Cross-sectional study was conducted among the secondary school students in the city of Chennai, Tamil Nadu, during the month of September and October 2019 after the approval from the institutional ethics committee of Schizophrenia Research Foundation. The schools were approached and convinced about the program and its logistics. The five schools which provided consent and approval for conducting the study were included.

All the students of class 9 were approached to take part in the study. As the students were <18 years, parental information sheet and parental consent form was sent to the parents. The opt-out method was used when obtaining the consent from the parents/guardian, where the parents/guardian who did not want their children to participate had to send the forms back stating so. After the consent from the parents/guardian, assent was obtained from the student before inclusion into the study. To maintain the anonymity and confidentiality, unique identification numbers were created for each participant. There was no exclusion criteria for the recruitment of the participants.

Socio-demographic profile: Semi-structured proforma was developed to collect the socio demographic details. The socio demographic details are grouped into three broad categories namely basic information, education, and family details.

Perceived stress: The perception of stress was measured using Cohen’s Perceived Stress Scale.¹⁴ It is the most widely used scale for the measurement of perception of stress. PSS includes a number of direct questions that measures the level of experienced stress during the last month. The scores ranging from 0 to 4 for each question.

Psychological well-being: The psychological well-being of the students was assessed using the World Health Organisation- Five Well-Being Index (WHO-5).¹⁵ The short self-reported and generic global rating scale measures the subjective mental well-being during the last 14 days. It consists of five positively phrased statements and have the scores ranging from 0 (none of the time) to 5 (All the time).

Anxiety: General Anxiety Disorder (GAD-7) is a self-administered patient questionnaire and was used to screen and measure the severity of generalized anxiety disorder.¹⁶ The seven items in GAD-7 rates the severity of the symptoms over the last two weeks.

Depression: Patient Health Questionnaire (PHQ-9), 9 item scale was used to assess and monitor the severity of depression.¹⁷ Each item is scored 0 to 3 providing a severity score of 0 to 27. The cut points were calculated from the score and each represents the severity, mild, moderate, moderately severe and severe depression.

Before administering the questionnaires to the students, they were explained about the purpose of the study, and assured of anonymity and confidentiality. Research assistants distributed the paper-based questionnaires to all the participants present in class on the day of administration and responses obtained. Questionnaire was administered both in English and Tamil based on the preference of the study participants. The time taken for the introduction and administration of questionnaire was around 1 hour 20 minutes.

The collected data were entered in Microsoft Excel and Statistical Package for Social Sciences (SPSS) Version 17.0 was used for the analysis of the collected data. The categorical variables were expressed in frequency and percentages and the continuous variables in mean. Chi-square test were used to find out the association between the Socio-demographic variables and Anxiety, Depression. It was considered that p-value < 0.05 statistically significant. Multivariate logistic regression was used to examine the association between different variables and anxiety & depression.

A total of 569 students were included in the study. The mean age of the participants was 13.6 ± .6. Of the 569 participants, 323 (56.8%) of them were male. 561 (98.8%) of them did not discontinue school ever. Only few students (10, 1.8%) had already received a course on mental health via other available resources, while most of them did not receive any kind of structured information on mental health. The summary of the socio-demographic profile of the study participants is depicted in Table 1.

Table 2 provides the prevalence of anxiety and depression among the study participants. The prevalence of anxiety was 46.3% (95%CI: 42.2-50.6 CI) and depression was 39.6% (95%CI: 35.6-43.8 CI).

Results from the univariate (Supplementary Table 1a, 1b) and multivariate logistic regression analysis (Table 3) showed blue collar occupation of the father, monthly household income and high perceived stress levels were statistically associated the prevalence of anxiety among the study participants. Whereas, gender, previous discontinuation of schools, parents living status, and school attendance in last 30 days were not significantly associated with the prevalence of anxiety. DISCUSSION

After univariate analysis (Supplementary Table 2a, 2b), multivariate logistic regression analysis was done to identify the factors associated with the prevalence of depression (Table 4). Results indicate that higher prevalence of depression was associated with students’ fathers who has blue collar job and students who expressed high perceived stress. No gender differences were observed in the prevalence of depression among the study participants.

The study aimed to estimate the prevalence of depression and anxiety and assess the socio-demographic factors associated with them among secondary school students in Chennai.

Our study found that the prevalence of anxiety was 46.3% and depression was 39.6% among secondary school students. As per the national mental health survey 2016, the prevalence of metal health problems in adolescents between the ages 13 – 17 was 7.3% and the rates were nearly double (13.5%) in urban metros.8 In comparison, a meta-analysis by Malhotra and Patra summarized that the prevalence of mental health problems in school settings was 23.3%.¹⁸

Previous studies from India have shown a wide variation in the prevalence of anxiety among the secondary school students, ranging from 17% to 85%.^{19 , 20 , 21 , 22} Previous studies carried out in India have reported an even greater prevalence of depression among school going adolescents.^{23 , 24 , 25} Especially, study by Daya & Karthikeyan in 2018 reported that 73.6% of school going adolescents of Tirunelveli district, Tamilnadu have depression which is very higher than our findings.10 Jayanthi et al., in 2014 found school going adolescents in Tamil Nadu had 45.7% moderate, 25.4% mild, 19.6% severe and 9.3% minimal depression.26 The differences in the tools and their cut off scores used, the site of study, socio-economic status of the population and age group of study participants included could explain the variability in the results.

Manystudieshaveidentifiedtheroleofparent’sSES(Socio Economic Status) in children’s growth, development, and health outcomes.²⁷ SES is not only determined by the income but also by the occupational status and the level of education. Our study shows association between household income and anxiety and depression in the univariate analysis, however in the multivariate analysis only father’s occupation which is also a part of SES has been found to have significant association with anxiety and depression among secondary school students. It was noted that students of fathers who are blue collar workers are more prone to anxiety and depression than students of fathers who are white collar workers. This possibly suggests that the parental occupation is a higher order factor that might be associated with the mental health of the students. Parent’s occupation plays an important role in psychological wellbeing of the children. Specific working conditions influence certain psychological qualities, parent child relationships and other behaviours such as alcohol consumption and domestic violence.

A study conducted by Whitbeck, Les B., et al. in 1997 showed that the father’s working conditions affect the fathers parenting behaviours.²⁸ Previous studies have observed the association between parenting styles adopted and depression and anxiety in children.^{29 , 30}

Our study shows significant association between the outcomes of anxiety and depression and stress. Many previous school-based cross-sectional studies among adolescents have also showed strong association between academic stress and anxiety and academic stress and depression.^{31, 32}. Gender was not associated with the presentation of anxiety and depression among the students. This is unlike most of the other studies which have found a significant association between these internalizing disorders and female gender.^{33 , 34} There are few previous studies that showed no association between and gender and depression similar to our study.^{13 , 24} This may be related to the higher overall prevalence of anxiety and depression reported by the students which would suggest common adverse environmental circumstances that might be associated with both genders.

The present study has few limitations that include that this is a cross-sectional, school-based study that utilized self-administered screening questionnaires. Due to this, we would not be able to access the students who had dropped out of schools. The use of self-reported questionnaires may be associated with under or over reporting of symptoms. However, all students were given information about the availability and method to access mental health services. Students found it hard to answer specific socio demographic variables like monthly household income and educational qualification of parents, leading to errors.

In conclusion, results of this study indicate a high prevalence of internalizing symptoms of depression and anxiety among adolescents and shows a strong indication towards the link of mental health problems and social determinants of health. The role of stress in the development of mental health issues indicates the need for future mental health promoting interventions to focus on building stress coping skills in school settings. The role of paternal occupation and its effects on the child’s mental health warrants further investigation. On a systemic level, addressing the modifiable risk factors and strengthening the protective factors would be essential to ensure the health and wellbeing of the students. More studies must be conducted in under privileged communities, schools, rural areas, and in community settings to get good representation of the sample.