Prevalence and correlates of risky sexual behaviour among in-school adolescents in Bulawayo, Zimbabwe: a cross-sectional study

Abstract

Background

Adolescents are a key target for sexual health interventions because risky sexual behaviour (RSB) raises public health concerns, including sexually transmitted infections and unintended pregnancies. In Bulawayo, Zimbabwe, rising HIV rates among adolescents highlight the need for targeted research. This study aims to assess the prevalence of RSB among in-school adolescents and identify associated sociodemographic and sexual health educational factors.

Methods

An analytical cross-sectional survey was conducted among 314 adolescents aged 13–19 from four randomly selected high schools in Bulawayo Central. A self-administered, structured questionnaire was used to collect data on sexual behaviours, knowledge and risk factors. Data were analysed using descriptive statistics, χ² tests and multivariable logistic regression to identify significant predictors of sexual activity. Prevalence estimates are reported with 95% CIs.

Results

The analysis revealed that 52 of 314 respondents (16.6%, 95% CI 12.5% to 21.4%) reported ever having engaged in sexual intercourse. A higher prevalence was observed among males (28/122, 23.0%) over females (24/192, 12.5%). Adolescents in Forms 5–6 were more likely to engage in sexual activity (adjusted OR (AOR) 3.46, 95% CI 1.24 to 9.63, p=0.018), and those living with a single parent were also at increased risk (AOR 2.56, 95% CI 1.13 to 5.81, p=0.024). Among the 52 sexually active respondents, 38 (73.1%) reported unprotected sex at last intercourse, with higher rates among females (22/26, 84.6%) compared with males (16/26, 61.5%) (p=0.049). Although 268 participants (85.4%) reported receiving sexual health education, only 149 (47.5%) found it sufficient and 296 (94.3%) desired safe spaces for discussion.

Conclusion

There is a significant gap between sexual health knowledge and safe practices, necessitating interventions that build skills, address communication gaps and foster supportive school environments to reduce RSB.

WHAT IS ALREADY KNOWN ON THIS TOPIC

• Risky sexual behaviour among adolescents poses a significant global public health issue, linked to HIV, sexually transmitted infections and unintended pregnancies. Research in Sub-Saharan Africa has identified age, gender and family structure as factors influencing sexual activity. In Zimbabwe, studies indicate persistent high-risk behaviours among youth, yet there is a scarcity of recent, context-specific data focusing on in-school adolescents in urban areas like Bulawayo.

WHAT THIS STUDY ADDS

• This study provides recent quantitative evidence from Bulawayo, revealing a critical disconnect between high sexual theoretical knowledge (85.4% received sexual health education) and unsafe practices, with 73.1% of sexually active adolescents engaging in unprotected sex, particularly females (84.6%). It identifies key correlates, such as older school year levels (adjusted OR (AOR) 3.46) and single-parent households (AOR 2.56). An overwhelming demand (94.3%), n=296/314, for confidential school-based safe spaces highlights the inadequacy of current educational delivery.

HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY

• The findings suggest a shift from knowledge-based curricula to comprehensive sexuality education that fosters negotiation and critical thinking skills. Policymakers and schools should prioritise creating non-judgmental safe spaces for students. The results also call for gender-transformative interventions to address power dynamics impacting condom use, particularly for adolescent girls.

Introduction

Adolescence, defined globally as the period between the ages of 10 and 19, represents a critical phase for health and development.¹ In the Zimbabwean context, this largely corresponds to the secondary school-going population (Forms 1–6), who are a vital target for sexual and reproductive health (SRH) interventions. Globally, the prevalence of risky sexual behaviours (RSBs) among adolescents remains substantial. Recent data from 69 low- and middle-income countries (LMICs) indicate that approximately 6.9% of adolescents aged 12–15 years have ever had sexual intercourse, with higher rates among boys (10.0%) compared with girls.² In the USA, nearly half of high school students report having had sexual intercourse, and 40% do not use condoms consistently during sex.³ A range of factors, including age, gender and socioeconomic status, influences these behaviours.⁴ RSB among adolescents, encompassing early sexual debut, unprotected sex and multiple partners, remains a major global public health concern due to its strong association with adverse outcomes like sexually transmitted infections, including HIV, unintended pregnancies and psychosocial challenges.⁵

Across different regions, several demographic factors are consistently associated with RSB among adolescents: older adolescents are more likely to engage in risky sexual practices compared with their younger counterparts.⁶ Males often report higher prevalence of RSBs, but females are particularly vulnerable to negative outcomes such as sexual violence and exploitation. Adolescents not attending school are at higher risk, as school environments can offer protective structures and information.⁵ Poverty and low household wealth are linked to higher rates of RSB, often due to transactional sex or lack of access to sexual health resources.¹ Strong parental involvement and emotional support are protective, reducing the likelihood of engaging in RSBs.⁷ Association with deviant peers and exposure to sexually explicit materials increases the risk of engaging in unsafe sexual practices.²

In Sub-Saharan Africa (SSA), the burden is particularly high. In South Africa, studies reveal that up to 50% of adolescents engage in sexual intercourse before the age of 16, and as many as 60% of these adolescents reported not using condoms during sexual activity.⁷ Multiple sexual partnerships are also common, with about 30% of adolescents with early sexual debut not in monogamous relationships.⁷ Socioeconomic challenges, peer influence and family dynamics are significant contributors to these behaviours. Research in Nigeria highlights a high prevalence of RSBs, including multiple sexual partnering, transactional sex, age-disparate sex and non-consensual sex.⁵ In Kenya, risky sexual experiences among adolescents are characterised by high rates of sexual violence, intergenerational sex and transactional sex.⁸

In Zimbabwe, adolescents and young people continue to bear a disproportionate burden of HIV and poor SRH outcomes.⁹ The Zimbabwe Demographic and Health Survey reported that among women aged 20–24, 21% had begun childbearing before age 18, pointing to early sexual activity and potential unprotected sex.¹⁰ Studies in Zimbabwe have identified factors such as poverty, gender inequality, peer pressure and limited access to youth-friendly SRH services as key drivers of RSB.^{11 12} However, much of the existing research is either dated, national-level or focused on out-of-school youth. There is a paucity of recent, school-based quantitative studies specifically examining the prevalence and multifaceted correlates (eg, family structure, quality of sexual health education) of RSB among adolescents in Bulawayo, Zimbabwe’s second-largest city.

Given the potential for adverse health outcomes and the long-term impact on both individual well-being and community health, this study sought to address this gap. The specific research questions were: (1) What is the prevalence of RSB among in-school adolescents in Bulawayo? (2) Which sociodemographic and sexual health educational factors are associated with RSB in this population? Consequently, this study aimed: (1) to determine the prevalence of RSB among in-school adolescents in Bulawayo and (2) to identify the sociodemographic and sexual health educational factors associated with such behaviour. This study provides critical evidence to inform targeted interventions and strengthen adolescent SRH programmes in the region.

Materials and methods

Study setting

This study was conducted in Bulawayo, Zimbabwe’s second-largest city and a major urban centre in the Matabeleland region. Bulawayo Central District was purposively selected for its diverse range of public and private secondary schools, providing a representative urban educational setting.

Study design and population

An analytical cross-sectional study was conducted in Bulawayo, Zimbabwe’s second-largest city. The study population comprised adolescents aged 13–19 (the typical age range for Zimbabwean secondary school students) from four public secondary schools in Bulawayo. This study targeted all adolescents who are still in high schools for quantitative data collection (survey). Official Ministry of Education records for Bulawayo Central District for the 2024 academic year estimated the target population to be 12 450 students. The inclusion criteria were all adolescents aged 13–19 enrolled in the selected schools and willing to participate and provide informed consent.

Sample size and sampling technique

The sample size was calculated using the Cochran formula for a finite population. Using a population size of 12 450, a 95% confidence level, a 5% margin of error and an expected prevalence of 50% to maximise variability, a minimum sample size of 373 was required. This was rounded up to 400 to account for a 15% non-response rate. However, due to administrative constraints and lower-than-expected recruitment, a final sample of 314 was achieved, providing a margin of error of approximately 5.5%.

A multistage sampling technique was employed. First, from a list of all public secondary schools in Bulawayo Central, four schools (two single-sex male, two single-sex female) were selected using simple random sampling. The selection of four schools was based on feasibility, administrative permissions and the need to include a mix of single-sex schools (two male, two female) to ensure gender balance in the sample frame. Second, within each school, classes (Forms 1–6) were selected as clusters to ensure representation across all grade levels. Finally, within each selected class, a simple random sampling method was used to select the predetermined number of students proportional to the school size.

Data collection tools and procedure

A structured, self-administered questionnaire was developed by the researchers based on a review of existing literature and questions adapted from validated instruments, such as the Youth Risk Behaviour Surveillance System and similar adolescent health surveys conducted in SSA.¹³ The tool was developed and administered in English, the primary language of instruction in Zimbabwean secondary schools and structured into three sections.

Section 1 collected sociodemographic data (gender, age, form level, living arrangement) using closed-ended, categorical response options. Section 2 assessed sexual behaviour; the primary outcome, ‘Ever had sexual intercourse’, was measured with a dichotomous ‘Yes’/‘No’ response. Participants who responded ‘Yes’ were directed to follow-up questions, including age at first intercourse (open-ended numerical response), number of sexual partners (categorical: ‘1’, ‘2’, ‘3 or more’) and condom use at last sex (dichotomous: ‘Yes’/‘No’). Section 3 evaluated awareness and perceptions of sexual health education, using ‘Yes’/‘No’ response options for questions regarding receipt of education, its perceived sufficiency and the need for school-based safe spaces. The full data collection tool is attached as an online supplemental file 1.

For validation, the questionnaire was pretested with 30 adolescents from a non-participating school to assess content validity (relevance and comprehensiveness) and face validity (clarity and understandability). Feedback from the pretest was used to refine question wording. Internal consistency for key multi-item scales (such as perceptions of education) was assessed using Cronbach’s alpha, which was found to be above the acceptable threshold of 0.7. The final tool was administered using Google Forms. The researcher partnered with school administrators to gain permission, and a brief information session was conducted to explain the study’s purpose, confidentiality and the voluntary nature of participation.

Data analysis

Data were extracted from Google Forms to Microsoft Excel and imported into SPSS V.28 for analysis. Analyses included: (1) descriptive statistics (frequencies, proportions with 95% CIs) to summarise characteristics and prevalence; (2) bivariate analyses using the χ² test to assess associations between categorical independent variables (gender, age, form level, living arrangement) and the primary outcome (ever had sex); (3) multivariable logistic regression to identify independent predictors of ever having had sexual intercourse, including all variables from the bivariate analysis regardless of significance. Adjusted ORs (AORs) with 95% CIs were reported. For the secondary outcome (unprotected sex among the sexually active), similar bivariate and multivariable analyses were performed. A p value of <0.05 was considered statistically significant.

Ethical considerations

Permission was also sought from the Bulawayo Provincial Education Director and respective school heads. Written informed assent was obtained from all participating adolescents, and written informed consent was obtained from their parents or guardians. Participants were informed of their right to withdraw at any time without consequence. To ensure anonymity, no personally identifying information was collected. Confidentiality was maintained, and data were stored on a password-protected server. These measures were explicitly communicated to participants to minimise social desirability bias and encourage honest responses.

Patient and public involvement statement

The research was designed and conducted by an academic researcher as part of a Master’s degree project with logistical and time constraints. However, the research questions were informed by existing public health literature and regional priorities concerning adolescent health. Results will be disseminated to participating schools and the provincial education director via a summary report.

Results

Demographic information of participants

The final sample consisted of 314 adolescents. There were 192 (61.1%) female respondents and 122 (38.8%) male respondents, all between the ages of 13 and 19. The majority of respondents (188, 59.9%) lived with both parents. The sociodemographic characteristics are summarised in table 1.

Table 1. Sociodemographic characteristics of respondents (N=314).

Characteristic	Category	Frequency (n)	Percentage (%)
Gender	Female	192	61.1
	Male	122	38.9
Age group	13–15 years	146	46.5
	16–17 years	88	28.0
	18–19 years	80	25.5
Form level	Form 1–2	84	26.8
	Form 3–4	132	42.0
	Form 5–6	98	31.2
Living arrangement	Both parents	188	59.9
	Single parent	62	19.7
	Guardian	26	8.3
	Alone	6	1.9
	Other	32	10.2

Prevalence of risky sexual behaviour

The prevalence of ever having had sexual intercourse, the primary indicator of RSB, was 16.6% (52/314; 95% CI 12.5% to 21.4%). The prevalence was not uniform across subgroups. A significant gender disparity was evident: 21.4% of males (28/122) were sexually active compared with 13.5% of females (24/192). This represents a 58% higher crude prevalence among males. Furthermore, prevalence increased markedly with school progression: from 7.0% (5/84) in Forms 1–2 to 15.5% (28/132) in Forms 3–4, and 30.6% (19/98) in Forms 5–6. Adolescents living with a single parent reported the highest prevalence among family structures at 25.0% (18/62), compared with 12.9% (21/162) for those with both parents.

Factors associated with sexual activity

To address the second objective of identifying associated factors, both bivariate and adjusted analyses were conducted. Table 2 presents the detailed results. In bivariate analysis, all four sociodemographic factors showed significant associations with sexual activity: gender (p=0.041), age group (p=0.026), form level (p=0.012) and living arrangement (p=0.011). The multivariable logistic regression model, which controlled for potential confounding between these factors, provided a clearer picture of independent associations. Two factors emerged as statistically significant independent predictors of sexual activity.

Table 2. Factors associated with ever having had sexual intercourse (N=314).

Variable	Category	Ever had sex n (%)	Never had sex n (%)	Χ² p value	Unadjusted OR (95% CI)	Adjusted OR (95% CI)
Gender	Male	28 (21.4)	103 (78.6)	0.041	2.01 (1.02 to 3.96)	1.88 (0.93 to 3.80)
	Female	24 (13.5)	159 (86.5)		1.00 (Ref)	1.00 (Ref)
Age group (years)	13–15	11 (10.4)	95 (89.6)	0.026	1.00 (Ref)	1.00 (Ref)
	16–17	25 (19.4)	104 (80.6)		2.05 (0.93 to 4.52)	1.70 (0.71 to 4.09)
	18–19	16 (27.1)	43 (72.9)		3.18 (1.29 to 7.82)	2.11 (0.74 to 6.04)
Form level	Form 1–2	5 (7.0)	66 (93.0)	0.012	1.00 (Ref)	1.00 (Ref)
	Form 3–4	28 (15.5)	153 (84.5)		2.43 (0.91 to 6.47)	2.12 (0.78 to 5.79)
	Form 5–6	19 (30.6)	43 (69.4)		5.75 (1.95 to 16.96)	3.46 (1.24 to 9.63)*
Living arrangement	With both parents	21 (12.9)	142 (87.1)	0.011	1.00 (Ref)	1.00 (Ref)
	Single parent	18 (25.0)	54 (75.0)		2.24 (1.07 to 4.69)	2.56 (1.13 to 5.81)*
	Guardian	11 (18.0)	50 (82.0)		1.48 (0.64 to 3.42)	1.40 (0.58 to 3.37)
	Alone	1 (14.3)	6 (85.7)		1.12 (0.13 to 9.94)	1.02 (0.11 to 9.61)
	Other	1 (12.5)	7 (87.5)		0.96 (0.12 to 7.91)	0.89 (0.10 to 7.85)

Ref denotes Reference group

^*Represent statistically significant factors

CI, Confidence Interval ; OR, Odds Ratio.

1. School progression: Adolescents in Forms 5–6 had 3.46 times higher odds (AOR 3.46; 95% CI 1.24 to 9.63; p=0.018) of being sexually active compared with those in Forms 1–2, even after accounting for age, gender and family structure.

2. Family structure: Adolescents living with a single parent had 2.56 times higher odds (AOR 2.56; 95% CI 1.13 to 5.81; p=0.024) of being sexually active compared with those living with both parents.

While the male gender showed elevated odds in the bivariate model (OR 2.01), this association attenuated and lost statistical significance in the adjusted model (AOR 1.88; 95% CI 0.93 to 3.80; p=0.078), suggesting its effect may be mediated through other variables like age or form level. Similarly, the age group association was no longer significant after adjustment, indicating that school progression (form level) may be a more precise measure of developmental stage in this school-based context. The findings are presented in table 2

Sexual behaviour among the sexually active (n=52)

To contextualise the prevalence findings, a detailed analysis of the sexually active subgroup was conducted. Among the 52 sexually active adolescents, the mean age at sexual initiation was 15.8 years (±1.6 SD). The largest proportion (10/52, 19.2%) initiated sex at age 16. The most significant reasons for engaging in sexual intercourse were peer pressure (24/52, 45.5%) and curiosity (11/52, 21.8%). The fact that more than a quarter of teenagers (14/52, 26.9%) reported having several sexual partners emphasises the higher risks of sexually transmitted diseases in this demographic. The findings are presented in table 3.

Table 3. Number of sexual partners by demographic characteristics (n=52).

Characteristic	Category	One partner n (%)	Two partners n (%)	Three+partners n (%)	Total (n)
Gender	Female	18 (75.0)	4 (16.7)	2 (8.3)	24
	Male	20 (71.4)	6 (21.4)	2 (7.1)	28
Age group	13–15	9 (81.8)	2 (18.2)	0 (0.0)	11
	16–17	19 (70.4)	6 (22.2)	2 (7.4)	27
	18–19	10 (71.4)	2 (14.3)	2 (14.3)	14
Form level	Form 1–2	4 (80.0)	1 (20.0)	0 (0.0)	5
	Form 3–4	21 (75.0)	5 (17.9)	2 (7.1)	28
	Form 5–6	13 (68.4)	4 (21.1)	2 (10.5)	19
Living arrangement	With both parents	16 (76.2)	4 (19.0)	1 (4.8)	21
	Single parent	12 (66.7)	4 (22.2)	2 (11.1)	18
	Guardian	8 (72.7)	2 (18.2)	1 (9.1)	11
	Alone	1 (100.0)	0 (0.0)	0 (0.0)	1
	Other	1 (100.0)	0 (0.0)	0 (0.0)	1
Overall total		38 (73.1)	10 (19.2)	4 (7.7)	52

Prevalence and factors associated with unprotected sex

Among the 52 sexually active adolescents, a high proportion (38, 73.1%) reported engaging in unprotected sex at their last sexual encounter. Bivariate analysis indicated that females were significantly more likely to report unprotected sex than males (22/26, 84.6% vs 16/26, 61.5%, p=0.049). The unadjusted OR was 3.44 (95% CI 1.01 to 11.75). However, in the multivariable model adjusting for age, form level and living arrangement, the association was attenuated and no longer statistically significant (AOR 3.12, 95% CI 0.81 to 12.01). No other factors were significantly associated with unprotected sex (table 4).

Table 4. Factors associated with unprotected sex among sexually active adolescents (n=52).

Variable	Category	Unprotected sex n (%)	Protected sex n (%)	Χ² p value	Unadjusted OR (95% CI)	Adjusted OR (95% CI)
Gender	Male	16 (61.5)	10 (38.5)	0.049	1.00	1.00
	Female	22 (84.6)	4 (15.4)		3.44 (1.01 to 11.75)	3.12 (0.81 to 12.01)
Age group (years)	13–15	8 (72.7)	3 (27.3)	0.871	1.00	1.00
	16–17	21 (77.8)	6 (22.2)		1.31 (0.29 to 5.96)	1.55 (0.30 to 8.03)
	18–19	9 (64.3)	5 (35.7)		0.68 (0.12 to 3.73)	0.89 (0.14 to 5.75)
Form level	Form 1–2	3 (60.0)	2 (40.0)	0.518	1.00	1.00
	Form 3–4	22 (78.6)	6 (21.4)		2.44 (0.39 to 15.11)	2.67 (0.41 to 17.29)
	Form 5–6	13 (68.4)	6 (31.6)		1.44 (0.21 to 9.77)	1.21 (0.16 to 9.32)
Living arrangement	With both parents	15 (71.4)	6 (28.6)	0.913	1.00	1.00
	Single parent	14 (77.8)	4 (22.2)		1.40 (0.38 to 5.14)	1.31 (0.33 to 5.21)
	Guardian	8 (72.7)	3 (27.3)		1.07 (0.24 to 4.77)	1.01 (0.21 to 4.88)
	Alone	1 (100.0)	0 (0.0)		Undefined	Undefined
	Other	1 (50.0)	1 (50.0)		0.40 (0.02 to 8.03)	0.45 (0.02 to 10.51)

CI, Confidence Interval; OR, Odds Ratio.

Reasons for engaging in unprotected sex

Among the 38 who engaged in unprotected sex, the primary reasons for unprotected sex were ‘having a trusted partner’ (8, 20.8%) and ‘not thinking about using protection’ (8, 20.8%). Other reasons included the partner refusing (6, 16.7%), lack of availability (5, 12.5%) and being under the influence of substances (3, 8.3%). This is reported in figure 1.

Figure 1. Reasons for engaging in unprotected sex.

Availability of sexual health education

The vast majority of respondents (85.4%) reported having received sexual health education. However, only 149 (47.5 %) believed the education provided was sufficient. A bivariate analysis was conducted to examine associations between the perceived sufficiency of sexual health education and sexual activity. Adolescents who reported that the sexual health education they received was not sufficient had higher odds of being sexually active, although this association did not reach statistical significance in this sample (OR 1.45, 95% CI 0.81 to 2.60, p=0.210). An overwhelming majority (296, 94.3%) felt their school needed a dedicated, confidential safe space for students to discuss sexual health issues. Furthermore, 231 participants (73.6%) reported that topics like condom negotiation and communication skills were not adequately covered in their current education.

Discussion

This study provides critical insights into the sexual behaviour of in-school adolescents in Bulawayo, Zimbabwe. The key findings reveal a troubling disconnect between high levels of sexual health knowledge and the practice of unsafe sex, particularly among younger adolescents and those in seemingly stable family environments. The overall prevalence of sexual intercourse (16.6%) among our sample is consistent with rates reported in other studies from LMICs.² Notably, this figure, while lower than rates in some Western contexts,³ underlines that a significant minority of in-school adolescents are sexually active and thus a key population for sexual health interventions. The observed gender disparity, with higher rates among males (21.4%) than females (13.5%), aligns with well-documented global trends often attributed to societal norms that encourage male sexual adventurism.² However, the fact that this association lost statistical significance in the adjusted multivariable model (AOR 1.88, 95% CI 0.93 to 3.80) suggests that in this specific context, the effect of gender may be mediated or confounded by other sociodemographic factors such as age and family structure.

The alarmingly high rate of unprotected sex (73.1%) among sexually active participants represents the study’s most pressing public health finding. This exceeds rates reported in a study of in-school adolescents across five sub-Saharan countries found that 43.7% reported ever having sexual intercourse,¹⁴ indicating a particularly severe knowledge–practice gap in our setting. The gendered pattern with females reporting higher, though not statistically significant, adjusted odds resonates deeply with a study from Bulawayo that documents how gendered power dynamics impede girls’ condom negotiation.¹¹ The primary reasons for unprotected sex (‘trusted partner’, ‘didn’t think about it’) reveal not ignorance of risks but rather cognitive and relational barriers to protection, similar to findings from adolescent studies in Liberia and South Africa.^{15 16} This suggests that factual knowledge alone is insufficient; interventions must equip adolescents with the socio-emotional skills to navigate complex intimate relationships, accurately perceive their own vulnerability and translate knowledge into consistent protective behaviour.^{17 18}

The study findings reported a notable disparity in gender factors associated with unprotected sex. In the bivariate analysis, females were significantly more likely to report unprotected sex than males (84.6% vs 61.5%, p=0.049); the magnitude of the point estimate and the high prevalence among females constitute a major public health concern. This trend suggests that gendered power dynamics in relationships may impair girls’ ability to insist on condom use, a challenge well-documented in the literature.^{3 19} Crucially, no other demographic factor was a significant predictor of unprotected sex. This indicates that the propensity for unprotected sex is a pervasive issue cutting across all demographic subgroups in this sexually active population, rather than being confined to a specific at-risk profile.

The study findings on independent correlates for sexual initiation provide important insights for targeting. Identification of advanced school progression (Forms 5–6) as a strong independent predictor (AOR 3.46) aligns with developmental theories and regional studies from Zambia and South Africa, where each additional school year increases exposure to peer networks, romantic opportunities and autonomy.^{20 21} More novel is the finding regarding single-parent households (AOR 2.56), which revealed that ‘family dysfunction’ is a risk factor by pinpointing a specific vulnerable structure. This does not imply single parenting causes RSB, but rather that these households may face challenges in providing consistent supervision or open parent–child communication about sexuality, a protective factor well-established in local studies.

Furthermore, data on sexual health education reveal a critical implementation gap. While the vast majority of respondents (85.4%) reported having received formal sexual health education, nearly half (47.5%) found it insufficient. The non-significant trend suggesting higher odds of sexual activity among those deeming education insufficient points to a potential quality gap. This discrepancy echoes evaluations of Zimbabwe’s Family Life Education curriculum, which noted that didactic teaching methods and teacher discomfort inhibited effectiveness.²² The overwhelming demand (94.3%) for safe spaces provides direct student input for intervention design, aligning with successful peer-educator models across different countries that created trusted forums for SRH discussion.²³ The fact that theoretical awareness does not translate into safe practices reinforces the urgent need to shift from a didactic, knowledge-based curriculum to an evidence-based, comprehensive sexuality education (CSE) model that builds socio-emotional competencies, critical thinking and self-efficacy.^{24 25}

The findings have several key implications for policy change and practice. Policymakers should mandate and resource the implementation of skills-based CSE within the national school curriculum, moving beyond biological facts to include critical thinking, communication and negotiation skills for safer sex. Programme implementers should develop targeted interventions for higher-risk groups identified in this study, specifically adolescents in upper forms (5–6) and those from single-parent households, focusing on strengthening protective family communication and support. Given the gendered patterns in sexual activity and potential vulnerability in condom use, gender-transformative programmes are needed. These should include specific interventions engaging adolescent males to promote responsible sexual behaviour and positive masculinity, and programmes to empower female adolescents with negotiation skills and agency to insist on safer sex. Community programmes must equip parents with skills and confidence to engage in open, positive and ongoing conversations about sexuality with their children to reduce the prevalence of high-risk behaviours in this population.

Limitations

The small sample size for the analysis of factors associated with unprotected sex (n=52) limited the statistical power to detect significant associations and resulted in imprecise effect estimates, as reflected in the wide CIs. The gender imbalance in the sample, while reflective of the school populations, may affect the generalisability of the findings. The focus on in-school adolescents means the findings are not representative of out-of-school youth, who may be at even higher risk. Potential biases include social desirability bias (under-reporting of stigmatised behaviours) and recall bias. To mitigate these, the study employed anonymous self-administered questionnaires, assured confidentiality and used clear, simple questions.

Conclusion and recommendations

In conclusion, this study identifies a critical knowledge-practice gap among adolescents in Bulawayo, characterised by high-risk sexual behaviours despite widespread access to sexual health education. The findings call for a paradigm shift in intervention strategies. This study recommends

• Policymakers to integrate CSE into the national school curriculum, with a strong focus on practical life skills, negotiation and critical thinking.

• Schools are to establish the ‘safe spaces’ students are demanding, facilitated by trained counsellors or peer educators, to foster open and non-judgmental discussions about sexual health.

• Programme implementers to develop and deploy targeted interventions for the most vulnerable, including younger adolescents (13–15 years), focusing on risk perception and condom use skills.

• Development of programmes that equip parents with the skills and comfort level to engage in open, positive and ongoing conversations about sexuality with their children.

• Design and implementation of gender-specific programmes: initiatives targeting adolescent males to promote responsible behaviour, and empowerment programmes for adolescent females to strengthen negotiation skills for safer sex.

Data availability statement

All data relevant to the study are included in the article or uploaded as supplementary information.