DHS are nationally representative that collect information on a wide range of public health related topics such as anthropometric, demographic, socioeconomic, family planning, and domestic violence amongst others. The survey covered men and women aged between 15 and 49 years and under-5 children residing in non-institutional settings. NDHS 2008 and NDHS 2013 were the third and fourth round of DHS to be conducted in the country. In Nigeria, DHS field work is implemented by the National Population Commission (NPC) with the financial and technical assistance by ICF International provisioned through the USAID-funded MEASURE DHS program. For sampling, a three-staged stratified cluster design was employed which was based on a list enumeration areas (EAs) from the 2006 Population Census of the Federal Republic of Nigeria. EAs are systematically selected units from the localities, which constitute the local government areas (LGAs). LGAs are subdivisions of each of the 36 administrative states (including the Federal Capital Territory called Abuja) and classified under six developmental zones in the country. EAs were used to form the survey clusters called primary sampling units. A more detailed version of the survey was published elsewhere (21).

Main inclusion criteria were being 15–49 years of age regardless of current marital status. In total 60,611 observations from last two rounds of the surveys met these criteria (28,502 respondents from 2008 DHS, and 32,109 from 2013 DHS).

The outcome variable was number of lifetime sexual partners. It was measured based on the responses to the question: total lifetime number of sexual partners. For analysis it was categorized as: 1, 2–3, and >3 partners.

The main explanatory variable was age at sexual debut, which was measured based on the responses to the question: age at first intercourse. The cut-off for early sexual debut ranges from <15 (22), <16 (23) to <18 years (23) in different studies. For this study, we measured the prevalence of both thresholds and categorized as: <15 years, 15–17 years, ≥18 years.

Besides the main explanatory variables, the following factors were likely to be correlated with age at sexual debut and number of sexual partners: Age groups (15–24/25–29/30–34/35–49), Region: North Central/North East/North West/South East/South South/South West, Residency (rural/urban) usually determined based on community population size, availability of basic amenities and predominant occupation, Religion (Christian/Islam/Other), Education (Nil/Primary/Secondary/Higher), DHS obtain wealth scores by principal components analysis, based on a list of household assets owned by respondents, which could include; type of floor, wall, roof, water supply, sanitation facilities, radio, electricity, television, refrigerator, cooking fuel, furniture, number of persons per room. Based on the weighted wealth scores, households are grouped into five wealth categories; poorest/poorer/middle/richer/richest.

Data were analyzed using SPSS version 24. The dataset was screened for missing values and then merged to perform pooled analysis. Owing to the clustered structure of DHS data, we used complex survey design method for all analyses. The prevalence of those experienced first intercourse before 15, 15–17 years and at 18 years or above were described by percentages with 95%CIs. Same was done to represent the prevalence of lifetime sexual partners. The basic sociodemographic variables were also described alongside Chi-square bivariate tests to examine the group differences (age of sexual debut below 15, 15–17, and at least 18 years, and number of sexual partners 1, 2–3, >3). The variables that showed significance at p ≤ 0.25 in the bivariate tests were retained for final regression analysis. The p ≤ 0.25 was chosen to allow the inclusion of more explanatory variables in the regression model. However, the level of significance in the regression model was set at p < 0.05. The association between age of sexual debut and number of sexual partners was calculated by multinomial logistic regression. The first model included no other variables to measure the unadjusted odds ratios, while the second model controlled for the potentially confounding variables to produce the adjusted odds ratios.

We did the analyses using publicly available data from demographic health surveys. Ethical procedures were the responsibility of the institutions that commissioned, funded, or managed the surveys. All DHS surveys are approved by ICF international as well as an Institutional Review Board (IRB) in respective country to ensure that the protocols are in compliance with the U.S. Department of Health and Human Services regulations for the protection of human subjects.

Table 1 summarizes the basic sociodemographic characteristics of the sample population. Mean age of the respondents was 30.68 (9.0). Most of them were aged between 25 and 29 years (21.1%), from North West region (26.8%), rural residents (66.9%), followers of Islam (51%), and had no formal education (43.9%).

Median age at sexual debut was 16 years, which tended be lower among those currently aged below 19 years, located in the North East and North West region, rural residents, followers of Islam, had no formal education, and from households with lower wealth status.

As shown in Table 2, the prevalence of those having first sexual intercourse before reaching 15 years, 15–17 years, and after 17 years was respectively 26.6% (25.8–27.4), 29.2% (28.6–29.8), and 44.2% (43.3–45.2); The likelihood of first intercourse before 15 years and between 15 and 17 years were significantly higher among those, located in the North West region, living in rural areas, followers of Islam and had no formal education. In contrast those who experienced first intercourse at 18 and above were more likely to be from North Central region, followers of Christianity and had secondary level education. Prevalence of sexual debut before reaching 15 years were 45.4% (95%CI = 42.9–47.9) and that of between 15 and 17 years was 49.8% (95%CI = 47.8–51.7).

Results from Table 3 showed that having 1, 2–3, and >3 lifetime sexual partners were 67.8% (67.0–68.6), 27.8% (27.2–28.5), and 4.3% (4.0–4.7), respectively. Regarding the number of lifetimes sexual partners, those who were monogamous were more likely to be from North West, followers of Islam, had secondary level education, and lived in poorest to poorer households. While the prevalence of having 2–3 and more than three were highest among those aged between 25 and 29 years.

Table 4 contains the results of the association between age at sexual debut and number of lifetime sexual partners. It shows that those who had first intercourse before reaching 15 years and between 15 and 17 years had higher odds of having 2–3 and more than 3 lifetimes sexual partners. In the second model after adjusting for all potential confounding variables, compared with those who had first sex after age 18 years, the odds of having 2–3 lifetime sexual partners were 1.4 times higher among those who had first sex before reaching 15, and 1.5 times higher among those who had first sex before 18 (between 15 and 17 years). The corresponding odds of having >3 were respective lifetimes partners were 1.3 and 1.7 times higher among those who had first intercourse before 15 and before 18.

Based on the analysis of Nigeria Demographic and Health Survey, the findings of the study suggest a considerably strong relationship between early sexual debut and lifetime experience of having multiple sexual partners. More than half of the respondents reported experiencing sexual debut before reaching 18 years, among which more than quarter reported same before becoming 15 years. Median age at sexual debut was 16 with significant disparities among different socioeconomic and regional groups. For instance, the percentage of women who had experienced first intercourse before 15 years and between 15 and 17 years were significantly higher among those living in the Northern region compared with those in the Southern region. This difference might be explained by the educational, economic, and healthcare disparities between these regions. Inhabitants in the Southern part of the country usually enjoy better socioeconomic prosperity and better healthcare facilities. The relationship between sexual debut, fertility, and educational status is a bidirectional one. While getting married or becoming pregnant usually mark the end of school life among adolescent girls in most societies in Africa, not being a school-goer can also enhance the possibility of early marriage/sexual experience. Evidence of early marriage among girls who are poor, have low education and living in rural areas are available from Asian settings (24). Thus, it is arguable that higher rates of early sexual debut and having multiple sexual partners in the Northern region are reflection of the socioeconomic disparities in comparison with the Southern. These factors should be taken into consideration when designing intervention programs.

Significant differences in the age at sexual debut and MSPs were observed for religious groups as well. The relationship between religion and individual's sexual behavior is not straightforward, and may vary substantially across gender, family structure, level of education. However, there are empirical evidences on the positive role religious beliefs delayed onset of sexual intercourse among adolescents (25, 26). These findings imply the role that religious institutions can play in influencing the collective standards and values that direct individual sexual behavior. There are several religious teachings that stand against certain medical practices such as abortion and contraceptive use, and this could prevent women in having sexual affairs capable of leading to pregnancy outside legal union including marriage. Christianity movement, for instance, has promoted the dissemination of clear doctrines and standards, along with disciplinary sanctions, especially to the sexuality of its followers in regard to fornication. Besides, religious groups have used various resources to create room to encourage young followers to participate in a religious environment. Dating groups, youth group, counseling services, and other activities have frequently been established in religious groups, which may have been successful at creating channels through which religion can influence the lives and behavior of young people and adolescents.

Comparison of early sexual debut from previous studies is tricky, as the definition of early age seem to vary across regions. In a recent study on secondary school students in Ido-ekiti, of South-West Nigeria found that the mean age of sexual debut was 13.10 with more than two-third of the students had early sexual debut (27). In contrast with the prevalence of sexual debut of 26.6% (<15 years) and 29.2% (<17 years) in our study, the rates were lower than in Zimbabwe (11.8% before reaching 18 years) (28), but higher than in Malawi (29.6% before reaching 16 years), South Africa (39% before reaching 16 years) (29), and Tanzania (57.8% before reaching 15 years) (30). Our finding on the association between age at sexual debut and having multiple sexual partners among women is in line with several previously published studies. Multiple Indicator Cluster Survey study on Vietnam found significant association between early sexual debut with having lifetime multiple sexual partners (31). The same study also reported that first sexual intercourse before 19 years increased the likelihood of having multiple sexual partners by a factor of five. In south Africa, MSP was found to be significantly more common among men and women who reported early sexual debut compared with those who had late sexual debut (29). In the U.S., a longitudinal study on minority adolescents also reported that intercourse at an early age significantly increased the likelihood of high risk behaviors including having MSPs (32).

Findings of this study on Nigeria support the findings from other countries in Africa regarding the positive the relationship between age at sexual debut and MSPs. The findings underscore the importance of taking programmatic measures to discourage early sexual debut among women. Currently there is insufficient evidence on sexual behavior among men and women in Nigeria. The country has in place since 2000 a national health policy to preventing adolescent behavioral risk factors of STIs, including HIV, associated adolescents. Despite that, developing intervention programs have been hampered by inadequate information on contextual factors associated with sexual behavior of adolescents (15). There is a need for more in-depth researches to explore the underlying cultural and sociodemographic factors that might be contributing to early sexual intercourse in Nigeria. Given the high prevalence rates of HIV/AIDS in the country, policy makers should invest on developing strategies to prevent of risky sexual behavior as an effort to reduce the risk of transmission. Promoting school-enrolment among adolescent girls can prove to be highly rewarding in this respect. Special attention should be given to adolescent girls contribute to a large proportion of all clandestine abortions (10) and HIV sero-prevalence in the country (14).

We have several important limitations to declare. Firstly, this was a cross-sectional study which means that the associations cannot guarantee any causal relationship. The datasets were secondary; therefore, we had no control over the selection and measurement of variables. It is very likely that some important variables were not adjusted for, especially those relevant to sociocultural values. The analysis includes only women, therefore the findings cannot provide the scenario for the general population. Lastly, the possibility of recall bias cannot be ignored as most of the respondents were above adolescent age. These limitations notwithstanding, this study provides important insights on sexual behavior among Nigerian women. This is the first country representative study to show the association between age at sexual debut and number of life times sexual partners among Nigerian women. Sample size was large and findings are generalizable for women aged between 15 and 49 years. As previous studies used different thresholds for early sexual debut, we used two different cut-offs instead of one.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.