Mongolian Adolescents’ Sexual Behavior and Beliefs in Light of Their Health Risk Behavior, Social Support, and Well-Being

Abstract

Adolescents’ health risk behavior may have a longstanding impact on their later adult life and health. The aim of this cross-sectional study is to report on health risk behaviors and sexual attitudes in a sample of Mongolian adolescents (N = 312) in light of their levels of social support and well-being. Measurements were derived from the Health Behavior in School-aged Children (HBSC), including substance use and sexual behavior, whereas social support from parents and peers was measured with the Multidimensional Scale of Perceived Social Support (MSPSS). Student well-being was evaluated using the WHO Well-Being Questionnaire (WBI-5). The frequency of current smokers was 13.1% (boys: 18.9%, girls: 10.2%; χ² = 4.61, Phi = −0.12, p = 0.032). The prevalence rate of current alcohol consumption was 11.2% (boys: 14.2%, girls: 9.7%; χ² = 1.38, Phi = −0.07, p > 0.05), similar to the nationwide results. The percentage of students who have ever had sexual intercourse was 6.4 (12.4% of boys and 3.4% of girls, χ² = 9.33, Phi = 0.17, p = 0.006), similar to Central Asian countries included in the HBSC survey. Boys were more likely to associate sexual activity with self-esteem, intimacy with a partner, and positive perceptions among peers. Conversely, girls expressed greater concern about the negative consequences of sexual activity, such as STDs and pregnancy. Social support from the family played a protective role in frequencies of smoking and drinking. However, levels of the social support subscales did not differ by the students’ relationship status or previous sexual intercourse. These results suggest that there is a need for a complex prevention program to avoid increases in Mongolian adolescents’ smoking and drinking and promote their healthy sexual attitudes and behavior.

1. Introduction

Preventing adolescents’ health risk behavior and promoting positive health outcomes and well-being is a worldwide priority in public health; however, we know less about these issues in several regions, e.g., the countries of Central Asia, such as Mongolia. Therefore, in this exploratory study, we focus on Mongolian adolescents’ sexual behavior and attitudes and their health risk behaviors (smoking and drinking). We aim to test a pertinent question about whether studies of adolescent risk-taking behaviors can be generalized to Mongolia.

Adolescence is the transitional stage between childhood and adulthood, spanning the ages of 10 to 19 years, as defined by the World Health Organization (WHO) [1]. During this stage, young people experience social, emotional, and cognitive changes, beyond the physical ones [2]. Physically, adolescents face significant changes, including the development of secondary sexual characteristics, growth spurts, and changes in the brain structures and functions. Emotionally, adolescents experience intensive expression of emotions, engage in more risk-taking behaviors, and seek greater independence from their families [3]. Puberty and brain development lead to new behaviors and capacities that facilitate transitions in social domains [4]. Preventing adolescents’ health risk behaviors and promoting their positive sexual health in particular is a great challenge for both parents and health educators. Adolescent sexual and reproductive health needs have intensified in recent years [5]. Adolescents should receive age-appropriate education about their sexual health to help them develop a safe and positive view of sexuality [6]. In line with such a challenge, health education programs should take into account that adolescent problem behavior often includes interplays between various forms, such as substance use and early or risky sexual activities [7,8].

The World Health Organization (WHO) defines sexual health as a condition of physical, emotional, mental, and social well-being concerning sexuality rather than just the absence of disease, malfunction, or infirmity [9]. Adolescents’ involvement in risky sexual behavior is a pressing issue for public health on a global scale [10,11]. Some common risky sexual behaviors among adolescents include, e.g., having early and/or unprotected sex, substance use during sex, lacking knowledge about sexual health, or having sex with older partners [12,13]. These risky behaviors increase the likelihood of getting HIV/AIDS or other sexually transmitted infections (STIs) and unintended pregnancy [14,15]. However, teaching adolescents about the anatomy and physiology of biological sex and reproduction is only one aspect of sex education. It should also address healthy sexual development, body image, gender identity, sexual development, interpersonal interactions, and intimacy [16,17]. Positive aspects of sexuality should also be mentioned as a part of the normative developmental process, as sexual well-being may contribute to youths’ psychological well-being [18]. Namely, they need to develop not only sexual literacy but also positive attitudes and skills to take responsibility for their sexual life [19].

Among health risk behaviors, adolescent substance use, particularly smoking and drinking, is a public health priority and a prevailing challenge for health educators [20]. Previous studies indicated that substance use could contribute to adolescents’ decisions regarding their sexual behavior [8,21]; thus, comprehensive health education programs are required to prevent this. Sexually active youth are more likely to spend their time with sexually active peers who also share more adult-like activities, such as substance use [22]. However, in a systematic review of cohort studies, it was found that the association between engagement in early sexual intercourse and substance use was dependent on familial factors, such as family attachment or the family’s living condition [23].

Numerous elements, such as the familial environment, pressure from peers, media exposure, cultural and religious views, and access to sexual health information and services, may affect adolescents’ sexual health behavior and substance use. Cultural and traditional norms, beliefs, and practices shape not only adolescents’ harmful habits but also their attitudes toward sexuality and sexual behavior [17,24]. Gender norms also vary widely across cultures, and they can impact adolescents’ learning about and engaging in sexual behaviors. For instance, in some cultures, boys are expected to be sexually active and may face pressure to have multiple partners, while girls are expected to be more passive and may face stigma and shame for engaging in sexual activity [25,26]. Further, many religions have specific teachings and expectations about sexual behavior, such as abstinence until marriage or prohibitions on sexual acts, such as in Islamic cultures [27]. These teachings may be reinforced by religious leaders and communities, shaping adolescents’ beliefs and behaviors related to sexual health.

In the literature, social relationships are often shown to serve as protective factors against health risk behaviors; in particular, social support from the family can provide protection against adolescent smoking and drinking [28]. Besides parental control, the quality of the relationship between the adolescents and their adult family members and significant others is the most relevant protective factor. In one study on sexual behavior, adolescents were more likely to delay sexual behavior, use contraception when they become sexually active, and have fewer pregnancies if they perceive their parents as warm, caring, interested, and responsive. Moreover, this study also found that communication about sex between parents and children increased young people’s intentions to use condoms [29]. Social support from friends seems less supportive in avoidance of health risk behaviors [30]. In a study among youth in Kolkata (India), peer pressure was found to significantly elevate the risk of initiation of smoking and drinking [31]. However, peers may also play an important role in adolescent identity formation, social skill development, and subjective well-being [32].

While monitoring trends in adolescent substance use and sexual behavior has been well documented in European countries, only a few countries from Asia have been included, for example, in the HBSC (Health Behavior in School-aged Children) studies [33]. There are sporadic studies reporting on Mongolian youths’ health and problem behavior; however, these are not comparable with international data due to the lack of validated questionnaires. The birth rate among adolescents in Mongolia is higher than average in Asia and the Pacific. The unmet need for family planning was reported in 2016 as the highest among females aged 15–24 [34]. The percentage of using condoms every time they have sexual intercourse was 57.4 in this age group [35]. A recent study found that Mongolian adolescents and youth have a high level of alcohol consumption; in addition, their awareness of alcohol-related harms is not sufficient. The lifetime prevalence of alcohol use was more than 50% among high school students, and they begin drinking alcoholic beverages at the age of 16 [36]. Another survey in Mongolia (Global School-Based Student Health Survey, GSHS–2023) indicated that 38.4% of adolescents aged 13–17 years reported ever smoking, and 11.1% reported currently smoking [37]. Although in most Asian countries, smoking and drinking are less frequent compared to the European rates, smoking and in particular heavy and risky drinking were found to be a common public health problem in the region, including in South Korea, China, Thailand, Mongolia, Vietnam, and Japan [38], where there are not sufficient data in some countries, e.g., in Mongolia.

There is a need for knowledge of Mongolian culture to interpret research data. In terms of culture, we should take into account that a Mongolian family’s culture and upbringing methods are very different from those of European families. Historically, in the typical Mongolian family, with its unique image of citizenship based on kinship ties, family education has been provided to the next generation in accordance with traditional upbringing customs and legal norms, and they have been trained in practical life skills. Mongolian parents pay a lot of attention to their children, love and care for them, and provide economic opportunities, but they are at a loss when it comes to choosing communication styles with their children and upbringing principles, and discussing sex education and sexuality with their children is always a closed and sensitive topic [39]. Discussing sexual health is still largely considered as a taboo. This cultural reluctance creates significant barriers for adolescents, parents, and teachers, leading to embarrassment and a lack of knowledge about sexual matters. As a result, misinformation persists, leaving young people ill equipped to make informed decisions regarding their sexual behavior. Traditional values often dictate that topics related to sex and sexuality are inappropriate for public or familial discussion, effectively marginalizing the issue. The double standard for gender roles also persists, and this perception hinders safer sex practices and reinforces gendered expectations regarding responsibility for contraception, with females facing the brunt of societal judgment. All these issues reflect the patriarchal norms ingrained in Mongolian society, which has a significant impact on adolescent sexual behavior [40]. In Mongolia, religion has been dominated by Mongolian Buddhism and Mongolian shamanism (the Mongolian ethnic religion). Mongolian religious norms are rooted in the worship of patrilineal ancestors, emphasizing masculine values and the subordinate role of women, which has only partially been changed by the Soviet-style revolutionary regime. On the other hand, it also decreased the role of religion in everyday life and children’s socialization [41]. Shamanic practices are more prevalent in rural areas, where the nomadic lifestyle aligns closely with shamanism’s deep connection to nature. In urban centers like Ulaanbaatar, shamanism coexists with modern lifestyles but might be less visibly practiced. In urban areas, the majority of adolescents may see Buddhism and shamanism as more symbolic or view them as part of their heritage rather than an active practice. Culture may also have an impact on substance use and attitudes toward smoking and drinking. Alcohol, in the form of a traditionally made ‘fermented horse milk liquor’, is often consumed, particularly by men [42]. Likewise, smoking is also a serious public health challenge, with higher rates among males, urban dwellers, and youth [43].

As mentioned above, Mongolian adolescents’ health risk behaviors and sexual life is a heavily under-investigated field of research, particularly with internationally accepted measurements. So, there is a need to collect data about adolescents’ behavioral status and problems to support adolescents’ reproductive health and prevent them from taking risky behaviors, and evaluate them in light of their cultural background. In this study, we aimed to examine (a) health risk behavior (smoking and drinking), (b) sexual behavior and beliefs, (c) social support, and (d) well-being in a sample of Mongolian (urban) adolescents. Our decision is based on the recognition that there are significant lifestyle and cultural differences between the urban and rural populations of Mongolia. Urban students may have greater exposure to sexual health education and media, while rural students may be more influenced by traditional values and family beliefs. In addition, Mongolia is well known for its nomadic traditions. Therefore, provinces are located in different geographical regions, which can cause difficulties in reaching them. Besides the prevalence rates of these behaviors, we also aimed to detect differences in social support, well-being, and sexual beliefs by gender, substance use, status of their relationship, and actual sexual behavior. Hopefully, this exploratory study triggers future research among this adolescent population, about whom we know little.

Beyond detecting frequencies of sexual behaviors and attitudes and health risk behaviors, the following hypotheses were formulated:

H1: 

We assume that gender would make a difference in the adolescents’ sexual behaviors and attitudes, as well as in their health risk behavior.

H2: 

We hypothesize that those who avoid substance use would report higher levels of social support (especially familial support).

H3: 

We also assume that sexual and relationship status would be related to higher frequencies of students’ smoking and drinking.

H4: 

Sexually active youth would report more positive sexual beliefs.

2. Materials and Methods

2.1. Participants and Procedure

Cluster sampling was carried out, and the primary sampling unit was the school class. We utilized cluster sampling to select schools based on district representation, school size, and their willingness to participate. Within each selected school (altogether 4), classes were randomly chosen (altogether 18). Inclusion criteria were the following: (1) the principal of the school must have given permission and (2) students must have had informed consent from their parents/guardians. We planned to include approximately 300–400 students from Mongolian secondary schools, ranging in age from 15 to 18 and in grades from 9 to 12, as participants in this survey. Our sample size (N = 312) proved acceptable with 95% CI and a 4.99% margin of error, based on the sampling frame of 1639 students. Thus, the selected age groups could represent the onset of adolescence (when young people face the challenges of physical and emotional changes) and the middle years, when they start to consider meaningful life and career decisions [36].

As city and rural living conditions in Mongolia are very distinct, we decided to collect data only for 15–18-year-old Mongolian adolescents in secondary schools in Ulaanbaatar and its Provinces. Research questions were translated into Mongolian languages with the help of professionals and prepared as paper-based questionnaires. Forms were fully self-completed after participants were given proper guidelines on how to complete them. It took around 20–25 min to finish the questionnaire. The questionnaire form-filling activities were held in the school class, where special attention was given to ensuring the comfort and privacy of participants during the survey.

We conducted the survey anonymously to reduce the potential for peer pressure or authority influence. We took great care in preparing the survey to assure students that their responses would remain confidential. For instance, when collecting paper-based survey data, we provided sealable envelopes for each student, which contained identical colored pencils and questionnaires. Students sealed their envelopes after completing the survey. Additionally, we arranged for students to be seated at a distance from one another to further protect their privacy. The distribution and collection of the questionnaires were conducted without any involvement from school officials or teachers, ensuring a neutral environment. Moreover, we provided participants with instructions that emphasized the integrity of the survey process and the voluntary nature of their participation. In order to minimize survey fatigue, besides keeping a logical flow in survey questions, we arranged to obtain feedback from a small group of students in the pre-test period. No one declined to participate in the survey (due to the class setting and the normative behavior of the students in a collectivistic culture); however, 12 students were absent, leading to a final sample size of 312 students (with a response rate of 96.3%).

Recognizing the cultural sensitivity surrounding sexual health topics in Mongolia, we implemented several culturally appropriate strategies to address students’ concerns. Before conducting the full study, we gathered feedback from both students and teachers and made necessary modifications based on their input. For example, after discussions with school principals and experts, we made some sexual health- and behavior-related questions optional and removed a few entirely. In order to promote clarity and transparency, we also organized two briefing sessions at each school prior to data collection. During these sessions, we explained the study’s purpose and addressed any questions or concerns from participants.

2.2. Ethical Considerations

This research received ethical clearance from the Institutional Review Board (IRB) of the Doctoral School of Education, University of Szeged, on 14 October 2023 (reference number: 18/2023). Furthermore, ethical approval was also received from the ethical committees of the Mongolian National University before conducting this research in Mongolia. Participation in this study was voluntary and anonymous. Initially, we contacted the principals of the selected schools, requesting permission to include students in this study. In addition, with the help of school staff, including the principal, teacher, and social workers, we also contacted the parents or legal guardians of the students in advance. Subsequently, we asked the students and their legal guardians or parents for written consent to participate in this study. Finally, we met the students in person to introduce our study and explain how it can contribute to their health education. We emphasize that the data collection would be anonymous and no personal information would be used in publications.

2.3. Measurements

We applied the methodology of the Health Behavior in School-aged Children (HBSC) study [44]. The Health Behavior in School-aged Children (HBSC) study was among the first international surveys on adolescent health. HBSC employs a standard methodology detailed in the international study protocol. Since this survey was the first in Mongolia to include these measurements, the items were translated and back-translated by bilingual translators. In addition, extensive efforts were undertaken to ascertain cultural and linguistic validity. Additionally, a professional English–Mongolian translator reviewed and refined the questionnaire for linguistic accuracy, while a Mongolian-language teacher ensured grammatical and contextual correctness. Data were collected through self-completion of questionnaires administered in the classroom. The questions cover a range of health indicators, health-related behaviors, life circumstances, and risk behaviors. The questions gather information on demographic factors, social context, health outcomes, health behaviors, risk behaviors, and student well-being.

To determine the prevalence of sexual intercourse, age at first sexual intercourse, and other items of sexual behavior, the Health Behavior in School-aged Children (HBSC) study questions were used [36]. In addition, questions from the Youth Risk Behavior Survey (YRBS: Centers for Disease Control and Prevention) were included to gather information about other aspects of sexual behavior, including number of partners, types of partners, and alcohol/drug use during intercourse in the last 3 months [45]. In terms of the questions on sexual activity, we gave a clear description of sexual intercourse and made sure that it was consensual and beneficial for the students (other types could not be included in this pilot phase of the project). Similarly, when we asked about the students’ relationship status (whether being in a relationship, i.e., having a girl- or boyfriend, or being single), a clear description was given which was unambiguous for adolescents at this age.

Sexual behavioral beliefs were assessed using the Integrative Model (IM) Belief Measures [46]. These beliefs can be evaluated on 7-point scales ranging from 1 (indicating extremely unlikely) to 7 (indicating extremely likely). Each assessment is preceded by the stem: “If I were to engage in sexual intercourse within the next 12 months, it would…”. In total, there are 12 items that have been grouped into 4 categories (with 3 items each) for the purpose of analysis. These categories encompass beliefs related to different aspects: beliefs about oneself (e.g., “Make me feel as though someone had taken advantage of me”); beliefs about relationships with partners (e.g., “Increase the quality of relationship with my partner”); beliefs about relationships with others, e.g., friends (e.g., “Gain the respect of my friends”); and beliefs about physical outcomes (e.g., “Give me a STD, HIV/pregnancy”, reverse item). Cronbach alphas for the subscales were as follows: self (α = 0.77); partners (α = 0.87); others (α = 0.84); and physical (α = 0.98).

Student well-being was measured using the 5-item version of the WHO Well-Being Questionnaire (WBI-5) [47]. The questionnaire includes five statements (e.g., “I have felt calm and relaxed”) about the students’ feelings during the past two weeks. Responses were measured on a 4-point Likert-type scale from 0 to 3, evaluating the summarized scores, where greater level of well-being was based on higher scores. The reliability coefficient (Cronbach alpha) was α = 0.93 with this sample.

The Multidimensional Scale of Perceived Social Support (MSPSS) measured the students’ social support [48]. The scale assesses social support in three dimensions: friend support (3 items, e.g., “My friends really try to help me”), family support (4 items, e.g., “I get emotional help and support I need from my family”), and significant other support (3 items, e.g., “There is a special person with whom I can share my joys and sorrows”). The students were asked to indicate how strongly they agree with each statement of the 12-item questionnaire. Responses were evaluated on a 5-point Likert-type scale from “strongly disagree” to “strongly agree”. Higher scores reflect higher levels of social support. In our study, the subscales were reliable with the following Cronbach alphas: family support (α = 0.93); friend support (α = 0.92); and significant other support (α = 0.91).

2.4. Data Analysis

IBM SPSS Statistics for Windows, Version 25.0 (Armonk, NY: IBM Corp, 2017) was used in the calculations, and the maximum significance level was set to 5%. Student t-tests (Welch tests in the case of non-equal SDs) were used to detect group differences in the scores for study scales. Chi-square tests were applied to test significance for differences in frequencies of smoking and drinking.

3. Results

3.1. Descriptive Statistics for Sample Characteristics

The sample included 312 participants, out of which 106 (34.0%) were boys and 206 (66.0%) were girls. The age distribution of the sample was as follows: 14.7% at the age of 15, 18.6% at the age of 16, 33.3% at the age of 17, and 33.3% at the age of 18. The distribution based on grade was as follows: 9th grade—14.7%, 10th grade—17.6%, 11th grade—34.3%, and 12th grade—33.3%. Fifty percent of the student had no religion, while Buddhism and Mongolian shamanism were the most commonly mentioned forms of religion. In total, 33% of participants stated they were in a relationship (boys: 43.4%, girls: 27.7%, χ² = 7.83, Phi = 0.16, p = 0.007), and 6.4% of them had ever had sexual intercourse (boys: 12.4%, girls: 3.4%, χ² = 9.33, Phi = 0.17, p = 0.006) (

Table 1. Descriptive statistics for characteristics of the study sample (N = 312).

Variable		N (%)	Age
			Age 15 n = 46 (14.7)	Age 16 n = 58 (18.6)	Age 17 n = 104 (33.3)	Age 18 n = 104 (33.3)
Gender
	Boys	106 (34.0)	17 (5.4)	24 (7.7)	30 (9.6)	35 (11.2)
	Girls	206 (66.0)	29 (9.3)	34 (10.9)	74 (23.7)	69 (22.1)
Family structure (with whom do you live in your home?)
	Mother	262 (84.0)	41 (89.1)	4 7(81.0)	94 (90.4)	80 (76.9)
	Father	216 (69.2)	30 (65.2)	44 (75.9)	78 (75.0)	64 (61.5)
	Stepmother	5 (1.6)	0	0	2 (1.9)	3 (2.9)
	Stepfather	7 (2.2)	1 (2.2)	0	3 (2.9)	3 (2.9)
	Other family members (siblings, grandparents)	99 (31.7)	24 (52.2)	14 (24.1)	30 (28.8)	31 (29.8)
	In a foster home	4 (1.3)	0	1 (1.7)	1 (1.0)	2 (1.9)
Religion
	Christian	10 (3.2)	0	5 (8.6)	1 (1.0)	4 (3.9)
	Buddhist	71 (22.8)	6 (13.0)	6 (10.3)	36 (34.6)	22 (21.4)
	Muslim	1 (0.3)	0	0	1 (1.0)	0
	Mongolian shamanism	51 (16.3)	10 (21.7)	14 (24.1)	18 (17.3)	9 (8.7)
	Other religions	23 (7.4)	5 (10.9)	4 (6.9)	7 (6.7)	7 (6.8)
	No religion	156 (50.0)	25 (54.3)	29 (50.0)	41 (39.4)	61 (59.2)
Relationship status
	Single	209 (67.0)	30 (65.2)	39 (67.2)	6 7(64.4)	73 (70.2)
	In a relationship	103 (33.0)	16 (34.8)	19 (32.8)	37 (35.6)	31 (29.8)
Sexual activity status
	Intercourse	20 (6.4)	1 (2.2)	0(1)	7 (6.7)	12 (11.5)
	No intercourse	291 (93.6)	45 (97.8)	57 (100)	104 (93.3)	104 (88.5)

). Variables of risky sexual activities were not appraisable due to non-responsiveness to these items.

3.2. Sexual Behavior and Beliefs

Table 2. Descriptive statistics for sexual behavioral belief variables by gender and relationship status (N = 312).

Variables	Boys	Girls	t-Value 95% CI (LL–UL) (Cohen’s d)	In a Relationship	Single	t-Value 95% CI (LL–UL) (Cohen’s d)
	Mean ± SD	Mean ± SD		Mean ± SD	Mean ± SD
Sexual Beliefs
Beliefs about oneself	11.98 ± 3.93	10.22 ± 3.55	3.99 (p < 0.001) (0.89–2.62) (0.47)	11.95 ± 4.11	10.26 + 3.46	3.80 (p < 0.001) (0.81–2.45) (0.44)
Beliefs about relationships with partners	11.54 ± 3.56	10.67 ± 3.37	2.10 (p = 0.036) (0.05–1.67) (0.25)	11.98 ± 3.72	10.47 ± 3.21	3.71 (p < 0.001) (0.71–2.31) (0.41)
Beliefs about relationships with others, e.g., friends	13.17 ± 3.21	11.74 ± 4.07	3.14 (p = 0.002) (0.53–2.32) (0.63)	13.01 ± 3.65	11.84 ± 3.89	2.54 (p = 0.012) (0.26–2.07) (0.31)
Beliefs about physical outcomes	8.18 ± 7.02	9.89 ± 7.53	−1.94 (p = 0.048) (−3.44–0.23) (0.23)	8.68 ± 7.63	9.64 ± 7.28	−1.15 (p = 0.251) (−2.78–0.73) (0.13)

Note: Student t-test.

shows results for sexual behavioral belief variables by gender and relationship status. There was a significant difference between genders in terms of beliefs about self; that is, more boys reported they would have self-respect due to having a sexual relationship in the future (t[310] = 3.99, p < 0.001, Cohen’s d = 0.47). Also, more boys than girls thought that a sexual relationship in the future would increase their intimacy with their partners (t[310] = 2.10, p = 0.036, Cohen’s d = 0.25). Likewise, more boys than girls thought that a sexual relationship in the future would increase others’ (e.g., friends’) positive opinion of them (t[310] = 3.14, p = 0.002, Cohen’s d = 0.63). There was also a significant difference in beliefs about negative physical consequences of a possible future sexual relationship (e.g., pregnancy, STDs): more girls agreed with these statements (t[310] = −1.94, p = 0.048, Cohen’s d = 0.23).

There were no significant differences between the students in their beliefs about physical health consequences by their relationship status (p > 0.05). However, there were significant differences in their beliefs about self-respect (t[310] = 3.80, p < 0.001, Cohen’s d = 0.44), beliefs about partners’ positive feelings (t[310] = 3.71, p < 0.001, Cohen’s d = 0.41), and beliefs about others’ positive opinion (t[310] = 2.54, p = 0.012, Cohen’s d = 0.31). Those who were in a relationship tended to have stronger positive beliefs about a possible future sexual relationship compared to single students.

Table 3. Descriptive statistics for sexual behavioral belief variables by sexual activity status (N = 312).

Variables	Sexual Intercourse	No Sexual Intercourse	t-Value 95% CI (LL–UL) (Cohen’s d)
	Mean ± SD	Mean ± SD
Sexual Beliefs
Beliefs about oneself	15.80 ± 3.38	10.48 ± 3.56	6.48 (p < 0.001) (3.70–6.93) (1.53)
Beliefs about relationships with partners	14.65 ± 3.41	10.72 ± 3.32	5.11(p < 0.001) (2.42–5.44) (1.17)
Beliefs about relationships with others, e.g., friends	10.50 ± 2.39	12.33 ± 3.90	−2.07 (p = 0.039) (−3.58–0.09) (0.57)
Beliefs about physical outcomes	15.45 ± 7.80	8.91 ± 7.20	3.91(p < 0.001) (3.25–9.83) (0.81)

Note: Student t-test.

shows results for sexual behavioral belief variables by sexual activity status (whether they have ever had sexual intercourse before or not). Not surprisingly, those who reported having sexual intercourse before would expect to have more positive beliefs about oneself (t[309] = 6.48, p < 0.001, Cohen’s d = 1.53) and more positive feelings from their partners (t[309] = 5.11, p < 0.001, Cohen’s d = 1.17). However, they would also expect more negative physical consequences (t[309] = 3.91, p < 0.001, Cohen’s d = 0.81). Finally, in terms of others’ opinion, fewer students would expect others’ positive opinion related to a future sexual relationship (t[309] = −2.07, p = 0.039, Cohen’s d = 0.57).

3.3. Sexual Behavior, Socal Support, and Well-Being

There were no gender differences in the levels of social support, while girls scored higher on the well-being scale than boys (t[310] = −3.49, p = 0.001, Cohen’s d = 0.41) (

Table 4. Descriptive statistics for variables of social support and well-being by gender and relationship status (N = 312).

Variables	Boys	Girls	t-Value 95% CI (LL–UL) (Cohen’s d)	In a Relationship	Single	t-Value 95% CI (LL–UL) (Cohen’s d)
	Mean ± SD	Mean ± SD		Mean ± SD	Mean ± SD
Social Support (MSPSS)
Family	13.92 ± 5.02	14.58 ± 4.60	−1.16 (p = 0.248) (−1.77–0.46) (0.14)	13.74 ± 5.11	14.67 ± 4.54	−1.62 (p = 0.105) (−2.05–0.20) (0.19)
Friends	13.67 ± 4.91	14.08 ± 4.59	−0.73 (p = 0.468) (−1.51–0.70) (0.09)	13.60 ± 5.22	14.10 ± 4.42	−0.89 (p = 0.374) (−1.62–0.61) (0.10)
Others	13.17 ± 4.89	13.41 ± 4.91	−0.41 (p = 0.685) (−1.39–0.91) (0.05)	13.97 ± 3.65	13.00 ± 4.69	1.63 (p = 0.103) (−0.20–2.11) (0.23)
Well-being	13.05 ± 7.21	15.96 ± 6.85	−3.49 (p = 0.001) (−4.55–1.27) (0.41)	13.92 ± 7.37	15.49 ± 6.93	−1.84 (p = 0.067) (−3.24–0.11) (0.22)

Note: Student t-test.

). Gender differences were not significant in terms of relationship status (p > 0.05) (Table 4).

According to

Table 5. Descriptive statistics for variables of social support and well-being by sexual activity status (N = 312).

Variables	Sexual Intercourse	No Sexual Intercourse	t-Value 95% CI (LL–UL) (Cohen’s d)
	Mean ± SD	Mean ± SD
Social Support (MSPSS)
Family	14.55 ± 4.28	14.34 ± 4.80	0.19 (p = 0.849) (−1.96–2.38) (0.05)
Friends	14.30 ± 4.30	13.91 ± 4.73	0.36 (p = 0.718) (−1.75–2.53) (0.09)
Others	14.00 ± 4.05	13.27 ± 4.95	0.64 (p = 0.521) (−1.50–2.96) (0.09)
Well-being	18.15 ± 8.22	14.73 ± 7.00	2.09 (p = 0.038) (0.20–6.63) (0.48)

Note: Student t-test.

, gender differences were not significant for the social support scales in light of previous sexual intercourse (p > 0.05). However, levels of well-being were higher for those who reported having sexual intercourse before (t[309] = 2.09, p = 0.038, Cohen’s d = 0.48). We also investigated religiosity, but neither having a boy- or girlfriend or having sexual intercourse was associated with it (p > 0.05).

3.4. Frequencies of Smoking and Drinking

According to the collected data, 29.8% of students had smoked at some point in their lives, while 13.1% had smoked in the past three months. There were differences noted between boys and girls in the prevalence of smoking. The statistical analysis showed a relationship between being a boy or girl and the lifetime prevalence of smoking (χ²[1, N = 312] = 16.20, p < 0.001, Phi = −0.23) and three-month prevalence of smoking (χ²[1, N = 312] = 4.61, p = 0.035, Phi = −0.12). Also, 24.7% of students had drunk alcohol in their lifetime, and 11.2% had drunk alcohol in the last three months. However, there were no significant relationships between sex and prevalence of smoking (p > 0.05) (

Table 6. Descriptive statistics for the frequencies of smoking and drinking (whole sample and by sex, n = 312).

Variables	Frequency n (%)	Boys n (%)	Girls n (%)	Chi-Square Tests for the Relationship of Sex and Substance Use
Lifetime prevalence of smoking				χ2 = 16.20 (p < 0.001) Phi = −0.23 **
No	219 (70.2)	59 (55.7)	160 (77.7)
Yes	93 (29.8)	47 (44.3)	46 (22.3)
Three-month prevalence of smoking				χ2 = 4.61 (p = 0.035) Phi = −0.12 *
No	271 (86.9)	86 (81.1)	185 (89.9)
Yes	41 (13.1)	20 (18.9)	21 (10.2)
Lifetime prevalence of drinking				χ2 = 2.62 (p = 0.127) Phi = −0.09
No	235 (75.5)	74 (69.8)	161 (78.2)
Yes	77 (24.7)	32 (30.2)	45 (21.8)
Three-month prevalence of drinking				χ2 = 1.38 (p = 0.259) Phi = −0.07
No	277 (88.8)	91 (85.8)	186 (90.3)
Yes	35 (11.2)	15 (14.2)	20 (9.7)

Note: Chi-square test. * p < 0.05; ** p < 0.001.

).

3.5. Smoking, Drinking, Socal Support, and Well-Being

Table 7. Descriptive statistics for variables of social support and well-being by current substance user status (N = 312).

Variables	Nonsmoker	Current Smoker	t-Value 95% CI (LL–UL) (Cohen’s d)	No Alcohol Use	Current Drinker	t-Value 95% CI (LL–UL) (Cohen’s d)
	Mean ± SD	Mean ± SD		Mean ± SD	Mean ± SD
Social Support (MSPSS)
Family	14.58 ± 4.73	12.88 ± 4.72	2.15 (p = 0.032) (0.15–3.26) (2.69)	14.64 ± 4.70	12.11 + 4.64	3.00 (p = 0.003) (0.87–4.18) (2.56)
Friends	13.95 ± 4.71	13.85 ± 4.67	0.12 (p = 0.901) (−1.45–1.65) (2.92)	14.06 ± 4.65	12.94 ± 4.96	1.33 (p = 0.183) (−0.53–2.78) (2.66)
Others	13.40 ± 4.93	12.85 ± 4.67	0.66 (p = 0.508) (−1.07–2.16) (2.65)	13.44 ± 4.95	12.46 ± 4.40	1.12 (p = 0.265) (−0.75–2.71) (2.63)
Well-being	14.38 ± 6.90	18.90 ± 7.28	−3.89 (p < 0.001) (−6.82–−2.23) (2.64)	14.77 ± 6.84	16.57 ± 8.86	−1.42 (p = 0.158) (−4.31–0.70) (2.07)

Note: Student t-test.

shows the results for mean scores of the social support and well-being scales by current substance user status (i.e., three-month prevalence). Among the social support variables, nonsmoking students reported higher levels of family support (t[310] = 2.15, p = 0.032, Cohen’s d = 2.69). However, smokers reported higher levels of well-being compared to nonsmokers (t[310] = −3.89, p < 0.001, Cohen’s d = 2.64). In terms of alcohol use, higher levels of family support were reported by those who had not drunk alcohol during the past several months (t[310] = −3.00, p = 0.003, Cohen’s d = 2.56).

3.6. Sexual Behavior and Beliefs, in Light of Smoking and Drinking

Sexual beliefs showed several differences between smokers and nonsmokers. Smokers expected higher levels of beliefs about self-respect in the case of a sexual relationship (t[310] = −2.25, p = 0.025, Cohen’s d = 3.38) as well as beliefs about partners’ positive feelings (t[310] = −2.01, p = 0.045, Cohen’s d = 3.43). Those who had consumed alcohol in the past several months stated similar feelings; in addition, they also expected more positive opinions from others (t[310] = −2.54, p = 0.012, Cohen’s d = 2.56) (

Table 8. Descriptive statistics for sexual behavioral belief variables by current substance user status (N = 312).

Variables	Nonsmoker	Current Smoker	t-Value 95% CI (LL–UL) (Cohen’s d)	No Alcohol Use	Current Drinker	t-Value 95% CI (LL–UL) (Cohen’s d)
	Mean ± SD	Mean ± SD		Mean ± SD	Mean ± SD
Sexual Beliefs
Beliefs about oneself	10.63 ± 3.82	12.05 ± 3.22	−2.25 (p = 0.025) (−2.65–−0.18) (3.38)	10.64 ± 3.71	12.29 ± 3.96	−2.36 (p = 0.019) (−2.91–−0.26) (3.17)
Beliefs about relationships with partners	10.81 ± 3.43	11.96 ± 3.48	−2.01 (p = 0.045) (−2.29–−0.03) (3.43)	11. 80 ± 3.31	12.34 ± 4.24	−2.52 (p = 0.012) (−2.76–−0.34) (3.21)
Beliefs about relationships with others, e.g., friends	12.22 ± 3.92	12.29 ± 3.39	−0.12 (p = 0.908) (−1.35–1.20) (3.32)	12.03 ± 3.87	13.77 ± 3.31	−2.54 (p = 0.012) (−3.09–−0.39) (3.79)
Beliefs about physical outcomes	9.08 ± 7.29	10.80 ± 8.02	−1.39 (p = 0.165) (−4.16–0.71) (2.08)	9.42 ± 7.45	8.43 ± 6.95	0.75 (p = 0.456) (−1.62–3.60) (1.16)

Note: Student t-test.

).

Finally, there was an association between relationship status and smoking (both in terms of lifetime and monthly prevalence) as well as ever having had sexual intercourse or not and smoking. In the case of alcohol, the relationships were significant only with having had earlier sexual intercourse (

Table 9. Descriptive statistics for the frequencies of smoking and drinking by sexual and relationship status (n = 312).

Variables	In a Relationship n (%)	Single n (%)	Chi-Square Tests	Sexual Intercourse n (%)	No Sexual Intercourse n (%)	Chi-Square Tests
Lifetime prevalence of smoking			χ2 = 14.16 (p < 0.001) Phi = −0.21 ***			χ2 = 9.49 (p = 0.004) Phi = −0.18 **
No	58 (56.3)	161 (77.0)		8 (40.0)	211 (72.5)
Yes	45 (43.7)	48 (23.0)		12 (60.0)	80 (27.5)
Three-month prevalence of smoking			χ2 = 11.37 (p = 0.001) Phi = −0.19 **			χ2 = 19.70 (p < 0.001) Phi = −0.25 ***
No	80 (77.7)	181 (91.4)		11 (55.0)	260 (89.3)
Yes	23 (22.3)	18 (8.6)		9 (45.0)	31 (10.7)
Lifetime prevalence of drinking			χ2 = 2.43 (p = 0.126) Phi = −0.09			χ2 = 4.70 (p = 0.034) Phi = −0.12 *
No	72 (69.9)	163 (78.0)		11 (55.0)	223 (76.6)
Yes	31 (30.1)	46 (22.0)		31 (45.0)	46 (23.4)
Three-month prevalence of drinking			χ2 = 2.87 (p = 0.126) Phi = −0.10			χ2 = 7.52 (p = 0.016) Phi = −0.16 **
No	87 (84.5)	190 (90.9)		14 (70.0)	262 (90.0)
Yes	16 (15.5)	19 (9.1)		6 (30.0)	29 (10.0)

Note: Chi-square test. * p < 0.05 ** p < 0.01 *** p < 0.001.

).

4. Discussion

Although adolescents make up a relatively small portion (17.5 percent) of Mongolia’s population, the lack of information and research about them results in their exclusion from policymaking processes and contributes to the continuation of traditional inequalities [49]. Exploring adolescents’ health risk behaviors is extraordinarily important as they may have a longstanding impact on their later adult life and health. Since there is a lack of studies about these issues, this study aimed to examine health risk behavior, sexual activity, and beliefs in a sample of Mongolian adolescents in relation to their well-being and social support.

In terms of smoking, according to the 2021/2022 HBSC survey including Europe, Central Asia, and Canada, 24% of boys and 26% of girls had ever smoked at age 15, where the highest rate among girls was reported in Denmark (64%, Greenland) and the lowest frequencies among girls (around 1%) in Tajikistan, Armenia, and Kyrgyzstan [50]. The Youth’s Health Behavior among Secondary School Children—Global School-Based Student Health Survey (GSHS–2023) reported Mongolian 13–17-year-olds’ health behaviors in the framework of a nationwide study [37]. Overall, 38.4% of students had ever tried smoking (48.7 among boys and 28.9% among girls). The monthly prevalence was 11.1% (boys: 17.2%, girls: 5.3%); in urban areas, it was 16.6% among boys and 5.5% among girls. Our results for the frequency of current smokers (13.1%; 18.9% among boys and 10.2% among girls) are similar to the national data, although the girls’ frequency proved slightly higher. However, the lifetime prevalence seemed lower in our study (29.8%). While among European countries, gender differences in smoking seem to disappear [33], Mongolian boys’ smoking remains more frequent compared to girls. This finding can be explained by the prevailing double standard for gender roles in traditional societies [40]. While historically, smoking has been viewed as a male habit, the prevalence of female smoking is increasing in low- and middle-income countries [50]. This trend is the same in Mongolia, especially in urban settings [43].

In terms of alcohol use, the results of the 2021/2022 HBSC survey reported a great variation in alcohol consumption among the countries included, with the highest rates in Denmark (84.4% of girls and 84.3% of boys) and Hungary (78% of girls and 75% of boys), while Tajikistan, Kazakhstan, and Kyrgyzstan showed the lowest rates (between 1 and 16%) [33]. In a previous Mongolian study, 10.8% was the rate of current alcohol consumption (boys: 11.2%, girls: 10.5%), while in our study it was 11.2% (boys: 14.2%, girls: 9.7%), similar to the nationwide results [36]. Gender differences were not found in any cases. These findings are in line with previous research: in a systematic review, while heavier drinking seems to be related to masculinity on an ideological basis, the prevalence rates do not follow this expectation [51].

The frequencies of 15-year-olds who have ever had sexual intercourse were definitely greater in European countries (e.g., Greenland: 49% of girls vs. 30% of boys; Hungary: 19% of girls vs. 28% of boys) [52] compared to our results (6.4%; 12.4% of boys vs. 3.4% of girls). The Global School-Based Student Health Survey (GSHS–2023) reported higher rates: 18.9% in the whole sample, 21.7% among Mongolian boys, and 16.4% among Mongolian girls [37]. However, the Central Asian countries included in the HBSC survey reported similarly low frequencies, especially among girls [52]. For example, in Kazakhstan, the frequencies were as follows: 11% of boys vs. 2% of girls. In Kyrgyzstan, they were 5% of boys vs. 1% of girls.

In addition to sexual behavior, we also explored their behavioral, normative, and self-efficacy beliefs with respect to engaging in sexual intercourse in the future [46]. A previous study found that adolescents who reported having had sexual intercourse also had more positive sexual self-concepts and they felt better in engaging in sexual communication with their partners compared to those without earlier experience of sexual intercourse [53]. Our findings are in line with these results. However, regarding these sexual beliefs, significant differences were observed based on gender, relationship status, and previous sexual intercourse. Boys were more likely to associate sexual activity with self-esteem, intimacy with a partner, and positive perceptions among peers. Conversely, girls expressed greater concern about the negative consequences of sexual activity, such as pregnancy and sexually transmitted infections (STIs). These results are consistent with previous research demonstrating that cultural and gender norms heavily influence adolescents’ perceptions and behaviors related to sexuality [25,26,27,54,55]. In Mongolia, cultural norms such as traditional family structures and gender expectations (e.g., the prevailing double standard for gender roles) also persist, and this perception hinders safer sex practices and reinforces gendered expectations that uniquely shape adolescent health behaviors [39,40]. The association of boys’ sexual activity with social validation and self-esteem may reflect societal pressure to conform to the traditional masculine ideals that prioritize sexual conquests [40]. Meanwhile, girls’ concerns about the potential negative consequences of sexual activity may arise from societal expectations emphasizing female modesty and responsibility [55]. These societal expectations significantly impact how adolescents view self-esteem in relation to sexual activity. A further explanation for the gender variations in sexual self-concept and societal expectation could be the differences in peer socialization or exposure to media. We should also note here that the experienced students scored higher on the statement saying that if they had sexual intercourse in the next 12 months, it would cause them negative physical consequences (e.g., STDs or pregnancy), showing that they were aware of these risks. Gender differences in sexual health knowledge (e.g., about STDs) is also well recognized, with females being more educated in this respect [56]. This is not surprising, given the more serious health impacts of unsafe sexual activities for women (e.g., infertility).

Our results suggest that adolescent sexual behavior and beliefs vary in light of their substance use. Single adolescents reported a lower rate of smoking compared to those in a relationship. Moreover, youth who reported earlier sexual intercourse consumed significantly more alcohol and cigarettes. In terms of sexual beliefs, smokers and those who had drunk alcohol during the past several months also expected a positive change in their sexual self-concept in the case of a future sexual relationship. Previous studies described a close connection between substance use and early or risky sexual behavior [7,8,21]. Sexual debut may also be directly associated with substance use, both symbolizing adulthood for young people [57]. Adolescent problem behaviors often show a high co-occurrence, and so does the avoidance of substance use and having responsible sexual activity [58]. A potential contributor to this association could be the adolescents’ perceived social support and the influence of their social relationships. Besides peer influence, adolescents’ elevated risk-taking may also explain these findings [57].

Social support may serve as a strong protective factor against health risk behavior; in particular, parental involvement provides protection in reducing adolescent substance use [28] and risky sexual behavior [29]. Notably, adolescents who did not smoke or consume alcohol reported higher levels of family support in our study. Parental support may include monitoring adolescent behavior, but also discussing health issues and guidance as well as providing a positive role model. However, the picture is different in terms of sexual behavior. First, no gender differences were detected in levels of the social support subscales. These levels did not differ by relationship status or previous sexual intercourse either. These findings may be explained by the cultural environment in Mongolia, which is characterized by traditional family structures and sensitivities surrounding sexual relationships, presenting both unique challenges and opportunities for adolescent health education [39]. Our results suggest that Mongolian parents are not supportive in discussing sexual education with their children. Since 2018, Mongolia has integrated lessons on reproductive health and comprehensive sexual education into the general education curriculum [37]; however, supportive familial relationships would be even more effective in helping adolescents to develop healthy attitudes and well-being. Traditionally, there is an inverse relationship between adolescents’ sexual behaviors and their religiousness; however, our findings report no significant relationship between these variables. An Italian study found that that the influence of religiosity is strong only among those with strong religious identity commitment, not among uncertain individuals [59]. As mentioned earlier, the majority of Mongolian urban adolescents do not actively practice their religion.

Finally, in terms of sexual activity, studies usually concentrate on risky behavior, but sexual health can also be viewed as a positive concept which is relevant for health and well-being among adolescents [60,61]. Our finding supports this concept, as those with previous experience of sexual intercourse scored higher on the well-being scale. Unfortunately, the well-being scores were higher among smokers. This finding may be explained by the biochemical nature of nicotine, which may serve as self-medication against stress and negative mood states [62]. A potential alternative explanation is that smokers in adolescence may experience positive evaluation from other smoking friends. Moreover, a study found a relationship between self-efficacy and smoking status [62]. However, while nicotine may have short-term mood-elevating or social effects, the long-term consequences of smoking should be acknowledged.

Interpreting our results in the light of our previous hypotheses, boys had higher levels of experience with sexuality, more positive beliefs about future sexual relationships (except for the physical consequences), and more substance use (H1). The second hypothesis (H2) has been partially confirmed, as only family support showed higher levels among those who reported avoidance of smoking and drinking, while the other forms of social support did not. Further, frequencies of smoking and drinking were indeed related to the students’ sexual activity and relationship status (having a boy- or girlfriend); that is, the third hypothesis (H3) is supported. Finally, as H4 suggested, sexually active youth indeed reported more positive sexual beliefs about the future.

Overall, our results highlight some important information about (urban) Mongolian adolescents’ sexual beliefs and behaviors, substance use, social support, and well-being. While many studies are carried out on European and American adolescents, we know much less about Mongolian youth, who have not yet been included in international surveys. Therefore, this study adds to the literature on this special population. Some limitations also should be mentioned. First, our study is basically descriptive and exploratory, and due to its pilot nature, more sophisticated analyses were not feasible at this stage. For example, the use of bivariate analyses may limit its analytical value; future research should use multivariate analyses to account for the impact of confounding variables, such as parental socioeconomic status. Multivariate analyses, such as logistic regression analysis or cluster analysis, would help better identify associations among variables of health risk behaviors, sexual beliefs, social support, and socio-demographics. Investigating several sensitive questions is always a challenge, and thus some of the questions remained unanswered (e.g., sexual orientation). Anonymity allowed the students to reply with confidence; however, nonresponse may always happen due to the sensitive topic. Therefore, we removed those questions which had too many missing data from the analysis (e.g., about contraception). We should also acknowledge the potential impact of absent students. Although a careful procedure in translation and adaptation was followed, several items and scales need further validation. Further, excluding non-consensual experiences may unintentionally overlook an essential aspect of adolescent sexual health. The gender imbalance (66% girls, 34% boys) may influence the generalizability of the findings. Furthermore, in this study, we focused only on urban adolescents as they were easier to reach online due to comprehensive internet coverage; a more careful organization of the survey is needed to include schools in rural areas. The urban–rural disparities in adolescent health would give more insight into adolescent behaviors in different geographical places, providing further explanations about the prevailing roles of traditional factors or the changing cultural milieu. Finally, given this study’s cross-sectional nature, it is essential to emphasize that these associations do not imply causation. In addition, in future research, we should also add a final question about how the students felt when answering these sensitive issues. Besides the applied social support scales, additional measures of social support would also be useful to capture objective support.

In conclusion, our study highlights the need for culturally appropriate programs that enhance knowledge and communication skills among both parents and adolescents. We believe that these findings may contribute to a complex prevention program in reducing health risk behaviors and promoting healthy sexual attitudes among Mongolian adolescents.