Mental health in adolescents is important because of its impact on the journey to adulthood. It's a time of rapid physical, social and emotional development, during which adolescents are susceptible to all forms of abuse, violence and stress. The period between 10-19 years is a critical time and experiences during this time can have lifelong implications, with many mental health disorders originating in adolescence [1]. An estimated 13% of adolescents worldwide suffer from a mental disorder, many of which go unrecognised [2]. Depression, anxiety and disruptive behavioural disorders are major causes of illness and disability of adolescents, and can lead to social isolation and exclusion, poor academic performance, risky behaviour and poor physical health [3]. Rates of mental health issues rise from around 2% in childhood to 10-20% in late adolescence, and then to adult levels [4].   Prevalence of psychiatric disorders is estimated to be 20% among adolescents, with anxiety and depression most prevalent [5]. Rates of anxiety disorders, depression, and coexisting anxiety and depression in older adolescents are approximately 5.5%, 3.5% and 2.5% respectively, which can lead to academic difficulties, social isolation and suicidal thoughts [9,10]. Suicide is a major cause of death among people aged 15-29 years, with India having one of the highest rates of youth suicide in the world [3,6,7]. Other common co-morbid disorders include substance abuse and conduct disorders [8].   Disruptive behaviour disorders (such as attention deficit hyperactivity disorder and conduct disorder) are more prevalent in early adolescence (10-14 years), and include inattention, impulsivity and aggression affecting learning and peer interactions. In India, which has a high number of adolescents (around 253 million), 7.3% of the 13-17-year-olds suffer from psychiatric disorders [9,10,11]. A meta-analysis shows higher rates in school samples than in the community [12,13].   Adolescent mental health issues are influenced by several factors, including poor family and home environments, punitive parenting, bullying, socioeconomic hardships, stigma and lack of access to services. These factors along with biological, psychological and social factors such as family, peer, school factors, substance use and unmet expectations play a role [10,14,15]. Research conducted after the COVID-19 pandemic also shows the impact of parents' mental health, physical activity and health complications on adolescents' behavioural problems [16].   The high burden and the complexity of emotional and behavioural problems (EBP) in adolescents make it necessary to conduct targeted research to estimate the prevalence and predict the factors associated with EBP in school-going adolescents. Earlier diagnosis is important for non-pharmacological interventions, preventing further complications and ensuring the well-being and human rights of adolescents.

The current study was an analytical study and carried out in November, 2022 with school going adolescents aged 13-16 years in selected schools in Thiruvananthapuram district, urban and rural, Kerala, India. Multi-level cluster sampling was used. The Thiruvananthapuram district has three educational districts, and of them, two were chosen at random. Out of these, schools were selected in a proportional manner in Government, Government-aided and unaided sectors. The study comprised adolescents aged between 13 and 16 years who were pursuing their studies in selected schools and gave assent, including the informed consent of their parents. The adolescents who had mental sub-normality or physical disability and those who were already on medication or under psychological treatment of any mental issues were not to be included in the study.   Sample Selection and Sampling Technique The sample size was estimated based on the formula 4pq/d 2 since a previous prevalence study in Kollam district (20172021) indicated a prevalence of 24.5% of emotional and behavioural problems. The lowest possible sample size was 680 considering the design effect. Eight hundred and forty-four teens were sampled through cluster-selection. Each sector had two schools. Random selection of one or two divisions of each standard was done to select adolescents aged 13 to 16 years who study in classes 8th, 9th, and 10th.   Data Collection Methods and Instruments Data were collected using a structured set of tools through a pen-and-paper self-report method. Socio-demographic information was obtained using a socio-demographic data sheet that included age, gender, class, and family-related details. Emotional and behavioural problems were assessed using the Strengths and Difficulties Questionnaire (SDQ 11–17) self-report version developed by Robert Goodman [17], and the validated Malayalam version was used [18]. Smartphone addiction was assessed using the Smartphone Addiction Scale–Short Version (SAS-SV) developed by Kwon M [19].   Ethical Considerations The study was reviewed and approved by the Institutional Review Board and the Government-approved Institutional Ethics Committee (No. 005/05/2022/IEC/SMCH) on 05/01/2022. Administrative approval was obtained from the Deputy Director of Education, Thiruvananthapuram District, and the school authorities. Parents gave written informed consent and students gave their assent before the data collection. The students were briefed about the study, and confidentiality was ensured. We then screened 804 students for emotional and behavioural difficulties by using the Strengths and Difficulties Questionnaire (SDQ).   StatisticalAanalysis Data were entered and coded using the Statistical Package for the Social Sciences (SPSS) version 20. Descriptive statistics were used to determine the prevalence of emotional and behavioural problems (EBP). Univariate analysis and binary logistic regression were applied to identify predictors of EBP. A p-value≤0.05 was considered statistically significant.

According to Table 1, the average age of the participants was 14.04 +1.2. Boys and girls were almost equal (50.1% and 49.9%). Most of the adolescents belonged to nuclear family (68.3%), and rural area (54.4%). The mother of the adolescent was of collegiate education over half of it (55.2%). Most of the adolescents (83.5%) reported to have social support; 85% have involvement in extracurricular activities and 71.1% used to spend 0-2 hours on mobile phone. Participation from Government and Government -aided schools were 49.6% and 40.4% respectively.   Table 1: Socio-Demographic Characteristics of Adolescents (n = 804) S.No. Variable Category Frequency Percentage 1 Age in years 13.0 228 28.4 14.0 344 42.8 15.0 200 24.8 16.0 32 4.0 2 Gender Male 403 50.1 Female 401 49.9 3 Area of residence Urban 367 45.6 Rural 437 54.4 4 Type of family Nuclear 549 68.3 Joint 255 31.7 5 Father’s education School 470 58.5 College 334 41.5 6 Mother’s education School 360 44.8 College 444 55.2 7 Monthly family income in Rupees < 25000 627 78.0 25001-40000 98 12.2 > 40000 79 9.8 8 Marital status of parents Married staying together 649 80.7 Single parent 155 19.3 9 Parental separation in childhood Yes 121 15.0 No 683 85.0 10 Family history of psychiatric illness Yes 30 3.7 No 774 96.3 11 Distress due to parental dispute Yes 290 36.0 No 514 64.0 12 Substance abuse in the family Present 230 28.6 Absent 574 71.4 13 An illness that needs regular treatment Present 76 9.5 Absent 728 90.5 14 History of psychoactive substances Present 44 5.5 Absent 760 94.5 15 Engagement in hobbies Present 553 68.8 Absent 251 31.2 16 Social support Present 671 83.5 Absent 133 16.5 17 Physical exercise Yes 229 28.5 No 575 71.5 18 Involvement in extracurricular activities Present 684 85.0 Absent 120 15.0 19 Time spent on mobile 0-2 hrs 572 71.1 3-5 hrs 140 17.4 6 hrs or more 92 11.5 20 The sector of schools Government 399 49.6 Govt-aided 325 40.4 Unaided 80 10.0   Figure1 indicates that on the basis of SDQ (11-17) Score, 10.4% of adolescents had clinically significant EBP (abnormal score 20-40), and 19.9% had slightly elevated score that indicate significant problem (borderline score 16-19) and sum of them 30.3 was the prevalence of EBP among adolescents.     Figure 1: Prevalence of Emotional and Behavioural Problems among Adolescents based on TDS   The Table 2 shows the prevalence of diverse emotional and behavioural problems (EBP) in adolescents as per SDQ scores. Most common were conduct problems, which were present in 31.1% of adolescents, 17.4% with clinically significant problems (abnormal), and 13.7% with slightly elevated problems (borderline), and indicates possible concern. Adolescents had hyperactivity (16.6) with 8.0% with abnormal score and 8.6% with borderline score. Embodiment issues existed in 12.9, 7.1 were abnormal and 5.8 were borderline. The clinical significance of peer problems was 11.3 and 53.6% of the adolescents scored borderline. In terms of prosocial behaviour which indicates strengths, only 8.2% scored in the abnormal range and majority (82.1) had demonstrated normal prosocial behaviour.   Table 2: Prevalence of Emotional-Behavioural Problems among Adolescents based on Scores of Different Domains of SDQ (n = 804) Sl. No Emotional-behavioural problems (EBP) Grading of EBP score Frequency Percentage 1. Emotional problems Abnormal (7-10) 57 7.1 Borderline (6) 47 5.8 Normal (0-5) 700 87.1 2. Hyperactivity Abnormal (7-10) 64 8.0 Borderline (6) 69 8.6 Normal (0-5) 671 83.4 3. Conduct problems Abnormal (5-10) 140 17.4 Borderline (4) 110 13.7 Normal (0-3) 554 68.9 4. Peer problems Abnormal (6-10) 91 11.3 Borderline (4-5) 431 53.6 Normal (0-3) 282 35.1 5. Pro-social behaviour Abnormal (0-4) 66 8.2 Borderline (5) 78 9.7 Normal (6-10) 660 82.1   Table 3 presents the univariate analysis of socio-personal variables associated with SDQ scores indicating emotional and behavioural problems (EBP). There was no significant association between EBP and gender and place of residence. There were significant correlations with family type, maternal education, childhood parental separation, distress because of parental quarrels, family substance abuse, extended illness treatment, teenage substance abuse, absence of hobbies, inadequate social support, physical inactivity, excessive cell phone use, and abnormal prosocial behaviour. On the whole, EBP was significantly linked to familial, psychosocial and behavioural factors, and maternal education and prosocial behaviour became significant predictors.   Table 3: Univariate Analysis of Socio-Personal Factors of Emotional and Behavioural Problems Socio-personal Variables Category SDQ Score Total c2 P OR 95%CI Borderline and Abnormal Normal No (%) No (%) No Gender Male 117 (29.0) 286 (71.0) 403 0.66 0.416 0.9 0.65-1.2 Female 127 (31.7) 274 (68.3) 401 Place of residence Urban 121 (33.0) 246 (67.0) 367 2.2 0.138 1.3 0.93-1.69 Rural 123 (28.1) 314 (71.9) 437 Type of family Nuclear 185 (33.7) 364 (66.3) 549 9.2 0.002 1.7 1.2-2.4 Joint/extended 59 (23.0) 196 (76.9) 255 Mother Education Primary 32 (71.1) 13 (28.9) 45 37.5 0.001 6.4 3.3-12.3 High school and above 212 (27.9) 547 (72.1) 759 Parental separation during childhood Yes 46 (38.0) 75 (62.0) 121 3.9 0.047 1.5 1.0-2.2 No 198 (29.0) 485 (71.0) 683 Distress due to parental dispute Yes 108 (37.2) 182 (62.8) 290 10.2 0.001 1.7 1.2-2.2 No 136 (26.5) 378 (73.5) 514 Substance abuse in family Yes 94 (40.9) 136 (59.1) 230 16.9 0.001 1.9 1.4-2.7 No 150 (26.1) 424 (73.9) 574 Prolonged treatment for illness Yes 35 (46.1) 41 (53.9) 76 9.8 0.002 2.1 1.3-3.4 No 209 (28.7) 519 (71.3) 728 Use of psycho active substance Yes 20 (45.5) 24 (54.5) 44 5 0.025 1.9 1.1-3.7 No 224 (29.5) 536 (70.5) 760 Hobbies No hobbies 89(35.5) 162 (64.5) 251 4.5 0.034 1.4 1.0-1.9 Hobbies 155 (28.0) 398 (72.0) 553 Social support No support 55 (41.4) 78 (58.6) 133 9.1 0.003 1.8 1.2-2.6 Yes 189 (28.2) 482 (71.8) 671 Physical exercise No 193 (33.6) 382 (66.4) 575 9.8 0.002 1.8 1.2-2.5 Yes 51 (22.3) 178 (77.7) 229 Time spent on mobile Long-term use 88 (37.9) 144 (62.1) 232 8.8 0.003 1.6 1.2-2.3 Mild or no use 156 (27.3) 416 (72.7) 572 Prosocial behaviour Abnormal 85 (59.0) 59 (41.0) 144 68.3 0.001 4.5 3.1-6.6 Normal 159 (24.1) 501 (75.9) 660 Smart phone addiction (SAS) Yes 110 (67.1) 54 (32.9) 164 132 0.001 7.7 6.3-11.2 No 134 (20.9) 506 (79.1) 640   Table 4 showed the binary logistic regression analysis was conducted to determine independent predictors of emotional and behavioural problems (EBP) in adolescents given the variables that are significant in the univariate analysis. Primary maternal education or lower, distress in parental conflicts, chronic physical illness, physical inactivity, and abnormal prosocial behaviour were found to be important independent predictors of EBP. The regression model did not show any statistically significant variables like family type, parental separation, adolescent substance use, hobbies, social support and mobile phone use. These results demonstrate that adolescent EBP is affected by various factors, especially maternal education, family environment, physical health, lifestyle behaviours and prosocial functioning.   Table 4: Binary Logistic Regression Analysis of Predictors of Emotional and Behavioural Problems Variables B S.E. Wald df Sig. Exp(B) 95% C.I. Lower Upper Gender -0.545 0.194 7.911 1 0.005 0.580 0.397 0.848 Type of family 0.295 0.198 2.229 1 0.135 1.343 0.912 1.978 Mother Education 1.268 0.409 9.631 1 0.002 3.555 1.596 7.922 Separation from Parents 0.322 0.244 1.745 1 0.187 1.379 0.856 2.223 Distress due to Parental Dispute 0.547 0.185 8.755 1 0.003 1.727 1.203 2.481 Substance Abuse 0.389 0.200 3.778 1 0.052 1.475 0.997 2.183 Prolonged Treatment for Illness 0.864 0.277 9.752 1 0.002 2.372 1.379 4.079 Use of Psychoactive Substance 0.493 0.383 1.655 1 0.198 1.638 0.772 3.472 Hobbies 0.029 0.192 0.023 1 0.880 1.029 0.707 1.499 Social Support 0.213 0.243 0.762 1 0.383 1.237 0.767 1.993 Physical Exercise 0.470 0.213 4.872 1 0.027 1.601 1.054 2.430 Times pent on mobile -0.045 0.208 0.047 1 0.829 0.956 0.636 1.438 SAS 1.741 0.230 57.040 1 0.000 5.700 3.629 8.955 Prosocial 0.964 0.231 17.457 1 0.000 2.623 1.668 4.123 Constant -4.657 0.699 44.428 1 0.000 0.009 - -

In the current research, the proportion of adolescents with abnormal scores according to the total difficulties score (TDS) of the SDQ was 10.4%, indicating a clinically significant level of emotional and behavioural problems. The abnormal scores of emotional problems, conduct problems, hyperactivity, and peer problems were 7.1%, 17.4%, 8.0%, and 11.3%, respectively. Although a significant association between gender and TDS was observed (χ² = 9.0, p = 0.011), binary logistic regression did not identify gender as a significant predictor of EBP. Univariate analysis revealed predictors such as type of family, maternal educational status, parental separation, parental conflict, substance abuse in family members, long-term physical illness, psychoactive substance use among adolescents, lack of leisure activities, poor social support, physical inactivity, excessive mobile use, and abnormal prosocial behaviour. Logistic regression further identified maternal education, parental conflict, long-term illness, physical inactivity, and abnormal social behaviour as key predictors.   A pre-university study in Bangalore reported similar findings, with abnormal TDS scores in 10.1% and subscale abnormalities of 9%, 13%, 12.6%, and 9.4% for emotional, conduct, hyperactivity, and peer problems, respectively. Predictors included gender, maternal education, parental discord, and substance abuse [20]. A cross-sectional study also identified intrafamilial communication and parental mental health disorders as predictors [21]. The present study supports parental conflict as a predictor but not family psychiatric history.   An Egyptian rural adolescent study reported an abnormal impact score of 13.7%, comparable to the present findings [22]. However, parental loss was a predictor in that study but not in the current one. A Turkish orphanage-based study also showed no association between parental psychiatric illness and EBP, aligning with present findings [23]. Regression analyses from similar studies highlighted family environment and maternal education as major determinants.   The prevalence of EBP in Thiruvananthapuram was 30.3% (borderline + abnormal), comparable to Kollam (24.5%) and Chandigarh (30%) studies [24]. Conduct problems were the most prevalent behavioural issue (31.1%), consistent with school-based surveys reporting 36.4% conduct disorders [25]. A meta-analysis across India also identified emotional, conduct, hyperactivity, and peer problems as major domains influenced by socioeconomic, environmental, and lifestyle factors [26]. Additional supporting evidence indicates that parental absence, alcoholism, and adverse family environments significantly predict behavioural issues [27,28].

Emotional and behavioural problems among children and adolescents result from the interplay of personal attributes, family dynamics, and environmental stressors, including parental mental health, family conflict, socioeconomic factors, and school-related pressures. The findings highlight the importance of implementing school-based screening to facilitate early identification and enable timely, targeted interventions to address these risk factors effectively.   Limitations The study is limited by its cross-sectional design, which restricts causal interpretation of the identified associations. Reliance on self-reported data may introduce reporting bias. The use of a single geographic setting and specific school population limits generalizability. Potential confounding variables, such as detailed family psychiatric history and environmental factors, were not fully controlled.