Adolescents and young women's sexuality continues to be a global source of social concern due to its link to negative outcomes such as adolescent pregnancy and sexually transmitted infections. Sexual debut is a critical life event in the life of a young woman worldwide. Modern lifestyle changes brought on by civilization and urbanization have had a significant impact on sexual and reproductive health. Thus, the timing of an adolescent's first sexual encounter is a critical influencer of those negative outcomes, hence the need to study the determinants of early sex debut among adolescents. Research has shown that early age at first sex, before 16 years as per WHO definition marks the initiation of exposure to unwanted and complicated pregnancies, HIV and sexually transmitted infections, unsafe abortions, unwanted and teenage pregnancies, and increased incidence of multiple sexual partners ^{1– 4} . Early sexual activity is a global public health problem that is particularly prevalent in developing countries ^{5, 6} .

Early sexual debut is a major public health concern around the world among adolescents and young women. A 50 countries study on correlates of early sexual initiation among adolescents aged 12–15 years reported that 14.2% of adolescents had early sexual initiation ⁷ . In Brazil, a study among adolescents found that the prevalence rate of early sexual debut was about 18% among boys and 7% among girls ⁸ . About 27% among boys and 20.7% among girls in the United States of America reported being sexually active ⁹ . Among the Jamaican adolescents, girls had an earlier mean age at sexual debut (11 years) than boys (15 years). In Ohio State in the USA, 31.2% of the adolescents were found to be sexually active before age 15 ⁴ .

In Sub-Saharan Africa, about 54.3% of adolescent women aged 15–19 years reported having had sex before the age of 18 and 12.3% before the age of 15 ¹⁰ . A study on the impact of sexual behaviors of adolescent girls in 12 sub-Saharan African countries showed that 19.9% of all adolescent girls had made their coitarche, while their mean age at sexual debut was 13.2 years old ¹¹ . In central Africa, 32.4% and 8.6% of the girls had a sexual debut before the age of 18 and 15 years respectively ¹⁰ . In East Africa, a study on the prevalence and associated factors of early sex debut found the prevalence rate of early sex debut to be 21.1% ¹² .

In Ghana, a study on sexually active women aged 15 to 24 years found that more women 33.1% than men 26.4% were more likely to initiate sex earlier ³ with a national prevalence of early sexual debut among unmarried women of 56.9% ⁵ . The prevalence of early sexual debut among adolescent girls in Mozambique was 45.8%, 43.6% in Namibia and 7.0% in Swaziland ¹¹ . In Ethiopia, a study assessing the timing of sexual debut and its associated factors found that about 45% of the adolescents of 10–19 years old who were not in school had reported commencement of sex ¹³ . It was also evident that children who are semi or complete orphans are consistently more likely to report younger age of sexual debut ¹⁴ . In Sierra Leone, orphan girls were more likely to have had sex and have multiple sexual partners than non-orphans ¹⁵ . A multi-country study found that in Burkina Faso, Ghana and Malawi, women who were total orphans had earlier sexual debuts compared to non-orphans ¹⁴ . In Kenya, a study on the impact of multisector cash plus programs found that 21% of adolescent girls reported ever having sex ¹⁶ , further ¹⁷ , found that about 50% of the adolescents and young women had ever had sex.

Several factors have been identified to influence the timing of first sexual commencement among adolescents. Findings from several studies have shown that demographic, socioeconomic status, peer influence, mothers' age at first sex, substance and alcohol abuse, large family size and religious factors among others have an impact on sexual debut among adolescents and young women ^{5, 13, 14, 18– 20} . One in every three girls aged 15–19 years compared to two in every three girls aged 20–24 years reported ever having had sex ¹⁷ . Schooling was also found to be protective against early sexual debut. The longer the adolescents and young women are in school, the less likely they are to initiate early sex ¹⁸ . Also, adolescents who live with their fathers are less likely to have an early sex debut or even engage regularly in sexual activity ¹¹ .

Various studies evaluating early sexual debut among adolescents and young women have been conducted in Kenya ^{6, 11, 12, 17} . The majority of these studies concentrate on adolescents in HIV settings, adolescents in school-based programs, data from the 2014 Kenya Demographic and Health Survey ^{21, 22} and data from surveys specific to certain counties like Nairobi, Kisumu and Mombasa ²³ . This study fills the gap by using the PMA data which is nationally representative to identify the correlates of early sexual debut among adolescents and young women in Kenya.

From the 2019 survey, a total of 3,706 adolescents and young women aged 15–24 years, who consented to voluntary participation were interviewed. More than half of the participants were adolescents (55.2%) while the rest were young women. About 32% of the participants were married, 72% were from rural areas and about 6 in every 10 had attained secondary or higher level of education. About 21% were from the lowest wealth quintile, while 15% were from the highest wealth quintile. About 29% were using a contraceptive while 97% were knowledgeable about contraceptives at the time of the survey. Most of the respondents were Protestants 69% while 18% were Catholics ( Table 1).

We found that 10.2% of adolescents and young women in Kenya had early sexual debut. Early sex initiation was more prevalent among respondents from West Pokot County 25%, unmarried sexually active respondents 18%, and among parous and respondents from the least wealth quintile 17%. Educational attainment, ever-given birth, contraceptive use and wealth status were significant determinants of early sexual initiation among adolescents and young women in Kenya.

Prevalence of 10.2% in our study is lower than that reported from other Sub-Saharan Africa, ranging from 55% in Ghana, 35.7%, in Nigeria, 21.14% in Mozambique, while in East Africa the prevalence rate was reported to be 21.14% ^{2, 5, 12} . Another study conducted in Kenya in a different setting found the prevalence to be 6.7% among youths aged 18–24 years ²⁷ .

Like in other studies, educational level, ever-given birth, contraceptive use and wealth status were determinants of early sexual initiation in Kenya ^{12, 27– 29} . Higher education attainment has been shown to have an impact on the likelihood of experiencing decreased early sexual initiation. These findings were supported by previous studies ^{9, 12, 30} . Higher education levels may be empowering adolescents and young women to make informed decisions, thus avoiding early sexual debut. Longer duration of schooling has also been reported to be protective against adolescent early sex debut and pregnancies ^{2, 18} . Further, education reduces the possibility of indulging in risks such as substance abuse, which may expose them to early sexual initiation ³¹ .

Respondents from the highest wealth quintile households were less likely to engage in early sex than respondents from the other household quintiles. This is consistent with other studies ^{7, 19, 30} that have shown women from high wealth quintile households delaying their sexual debut. This may be due to women from low-income families engaging in earlier sexual relations for monetary gain and other benefits, unlike those from wealthy families who have good quality of life and better health-seeking behavior ^{12, 32} .

A strong association exists between early sexual debut and ever giving birth. Young women who had given birth were 2-times more likely to have had an early sexual onset than their counterparts. Women who had early sexual debut likely did not use contraception at their sexual onset, and hence the likelihood of pregnancy.

The odds of early sexual debut among contraceptive users were 1.4 higher as compared to non-users (AOR = 1.39, 95% CI (1.01 - 1.90). Other studies have shown that women with a contraceptive use history were likely to engage in early sex ^{5, 12} , as it this would assure them of pregnancy protection hence the early sexual debut. The positive association of early sexual and contraceptive use could be indicative of a functional health system where sexually active adolescents are more likely to receive family planning/contraceptive services.

There are some limitations which needed to be considered when interpreting the results of this study. First, though we strived to include both in and out of school adolescents and young women, the sample may not be representative of all adolescents and young women. Sexual activities might vary between those included and those not included. Second, because of the cross-sectional nature of the study, we cannot establish the temporal direction of associations between early sexual debut and various factors. Third, information about sexual behaviours was collected using self-reported measures, which can introduce reporting recall bias.

The prevalence of early sexual debut among adolescents and young women is influenced by personal factors. Our findings suggest there is need to implement targeted program which can address covariates of early sexual debut including strengthening information, education and communication on adverse consequences of early sexual initiation and improving contraceptive knowledge of teenagers as well as promoting schooling.

Early sexual debut has life-long consequences

Adolescent stage is an experimental stage which may include early sexual debut

Contraceptives can prevent some of the consequences of early sexual debut

That among a general population generated via household survey of adolescents and young women in Kenya, early sexual debut remains a public health problem at 10.2%

There are covariates at different levels which if addressed could mitigate and reduce early sexual debut and its consequences.