An 8-month (June 2024-January 2025) cross-sectional study in which structured questionnaires were administered to students by guidance counselors in the 5 secondary schools of intervention in the Southwest Region was conducted. The chosen schools in the Fako division were the most populated in terms of the student population and had more internally displaced students due to subsidized school fees.

This study essentially targeted students aged 10 - 24 years of both genders in the five participating schools found in the Fako division, Southwest region of Cameroon.

All students aged 10 - 24 years in the five selected secondary schools who gave their consent.

Consecutive sampling was used to select the students aged 10 - 24 years in the five participatory secondary schools as identified by the assigned guidance counselors in these schools.

Coded questionnaires made up of standardized scales to assess knowledge of mental health disorders and screen for anxiety, depression, and substance use were shared with students in their classrooms during free periods by guidance counsellors with lived experiences, assisted by a research assistant. Clinical psychologists and research staff trained the guidance counsellors in applying the scales to their various school settings. The questionnaires were cross-checked, validated, and signed by guidance counselors. These questionnaires were collected every week, checked again for completeness, and validated.

The sample size was determined via Cochran’s formula [Z²p (1 − p)/d²], and the estimated proportion of mental disorders among students was 50%, 95% confidence level, and 5% margin of error. This gave a minimum sample size of n = (1.96)² × 0.5 (1 − 0.5)/(0.05)² = 384. A 10% contingency addition was performed on the sample size for nonresponses; the final sample size was 423 participants [20] . 423 was the minimum sample size calculated but a total of 958 participants were recruited to improve data quality and to enhance precision and reliability.

The questionnaire was divided into 3 main sections: section A had 6 questions designed to assess the sociodemographic characteristics of the study participants, section B comprised 18 questions aimed at evaluating students’ knowledge of mental health disorders, and section C had 34 questions that focused on assessing the diagnosis and prevalence of mental health disorders [anxiety (7), substance abuse (4) and depression (9)]. The knowledge section was adapted from the Mental Health Knowledge Schedule (MAKS), which evaluated participants’ understanding of mental health concepts.

For the prevalence assessment, the following standardized screening tools were employed: the Generalized Anxiety Disorder 7-item scale (GAD-7) to evaluate anxiety symptoms, the CAGE (Cut, Annoyed, Guilty, Eye) Substance Abuse Screening Tool to assess substance use, and the Patient Health Questionnaire 9-item scale (PHQ-9) to measure depressive symptoms. These tools were selected for their validity, reliability, and sensitivity in evaluating mental health disorders among adolescents and young adults.

For anxiety, a score ≥5 indicated anxiety, whereas for substance use, a score ≥1 suggested a problem, and for depression, a score ≥5 indicated depression. Anxiety was measured with seven statements scored from 0 to 21, and respondents with scores ≥5 were considered to have anxiety. Suicide risk was assessed with four questions scored from 0 to 4, and respondents with scores ≥1 were considered to be suffering from suicidal thoughts. Depression was determined with nine statements scored from 0 to 27, and respondents with scores ≥5 were considered to have depression. The overall prevalence was determined by the presence of any of these four disorders. For substance use, ≥2 indicated an alcohol problem.

The coded questionnaires were thoroughly checked for completeness, and the data were entered into the Kobo Toolbox via the Open Data Kit application installed on Android phones. The data were then downloaded as an Excel file from the secure Kobo Toolbox database, cleaned, coded, and exported to SPSS version 25 for analysis. The data were saved in a password-protected file to prevent unauthorized access.

Regarding knowledge of mental health disorders, each correct answer was awarded a score of 1. The total score for each participant was calculated by summing all the knowledge questions. Participants who scored ≤ 9 points were considered to have poor knowledge, whereas those who scored above 9 points were deemed to have good knowledge of mental disorders. The disparity in knowledge between schools and communities was determined by comparing knowledge scores between participants in schools and communities. Regarding. The prevalence of anxiety, suicidal thoughts, depression, and substance use was used to determine the prevalence of mental health disorders in the study. For specific mental health disorders, anxiety was assessed via 7 statements with a four-point Likert scale, where 0 = not at all and 3 = nearly every day. The scores were calculated for each respondent, with the total scores ranging from 0 to 21 points. The respondents had anxiety if they scored ≥5 and no anxiety if they scored <5. For the substance use questions, a response of “yes” was assigned a score of 1, whereas “no” was assigned a score of 0. The total scores ranged from 0 to 4 points. Respondents with 1 point had no problem, whereas all the respondents with ≥1 point had a problem with substance use. For depression screening, 9 statements were used on a four-point Likert scale, where 0 = not at all and 3 = nearly every day. The scores were calculated for each respondent, with the total scores ranging from 0 to 27 points. The respondents had depression if they scored ≥10 and no depression if they scored <10. For substance abuse, 4 statements were used on a four-point Likert scale. The scores were calculated for each respondent, with the total scores ranging from 0 to 4 points. The respondents had substance abuse disorders if they scored ≥2 and no substance abuse disorder if they scored <2.

Ethical clearance with number: 2023/2066-3/UB/SG/IRB/FHS was sought and obtained from the Institutional Review Board of the Faculty of Health Sciences, University of Buea. We collaborated with the Ministry of Youth and Civic Education and the School administration to reach out to these students in the various schools. The school administration and guidance counsellors stood in as legal guidance for students. Confidentiality was maintained for participant responses through proper storage of the data in an encrypted system. Data was then extracted from the records with confidentiality and securely managed using the Open Data kit database with a password for protection. Parental consent was obtained through a written consent form, which was distributed to parents or guardians through the school. The consent form was sent home with students and returned to the school. Participatory students aged 10 - 17 years provided assent through a separate age—age-appropriate assent form, which was administered by the guardian counsellors in the school

A total of 958 study participants were recruited successfully; the mean age of the respondents was 15.41 years, with a range of 10 - 24 years. Most of the study participants (614, 64.1%) were aged 15 - 19 years. Additionally, 585 participants (61.1%) were female, and the majority of the participants (370, 38.6%) were from the locality of Limbe, with 581 being from the host population ( Table 1 ).

Almost all (574, 59.9%) of the study participants knew that mental health disorders are a type of medical disorder, and the majority (774, 80.8%) recognized mental health disorders as treatable. Additionally, 776 (81.0%) of the study participants reported that mental health disorders are not contagious, and 804 (83.9%) indicated that living alone is not a treatment for mental illness. Additionally, many (655, 68.8%) of the respondents indicated that the development of mental health problems is identical for males and females. Talking or laughing alone and excessive or unusual happiness were reported as the major symptoms of mental illness by 648 (67.6%) and 601 (62.7%) of the study participants, respectively ( Table 2 ).

The median overall knowledge score value was 28, with minimum and maximum values of 4 and 41 out of 44 knowledge items, respectively. The overall knowledge score indicated that 55.3% of the respondents (530 individuals) had an adequate understanding of mental health problems ( Figure 1 ).

Univariate logistic regression analysis was used to test the associations between sociodemographic factors and knowledge of mental health disorders. The logistic regression analysis results revealed that age group, gender, particularly male sex [p = 0.029], OR: 0.748 (0.576 - 0.971)] and locality of Tiko [p = 0.001, OR: 0.384 (0.226 - 0.670)], were significantly associated with knowledge of mental disorders, with a cut-off of <0.2 ( Table 3 ).

All factors significant in the univariate analysis were analysed via multivariate logistic regression analysis. The associations between sociodemographic factors and knowledge of mental health disorders were tested. The logistic regression analysis results revealed that sex, particularly male sex [p = 0.021), OR: 0.730 (0.559 - 0.954)] and locality of Tiko [p = 0.000), OR: 0.368 (0.213 - 0.639)], were significantly associated with knowledge of mental disorders (p < 0.05) ( Table 4 ).

Among the 958 participants in this study, 73.6% (705) of the participants had mental health disorders, with 54.7% suffering from anxiety disorders, 42.2% suffering from substance use disorders, 19.8% suffering from suicidal thoughts, and 54.7% suffering from depression ( Figure 2 , Table 5 ).

Univariate logistic regression analysis was used to test the associations between sociodemographic factors and knowledge of mental health disorders. The logistic regression analysis results revealed that the age groups 10 - 14 years (p = 0.020, OR: 0.180), 15 - 19 years (p = 0.036, OR: 0.215), sex (p = 0.188, OR: 0.811) and suicidal thoughts (p = 0.000, OR: 0.213) were significantly associated with knowledge of mental disorders, with cut-off values <0.2 ( Table 6 ).

For all the factors that were significant in the univariate analysis, a multivariate logistic regression analysis was performed. The logistic regression analysis results, presented in Table 7 below, revealed that age groups 10 - 14 years (p = 0.025, OR: 0.188), 15 - 19 years (p = 0.040, OR: 0.219) and suicidal thoughts (p = 0.00, OR: 0.219) were significantly associated with the prevalence of mental health disorders, with p values <0.05 ( Table 7 ).

Notably, 55.3% of secondary school students demonstrated adequate knowledge of mental health disorders. While this level of awareness is encouraging, it is lower than the 65.3% reported by Siddique et al. (2022) [22] among university students. This difference likely reflects the increasing exposure to mental health information and resources students receive at the university level. The identification of “talking or laughing alone” and “excessive or unusual happiness” as major symptoms by a significant number of participants was consistent with the findings of Tesfaye et al. (2021) [23] . This reveals the persistence of misconceptions and stigmatizing stereotypes surrounding mental health. While the understanding that mental health disorders are not contagious and that social isolation is not a helpful treatment provides a foundation for accurate knowledge, the observed misconceptions underscore the critical need for comprehensive and nuanced mental health education in secondary schools.

Our analysis revealed a significant association between sex and knowledge of mental health disorders. Males presented significantly lower odds of possessing adequate knowledge than females did (OR = 0.730, 95% CI: 0.559 - 0.954). This finding aligns with existing research suggesting potential gender disparities in health literacy and access to health information. Sociocultural norms that discourage men from expressing vulnerability and seeking help for mental health concerns may contribute to this discrepancy [24] .

Further investigations are warranted to explore the underlying causes and develop targeted interventions to improve mental health knowledge among male students. Furthermore, a significant association was found between locality and knowledge of mental health disorders. Residents of the locality of Tiko presented significantly lower odds of adequate knowledge than those from Limbe did (OR = 0.368, 95% CI: 0.213 - 0.639). This stark difference underscores the potential for disparities in access to mental health resources and information across different geographic areas within the Southwest Region. Factors such as the limited availability of healthcare professionals, transportation barriers, socioeconomic disadvantages, and variations in the quality of education may contribute to this knowledge gap [25] . Targeted interventions are needed to address these disparities and ensure equitable access to mental health information for all students, irrespective of their location. These interventions should include community-based mental health awareness campaigns and the integration of mental health education into school curricula.

The overall prevalence of mental health disorders among the students in this study was 73.6%, which is alarmingly high. A substantial proportion of the sample experienced anxiety (54.7%), substance use disorders (42.2%), suicidal thoughts (19.8%), and depression (57.3%). This prevalence rate is significantly higher than the 50% reported by Sakthivel et al. (2021) [20] in a different population. This could be because our study evaluated a larger sample size. However, differences in population characteristics may have also contributed. This difference suggests a potentially greater burden of mental health issues among secondary school students in the Southwest Region of Cameroon, which may be related to the ongoing sociopolitical instability and associated trauma in the country. Prolonged conflict has resulted in displacement, economic hardship, and increased stress levels, all of which can significantly contribute to mental health problems, especially among vulnerable populations such as adolescents [19] .

Multivariate logistic regression analysis revealed significant associations between several factors and the prevalence of mental health disorders. Specifically, younger age groups (10 - 14 years: OR = 0.188, p = 0.025; 15 - 19 years: OR = 0.219, p = 0.040) and the presence of suicidal thoughts (OR = 0.219, p = 0.00) were less likely to develop mental health disorders. This may reflect their lesser vulnerability to environmental stressors and limited coping mechanisms. The strong association between suicidal thoughts and mental health disorders underscores the importance of early identification and intervention to prevent potential self-harm.