of becoming infected. About 78.8% of females and 79.7% of males knew that being faithful to one uninfected partner can reduce the risk of HIV infection, and 73.3% and 80.2%, respectively, knew that condoms use is a way to reduce the risk of HIV infection. The HIV/AIDS transmission knowledge was also high among never-married youths. Globally, the HIV/AIDS-related knowledge showed male youths had higher scores than female youths (4.2 compared to 4.0).

Thus, sexual abstinence practice significantly decreased with age among never-married Ivorian youths. Females of higher education level were significantly 11.14 times as likely as those of no education to practice either primary or secondary abstinence. Males who were animists, had no religion, or were practicing religions other than Christianity or Islam were significantly less likely than other male youths to practice sexual abstinence (OR = 0.53, 95% CI = 0.30–0.95). Living in the north-west region of the country significantly decreased the odds of sexual abstinence among never-married female youths. Likewise, female youths living in rural areas, compared to females in urban areas, were significantly more reluctant to practice sexual abstinence (OR = 0.42, 95% CI = 0.24–0.75). Knowing someone who has AIDS or has died of AIDS significantly decreased the odds of sexual abstinence practice among male youths (OR = 0.46; 95% CI = 0.26–0.80). The perception of delayed premarital sex practice among young men and women was positively related to sexual abstinence among female youths. For instance, female youths who reported that most young men they knew waited for sex until marriage were 1.85 times as likely as other females to practice either primary or secondary abstinence (albeit marginally). The perceived practice of faithfulness among sexually active unmarried men increased the likelihood of sexual abstinence practice among male youths (OR = 1.61; 95% CI = 0.81–3.19), while significantly decreased that of female youths (OR = 0.26; 95% CI = 0.07–0.98). Furthermore, female youths who reported that most married sexually active men they knew only had sex with their wives were 1.94 times as likely as the other females to practice either primary or secondary abstinence. In addition, male youths who knew that condom use can prevent HIV/AIDS were significantly less likely to abhor sexual abstinence behavior (OR = 0.45, 95% CI = 0.26–0.77). Knowledge of being faithful as a way to avoid HIV/AIDS was negatively associated with sexual abstinence practice among female youths (OR = 0.50, 95% CI = 0.26–0.98). Lastly, male youths who knew that a person cannot be infected by sharing food with a person who has AIDS were significantly 1.73 times more likely to adopt primary or secondary abstinence behavior.

One of the principal findings from this research was that the proportion of never-married male youths was larger than the proportion of never-married female youths, which reflect the fact that females get married at earlier ages than their counterpart males in Côte d'Ivoire [2]. We found that both male and female youths similarly adopted abstinence behavior (either primary or secondary abstinence). Other studies found that abstinence was higher among female youths [31].

Another important correlate of sexual abstinence behavior is age, with older youth being significantly less likely than their younger counterparts to report sexual abstinence practice. This points to the need to target youth with either primary or secondary abstinence messages early (preferably in the early teenage years) and thereby strengthen their capability to refuse sex before sexual activity becomes widespread. As Babalola et al. ideally put it, critical thinking, priority setting, problem solving, sexual negotiation, and coping with life challenges constitute appropriate skills that an intervention program should help youths to develop early and apply in every day situations [32].

Interestingly, higher education was significantly and positively associated with sexual abstinence among female youths, echoing the need to improve education attainment, particularly among this group, as previously highlighted [24, 33].

Female youths living in rural areas were significantly less likely than other females to abhor sexual abstinence practice. This suggests that marriages in rural areas occur at a young age and abstinence until marriage might be difficult as these areas feature communities with poor socio-economic status and less HIV/AIDS preventive behavior than urban areas. The finding also underscores the need to focus on urban youth specifically, in efforts aimed at promoting sexual abstinence. Moreover, developing holistic approaches that could shape rural female youth livelihoods is essential to empower female youth to control issues such as their sexual and reproductive health and rights.

Also notable, abstinence behavior was unevenly distributed throughout the country, with youths living in the north-west part of the country less likely to practice either primary or secondary abstinence compared to other youths. This region also has the lowest education level in the country [2], which may contribute to the high-risk sexual behavior prevalent among these inhabitants. Improving education attainment throughout the country, especially in the north-west part, is a strategy likely to foster delayed sexual debut.

The finding that male youths who were Christians or Muslims were significantly more likely than other male youths to adopt sexual behavior is quite interesting and not surprising. From the standpoint of these religions (Christianity and Islam) on sexual abstinence, sexual intercourse is meant to take place with the context of marriage; therefore, sexual abstinence is expected of unmarried people. Extensive researches have also documented the major impact of these religions on adolescents' sexual behavior [34, 35]. Therefore, HIV/AIDS prevention could include religious leaders in Côte d'Ivoire to serve as reliable medium for providing comprehensive sex education not only to their congregations, but also to the larger community. Besides, these religious institutions possess some resources such as schools, hospitals, clinics and orphanages that can be tapped for HIV/AIDS prevention programs aimed at encouraging youth to postpone the onset of sexual activity at least until marriage.

The present study also showed that, contrary to youths' perceptions, a substantial proportion of them were postponing sexual debut until marriage. Since the perceived practice of abstinence among young people who female participants knew was positively associated (albeit marginally) with abstinence practice, it is wise to design appropriate messages and activities in order to correct and provide visibility to positive deviants (youths who are delaying sexual debut or practicing secondary sexual abstinence). For instance, abstinent youths should be encouraged to be proud and vocal about their behavior and to advocate in favor of sexual abstinence behavior among their peers. Another study mentioned interpersonal channels as an effective means to persuade individuals to adopt new behaviors [36]. Understandably, the use of interpersonal channels could form a substantial part of any communication campaign in Côte d'Ivoire.

However, this study has several limitations. First, the study used a cross-sectional design; thus, it is impossible to establish causality. A study using a longitudinal design is necessary to assess the significance and stability of predictors of abstinence behaviors over time. Second, the study is based on a secondary analysis with archival data, which likely reflects the perspectives and purposes of the original investigators. We were challenged to shape the data to match our research questions, which required an intensive process of understanding the data set, recoding variables, and recasting research questions to match data available. Other limitations include self-reported information, which increases the possibility of inaccuracies, particularly with regard to information about HIV/AIDS and sexual behavior. And given an increasing social pressure to delay sexual initiation and avoid pregnancy or sexually infectious diseases such as HIV/AIDS, youths may be likely to underreport their sexual activity. Therefore, caution should be used when generalizing these findings. Lastly, many studies of youth sexual behavior, including the current one, only ask about participation in sexual intercourse, leaving a gap in our knowledge regarding other sexual behaviors that young people might participate in, such as oral and anal sexual behaviors [37]. Thus, further research should attempt to measure behaviors other than sexual intercourse.

In spite of these limitations, the findings of this study are important. They contribute to the field by examining the topic of youth abstinence behavior in a sub-Saharan African context.

Youth sexual abstinence is shaped by a number of interlinking forces that include individual and familial influences, as well as a network of more indirect forces, such as peers, school and community. Better identification and understanding of the interplay between these individual-level characteristics and a variety of social and cultural contexts is necessary for programmatic youth development.

Our findings demonstrate that age, education, and some religious institutions play a key role in promoting sexual abstinence. Living in rural areas decreased the likelihood of practicing sexual abstinence. The perceived practice of abstinence among young people who female participants knew was positively associated (albeit marginal) with abstinence practice.

HIV/AIDS prevention program components could include media campaigns, educational intervention improvement, as well as promoting policies that shape female youth livelihoods. Likewise, youth involvement at early ages in initiatives to design appropriate messages, and activities to promote positive behaviors or to change negative perceptions could impact on youths' decision to exert abstinence behavior. Lastly, the inclusion of Christian and Islamic leaders in HIV/AIDS work can be beneficial.