Socio-Demographic, Economic and Psychological Correlates of Risky Sexual Behaviour Among Sexually Active Young People in Nigeria

This study sought to identify the socio-demographic, economic, and psychological factors associated with risky sexual behaviour among sexually active youths in Nigeria with the view to providing more empirical information for the development of more effective interventions to improve safe-sex practices and the sexual health of the young people in Nigeria. The study analyzed the male and female datasets extracted from the 6 round of the Nigeria Multiple Indicator Cluster Survey data (MICS) (n=7,909) using descriptive statistics and multiple binary logistic regression to achieve the study objectives and test hypothesis. The results showed that 66% of the youths have had sex before reaching 18 years, 77% had unprotected sex, and 32% have had more than one-lifetime sexual partner. The significance of the association between socio-demographic (age, sex, marital status, ever fathered/mothered, awareness of AIDS, ethnicity, residence, and region), economic factors (employment status and wealth index), and risky sexual behaviour differ by the category of risky sexual behaviour. Overall psychological factor (satisfaction with life) was a significant correlate of the lifetime number of sexual partners. This study concludes that socio-demographic, economic, and psychological factors were predictive of risky sexual behaviour among young people in Nigeria. However, the significance of these predictors differs by type of risky sexual behaviour. The study recommends that more effective sexual health interventions must also address the prevalent psychological risk factors among young people in Nigeriaapart from different background characteristicswhich could predispose them to risky sexual practices.


Introduction
Globally there has been an exponential increase in the numerical size of young people. Young people constitute about a fifth of the global population (Sekoni & Soyanwo, 2014;UNDESA, 2018). Sub-Saharan Africa has been implicated as a significant contributor to the rising global population size, given the regions' widening base of her population pyramid.
A more in-depth examination of the population statistics in Africa reveals that Nigeria-the most populous country in Africa-has a large and teeming population of young people as a result of years of an increasing rate of natural increase and high fertility rate (NBS, 2017) resulting in high population momentum. This high fertility is not unconnected with the teeming population of sexually active young people who constitute the population exposed to risky sexual practices with a higher risk of having unwanted pregnancy and procuring a clandestine and unsafe abortion. Therefore, risky sexual practices among young people in Nigeria presents a disheartening situation that portends a significant threat to the sexual and reproductive health of many young people in Nigeria (Imaledo, of sexual activity among young people, many of whom had their sexual debut and had become sexually active at an early age (Manjula & Dutt, 2017). For instance, more than 40% of youths in Nigeria have ever had sex. Besides, data on the timing of last sexual intercourse among these young people also showed that more than 30% of those aged between 15-19 had last sexual intercourse in the last four weeks compared to 59% of those aged between 20-24 in the same period (NPopC & ICF Macro, 2014); while the average age at sexual initiation among females and males are 17.2 and 21.7 respectively (NPopC & ICF Macro, 2018). It has been observed that young people across the world struggle with risky sexual behaviour which takes different forms such as having multiple partners, having risky casual or unknown sexual partners, early sexual initiation, engaging in transactional sex, forced sexual intercourse, having sex under the influence of alcohol or other stimulating substances, having sex immediately after watching pornographic media and having unprotected sexual intercourse (Aliza Lodz et al., 2019;Berhan & Berhan, 2015;Odeigah et al., 2019;Sekoni & Soyanwo, 2014;Tarkang, Pencille, Amu, Komesour, & Lutala, 2019;Urassa, Moshiro, Chalamilla, Mhalu, & Sandstrom, 2008;Wendland et al., 2018). Unfortunately, in most cases, these actions result in deleterious health outcomes among which are HIV/AIDS and other sexually transmitted infections (STIs), unplanned pregnancies and unsafe abortions (Chawla & Sarkar, 2019;Pinyopornpanish et al., 2017). Again, empirical evidence holds that about 33% of New Infections for HIV in 2017 are among young people within the age of 15-24 (UNAIDS, 2018).
In view of these deleterious effects of risky sexual behaviours (among adolescents and youths in Nigeria) as well as its negative implication for achieving the UN Sustainable Development Goal (SDG) three, which focuses of promotion of health and well-being (UNDP, 2015), understanding the determinants of sexual behavior among young people would be of great benefit to the development of programs and interventions for the prevention and treatment of sexually transmissible infections (STIs) (Nguyen, Subasinghe, Wark, Reavley, & Garland, 2017). Despite the vast amount of studies geared towards investigating and understanding risky sexual behaviours among young people, understanding the processes linking risky sexual behaviours with its predictors remains limited in Nigeria. Previous studies have identified numerous factors associated with risky sexual behaviours among young people including age and gender, with men having higher odds of engaging in unprotected sexual intercourse (Abosetugn, Zergaw, Tadesse, & Addisu, 2015;Ganczak, Czubinska, Korzen, & Szych, 2017;Hadish, Mao, Gong, Hadish, & Tesfamariam, 2017;Noubiap et al., 2015;Sambisa, Curtis, & Stokes, 2010;Sekoni & Soyanwo, 2014;Shek, 2013). Besides, the relationship status of an individual has also been found to significantly predict sexual behaviour; single people have higher chances of having unprotected sex (Ganczak et al., 2017;Hadish et al., 2017). Studies have also found that the use of alcohol and substances like tobacco and marijuana is associated with risky sexual behaviour among young people (Chawla & Sarkar, 2019;Wendland et al., 2018;Yaw Amoateng, Kalule-Sabiti, & Arkaah, 2014).
Studies have also found that socio-economic status, living conditions, educational attainment, and ethnocultural factors are some of the factors that are associated with risky sexual behaviours among young people (Berhan & Berhan, 2015;Wendland et al., 2018;Yi et al., 2014). Furthermore, the extent of the influence of peer pressure on young peoples' engagement in risky sexual behaviours has also received scholarly attention (Cherie & Berhane, 2012). Yet, few studies-in Australia and Poland-have identified the psychological associates of risky sexual behaviours among young people (Nguyen et al., 2017).
In Nigeria and some other developing countries, little is known about the linkage between the self-reported level of happiness and satisfaction with life and the risk of engaging in risky sexual behaviour among young people in Nigeria considering the highly prevalent life-threatening risky sexual practices among them. Hence, this study identified the socio-demographic, economic, and psychological factors associated with risky sexual behaviours among sexually adolescents and youths in Nigeria. This study provided more empirical evidence needed for developing more impactful social and behaviour change communication (SBCC) interventions to address the risky sexual practices that have continued to subvert the sexual and reproductive health of young people in Nigeria.

Study Location
The geographic domain for the study was Nigeria. Nigeria is the most populous country in Africa, with over 180 million people clustering around 18 years of age (Bolarinwa, 2019). According to the 2018 Nigerian Demographic and Health Survey report, the median age at first sexual intercourse among men and women in Nigeria are 21.7 and 17.2 respectively while more young men aged 15-24 (than young women in the same age-group) have had multiple sexual partners in the last one year and used condom in their final sexual intercourse (NPopC & ICF Macro, 2018).

Research Methodology
The study analyzed secondary cross-sectional data. The male and female survey data were extracted from the 2016-17 Nigeria Multiple Indicator Cluster Survey (MICS), combined and analyzed in line with the study objectives. The survey was the fifth round jointly conducted by the National Bureau of Statistics (NBS) and the United Nations Children Fund (UNICEF) to provide indicators of child mortality, child health, child and maternal nutrition, reproductive health, water, and sanitation, etc. The survey methodology involved a two-stage random sampling technique and a weighting factor to adjust for under and oversampling.

Inclusion Criteria
Sexually active male and female, young people aged 15-24 were of interest in the study (n = 7,909). Thus, the female and male datasets were extracted, merged, and analyzed.

Variables
The outcome variables were age at first sexual intercourse, number of lifetime sexual partners, and condom use (in the first and second last sexual intercourses). The explanatory variables were socio-demographics (age, sex, marital status, ever fathered/mothered any children, ethnicity, residence, and zone) economic characteristics (level of education, awareness of AIDS, working status, and wealth index), and ratings of overall happiness and satisfaction with life. Indicators of psychological factors were self-reported levels of happiness and satisfaction with life. Respondents were asked to rank the level of their overall happiness and satisfaction using the 5-point Likert scale (which ranged from very happy to very unhappy for happiness, while the scale ranged from very satisfied and very happy unsatisfied).

Data Analysis
A weighted tabulation of all the respondents' socio-demographics, economic characteristics, and ratings of happiness and satisfaction with life was done. A weighted multiple binary logistic regression models fitted at different stages using Odds ratios (ORs) and the probability of type 1 error of 5% was also performed. The first stage (model 1) examined the relationship between all socio-demographic and economic variables on each indicator of risky sexual behaviour; the second stage examined the relationship between psychological factors and each indicator of risky sexual behaviour. The full model estimated the marginal effects of all the variables. All the analyses were performed with STATA version 14.
The multiple logistic regression is expressed as Where Y ij = log-odds of risky sexual behaviour (i.e. condom use in first and last sex, etc.) β 0 = Intercept (Average log-adds of risky sexual behaviour) β 1, β 2 = Coefficients for explanatory variables X 1, X 2 = Explanatory variables (socio-demographic, economic and psychological factors) The 2016-17 Nigerian Multiple Indicator Cluster Survey datasets are freely available. Therefore, no ethical approval was required. However, approval for use was obtained from https://mics.unicef.org/surveys before analysis.  Table 1 presents the distribution of the respondents (in 15-24 years and sexually active) by risky sexual behaviour. More than two-thirds of the respondents had sex for the first at below 18 years of age (66%) had only one partner in their life (67%) respectively while majority (77%) of the respondents said they never utilized any contraceptive methods the first and the last time they had sexual intercourse.  Table 2 presents the distribution of respondents by socio-demographic and economic characteristics. Above two-thirds of the respondents (69%) were in the age group 20-24. Majority of the respondent were females (84%); more than two-thirds (69%) were ever married; slightly below two-thirds (58%) had ever fathered or mothered any children; a slight below half of the respondents had at least a secondary education; majority were not working (74%); less than half of the respondents were in poor and rich with 42% and 37% respectively. Besides, a slight below half (49%) of the respondents were of the Hausa tribe; majority were rural residents (72%). Majority (71%) resided in the northern region of Nigeria and ever heard of AIDS (84%).  Table 3 shows the distribution of the respondents by their levels of overall satisfaction and happiness with their lives. Majority of the respondents claimed they were satisfied (88%) and were happy (91%) with their lives, respectively.  Table 4 presents the multivariable analyses -examining the relationship between socio-demographic and economic factors and age at first sexual intercourse. It also shows results for satisfaction and happiness with life and age at first sexual intercourse. In Model I, only age, marital status, ever fathered/mothered, level of education, employment status, wealth index, and ethnicity respectively had significant relationships with age at first sexual intercourse. Respondents aged 20-24 -compared to those aged 15-19 years-were eighty-nine percent less likely to have had first sex at minor age-group (OR= 0.11; CI = 0.09 to 0.14; P<0.05). Female respondents -compared to the males-were seventeen percent more likely to have had first sex at minor age-group (OR= 1.17; CI = 0.94 to 1.45; P>0.05). The ever-married were fifty-five percent less likely to have had first sex at minor ages (OR= 0.45; CI = 0.35 to 0.57; P<0.05) compared to those who never married. Those that ever fathered/mothered children were four times as likely to have had first sex at minor ages (OR= 3.60; CI = 2.87 to 4.50; P<0.05) relative to those who never fathered or mothered children.

Results
Respondents who had heard of AIDS were twenty-one percent more likely to have had first sex at minor ages (OR= 1.21; CI = 0.98 to 1.49; P>0.05) relative to those who said they never heard about AIDS. Those with primary education were twelve percent more likely to have had first sex at minor ages (OR= 1.12; CI = 0.82 to 1.54; P>0.05) relative to those who had no education while those who had at least a secondary school education were fifty-one percent less likely to have had first sex at minor age-group (OR= 0.49; CI = 0.37 to 0.65; P<0.05) and those who had nonformal education were forty percent more likely to have had first sex at minor ages (OR= 1.40; CI = 1.01 to 1.94; P<0.05) relative to those who had no education. In terms of wealth index, there was a dose-response relationship between wealth index and age at first sexual intercourse such that the higher the wealth index the lower the likelihood of having first sex at minor ages. Respondents who reported they were not working were twenty-one percent more likely to have multiple lifetime sexual partners (OR= 1.21; CI = 1.02 to 1.42; P<0.05) compared to those working. Compared those respondents in the poorest category, those that belonged to the poorer categories were ten percent less likely to have had first sex at minor ages (OR= 0.90; CI = 0.71 to 1.15; P>0.05); the middle group was associated with a twenty-three percent increase in the likelihood of having first sex at minor ages (OR= 0.77; CI = 0.59 to 1.00; P<0.05); the richer group yielded a thirty-eight percent increased likelihood of having first sex at minor ages (OR= 0.62; CI = 0.48 to 0.81; P<0.05) and the richest group yielded a fifty-five percent increased likelihood of having first sex at minor ages (OR= 0.45; CI = 0.33 to 0.61; P<0.05).  Vol. 12, No. 8; Respondents who were indifferent (neither satisfied nor unsatisfied)-compared to those who were satisfied with their lives were one percent less likely to have had sex at minor ages (OR= 0.99; CI = 0.75 to 1.31; P>0.05) while respondents who felt dissatisfied with their lives were thirty-six percent less likely to have had sex at less than 18 years old (OR= 0.64; CI = 0.41 to 1.00; P>0.05). Similarly, Respondents who were indifferent -compared to those who were happy with their lives-were seven percent less likely to have had sex for the first time at minor ages (OR= 0.93; CI = 0.75 to 1.14; P>0.05) while respondents who claimed they were unhappy with life were twenty-seven percent less likely to have had sex at minor ages (OR= 0.73; CI = 0.54 to 1.00; P>0.05).
In the full model, there was 2% increase in the likelihood of having first sex at minor ages among respondents who felt indifferent (neither satisfied nor unsatisfied) (OR= 1.02; CI = 0.80 to 1.30; P>0.05) compared to those who felt satisfied. There was a 5% decrease in the likelihood of having first sex at minor ages among respondents who felt dissatisfied with their lives (OR= 0.95; CI = 0.65 to 1.38; P>0.05) compared to those who felt satisfied. Respondents who were indifferent were associated with eight percent increase in the likelihood of having first sex at minor ages (OR= 1.08; CI = 0.79 to 1.46; P>0.05), while those who were unhappy with their lives were thirty-two percent less likely to have had first sex at minor ages (OR= 0.68; CI = 0.35 to 1.33; P>0.05) relative to those who felt happy.    Table 6. presents the multivariable analysis examining the relationship between the socio-demographic factors, economic factors, and condom use (in first and last sexual intercourse). Also, it presents the relationship between happiness and satisfaction with life and condom use.
Only sex of respondents, marital status, ever fathered/mothered any children, ever heard of AIDS, level of education, employment status, wealth index, and ethnicity respectively had significant relationships with condom use at first and last sexual intercourse. Respondents aged 20-24-compared to those aged 15-19 years-were sixteen percent less likely to have used condom (OR= 0.84; CI = 0.70 to 1.00; P>0.05). Female respondents -compared to the males-were fifty-six percent more likely not to have utilize condom (OR= 1.56; CI = 1.27 to 1.93; P<0.05). The ever married were six times more likely not to have utilized condom (OR= 5.83; CI = 4.52 to 7.52; P<0.05) compared to those who never married. Those that ever fathered/mothered children were twenty-eight percent more likely not to have utilized condom (OR= 1.28; CI = 1.01 to 1.63; P>0.05) relative to those who never fathered or mothered children.
Respondents who ever heard about AIDS were forty-five percent more likely not to have utilized condom (OR= 1.55; CI = 1.11 to 2.17; P<0.05) compared to those who never heard. Those with only primary education were forty percent less likely not to have utilized condom (OR= 0.60; CI = 0.38 to 0.96; P<0.05) relative to those who had no education; those who had at least a secondary school education were sixty-one percent less likely not to have utilized condom (OR= 0.39; CI = 0.25 to 0.61; P<0.05) relative to those who had no education, and those who had nonformal education were sixty-two percent more likely not to have utilized condom (OR= 1.62; CI = 0.92 to 2.87; P>0.05) relative to those who had no education. Those weren't working were twenty-five percent more likely not to have utilized condoms compared to those not working (OR= 1.25; CI = 1.04 to 1.50; P<0.05).
Compared to the poorest category, the poorer were twenty-two percent less likely not to have utilized condom (OR= 0.78; CI = 0.53 to 1.15; P>0.05); the middle group was associated with a thirty-five percent increase in the likelihood of non-use of condom (OR= 0.65; CI = 0.44 to 0.96; P<0.05), the richer group were associated with a fifty-seven percent increase in the likelihood of non-use of condom (OR= 0.43; CI = 0.29 to 0.65; P<0.05) and being in the richest group was associated with a sixty-eight percent increase in the likelihood of non-use of condom There was a statistically significant association between satisfaction and condom utilization. There was two percent decrease in the likelihood of non-use of condom among respondents who felt indifferent (neither satisfied nor unsatisfied) (OR= 0.98; CI = 0.72 to 1.34; P>0.05) compared to those who felt satisfied, while there was a four percent increase in the likelihood of not utilizing condom among respondents who were dissatisfied with their lives (OR= 1.04; CI = 0.71 to 1.51; P>0.05) compared to those who felt satisfied. Happiness was not associated with condom utilization in the first and last sexual intercourse. Among the respondents who were indifferent, there was twenty percent increase in the likelihood of non-use of condom (OR= 1.20 CI = 0.84 to 1.72; P>0.05), while those who were unhappy with their lives were one percent less likely not to have utilized condom (OR= 0.99; CI = 0.63 to 1.54; P>0.05) relative to those who felt happy.

Discussion
The study revealed that socio-demographic factors exhibited striking differences in relationship with risky sexual behaviour. Specifically, young people in older age-group (20-24) were likely to have more lifetime sexual partners but were less likely to have had sex at younger ages. Also, while females were less prone to having high lifetime sexual partners (Mlambo, Peltzer, & Chirinda, 2016), they were more likely to have had unprotected sex (in the first and last sexual intercourse). That suggests a high tendency to keep a few trusted sexual partners with whom young females believe they can continue to enjoy unprotected sex (Odimegwu & Somefun, 2017), thinking they are safe. Previous studies also revealed that unmarried youths who are not into marital vow have greater chances of engaging in risky sexual behaviour (Chawla & Sarkar, 2019;Hadish et al., 2017;Wendland et al., 2018). However, what passes for risky sexual behaviour could be examined through the lens of marital status as suggested by this study in which ever-married young people were more likely to not use protection (during their first and last sexual intercourse) probably because of their desire for children. Also, the high tendency among ever-married young people to keep low number of sexual partners in their lifetime may be attributed to the high (moral-based) disregard for extra-marital affairs or the decision of some to limit their access to out-of-wedlock sexual activities.
In terms of ethnicity, it has been established that ethnicity plays a crucial role in the practice of risky sexual behaviour (Odimegwu & Somefun, 2017). This is evident in the findings that, compared to the Hausas/Fulanis, young people of Yoruba and Igbo extractions (and other ethnic groups) were less likely to engage in unprotected sex. A plausible explanation for is that these categories of young people (i.e., Yoruba, Igbo, and others) must have been educated and knowledgeable enough about using protection, and they leverage the benefits of protection (against sexually transmitted infections and unwanted pregnancies). Also, their knowledge of how to prevent contracted STIs -through protected sex-could have increased their lack of fear in have more sexual partners. Only the Igbos were less likely to have had their first sexual experience at younger ages, which further emphasis the extent to which ethnicity account for variation in sexual behaviour (Imaledo et al., 2012) owing to some inherent and contextual ethnic factors that are beyond the scope of the study.
Furthermore, it was discovered that the economic characteristics of individuals play a vital role in determining risky sexual behaviour among young people. However, socio-economic variables do not predict three indicators of risky sexual behaviour similarly. For instance, while having either primary or at least secondary education yields a less likelihood of having unprotected sex (non-use of condom in the first and last sexual intercourse) which is a positive sexual behaviour (Odimegwu et al., 2019), it is associated with increasing propensity to have high lifetime sexual partners and having sex at young in the case of being exposed to nonformal education. Similarly, the richer the young people are, the less likely they are to use protection but more likely to have multiple sexual partners in their lifetime. Also, the odds of engaging in unprotected sex and having first sex at younger ages were higher among the non-working group making non-participation in economic activities a prominent risk factor for risky sexual practices among young people. Being aware of AIDs didn't translate to reduced indulgence in unprotected sex but brought about a reduction in the risk of having more sexual partners. This suggests a tendency to indulge in unprotected sex, which makes less than two sexual partners as ideal and logical.
As earlier stated, the relationship between psychological factors and risky sexual behaviour among sexually active young people in Nigeria is not one that is overly replete in public health literature in Nigeria. Hence, this study established that self-reporting low level of satisfaction with life is associated with high lifetime multiple sexual partners among young people in Nigeria. This suggests that the influence of psychological factors may differ by the type of risky sexual behaviour just like what is obtainable in a study conducted in America were self-esteem among adolescents significantly determine their exposure to risky sexual behaviour (Kerpelman, McElwain, Pittman, & Adler-Baeder, 2016). Nevertheless, this doesn't eschew the fact that psychological factors are indispensable predictor of risky sexual behaviour as seen in Australia and Poland, where psychological factors also have a significant impact on the life of individuals apart from their sexual behaviour. Specifically, in the study in Poland, lower self-esteem was a predictor of unprotected sexual intercourse. In Australia, individuals experiencing psychological distress possess an increasing tendency to adopt risky sexual behaviours (Ganczak et al., 2017;Nguyen et al., 2017), while perceived self-esteem among adolescents is found to be a significant correlate with their practice of risky sexual behaviour. Furthermore, one study among Mexican adolescents has found that self -efficacy is linkable sexual risk behaviour of young people while this scenario also plays out in Iran where it was found that personality pattern is a significant predictor of risky sexual behaviours (Palacios, 2018;Samadypoor & Kord Tamini, 2016).

Conclusion
The fact that the link between psychological factors and risky sexual behaviour among young people in Nigeria has not received the much-needed attention makes this study an important one in developing interventions that address the high penchant for more sexual partners (among the young people) by targeting their psychology in terms of their subjective perceptions about their lives and well-being. Also, this study identified the significant socio-demographic, economic factors associated with the different indicators of risky sexual behaviour among young people, which will further inform evidence-based programmatic interventions. This study has provided empirical information which-by implication-would further help in tackling the spread of Sexually Transmitted Infections and other fatal outcomes of risky sexual practices in the country. This would accelerate steps towards achieving good health and well-being in Nigeria by the year 2015 in accordance with the Sustainable Development Goal (SDG) three.

Recommendation
Apart from considering socio-demographic and economic factors among young people in Nigeria that exposes them to risky sexual practices, the study recommends that psychological risk factors should be also be given utmost priority in sexual health programmes (through social and behavioural change communication approach) to effectively tackle the menace of risky sexual practices and its negative consequences which could jeopardize all efforts to achieve good health and well-being (i.e., Sustainable Development Goal Three) in Nigeria.