Comprehensive Mental Health Facets And Association With Age, Stress And Coping Amid 'The New Normal' Pandemic: How Do Adolescents Differ From Adults?

Background: Given varied forms of stressors in the backdrop of the COVID 19 pandemic, adolescents are particularly vulnerable to specic mental health challenges. It pertinent to explore how adolescents differ from adults regarding the comprehensive mental health facets amid the 'new normal, i.e., the duration between the rst and second wave of COVID-19 in a developing country like India. The present study aimed to compare the comprehensive mental health facets of adolescents with the adults. Also, it aimed to explore the association of mental health facets with age, perceived stress, coping and impact of pandemic. Methods: The study followed a cross-sectional design with 1,027 participants [456 adolescents; 347 young and 224 middle age-adults] recruited from schools and colleges. The Comprehensive DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure, Perceived Stress Scale and Brief COPE Scale were used for assessment. The difference between groups was analyzed using the Chi-Square Test of Independence and correlational analysis was done using Spearman Rank Correlation. Multiple regression analyses were performed for the mental health facets along with bootstrapping method. Results: 33.77% of adolescents, 25.65% of young adults, and 17.41% of middle-aged adults reported that their symptoms started during the pandemic. Compared to adults, the adolescents reported higher depression, anxiety, suicidal ideations, anger, and somatic complaints. Signicantly higher adolescent females (39.9%) were found to have sleep disturbances compared to their male counterparts (25.5%). Therefore, it can be observed Adolescents are more clinically vulnerable in most domains. The correlational analysis showed that most mental health domains, except substance use showed moderate-to-low correlations with ‘Impact of COVID’. Perceived stress, impact of COVID, self-distraction, self-blaming were signicant independent positive predictors for all the mental health domains except substance use. ‘Age’ was negatively associated with depression anxiety, repetitive thoughts, personality changes, suicidal ideations and memory, and positively associated with ‘substance use’ at low levels. Maladaptive coping was moderately positively correlated with all the mental health domains. nature of data, the difference between groups was analyzed using the Chi-Square Test of Independence and correlational analysis was done using Spearman Rank Correlation. Multiple regression analyses were performed for the mental health facets along with bootstrapping method.

Moreover, it was indicated that students with poor nancial backgrounds, female students, and students living alone showed a higher prevalence of depressive and anxiety symptoms. [8] The pandemic has had severe nancial and work's implications for the adult population. With the rise in work-from-home culture, balancing professional and personal lives has also been challenging as individuals are expected to complete an increased amount of work at odd hours. [2] While no signi cant gender differences in levels of anxiety, stress, and depression have been found in most of the studies, [6,9] others do not accept this observation and show that females tend to be more vulnerable to these disorders. [10,11] Findings regarding education are also inconsistent; for instance, in Australia, distress was higher among less educated people, [12] whereas stress was higher among more educated participants in China. [13] Present study Given varied forms of stressors and experiences of the resultant stress due to COVID 19 pandemic, the adolescents could be particularly vulnerable to speci c mental health facets. In this backdrop there have been studies on stress, depression, and anxiety amongst people, including adolescents and adults, but most of them were conducted during the phase of 'outbreak' or 'lockdowns' [8][9][10][11][12] . To the best of the author's knowledge, these studies have neither covered comprehensive mental health outcomes in adolescents in a single research design nor provided a comparative analysis by comparing them with adults.
Hence, it was considered pertinent to explore how adolescents differed from adults regarding the comprehensive mental health facets amid the 'new normal' phase, i.e., the duration between the rst and second wave of COVID-19 in a developing country like India. The present study aimed to compare adolescents, young and middle-aged adults with comprehensive mental health facets (depression, anxiety, anger, repetitive thoughts and actions, somatic complaints, personality disturbances, memory, sleep problems, dissociation, suicidal ideation and substance use). Also, it aimed to explore the associations of mental health facets with age, perceived stress, coping and impact of pandemic. It was hypothesized that [1] adolescents would differ from adults on the comprehensive mental health facets [2] there would be an association of mental health facets with age, perceived stress, coping and impact of COVID 19 pandemic.

Participants and study characteristics:
This study employed a cross-sectional design by using a standardized tools through an online platform. It was part of an institutional project on mental health aspects during the pandemic. The proforma contained an assent form for adolescents and a consent form for adults, approved by the institute ethics committee. The ethical approval for conducting the study was obtained from the institutional ethics committee (ref code: 11th ECM COVID-19 1B/P7).
The study population comprises adolescents and adults between 13 to 60 years of age. All participants were selected by using a two-stage sampling technique. In the rst stage, schools/ colleges were selected, and in the second stage, participants were recruited from the selected schools/ colleges. They were further categorized as [a] adolescents (Ad = 13-19 years), [b] young adults (Y = 20-39) and [c] middle-age adults (MA = 40-60 years of age). Five schools and ve colleges were randomly selected from the city and contacted to participate in the study. Higher secondary schools and undergraduate/ postgraduate co-education colleges located within the city's urban limits were chosen.

Procedure
Three schools and four colleges agreed to participate in the study. The consenting school and college authorities were contacted for obtaining voluntarily informed consent to conduct a study in their institute and for coordination and implementation of data collection. The data collection was commenced on the noon of 1st December and concluded on noon of 31st December.
For the recruitment of adolescents, parents of the 13 to 18 years old adolescents were shared a consent form through email, which included information about the study, anonymity, and con dentiality. They were invited to give consent for the participation of their ward in the study. There was an option for parents to accept or decline the recruitment of their ward(s) in the study. After receiving the parents' consent, the assent form and 'the web link of study proforma' were mailed to the adolescents through email.
For the recruitment of adults, the web link of study proforma containing self-explanatory information about the study, informed consent form, questionnaire items, contact number, and o cial address of the investigators of the study, was disseminated through email to students of the colleges, parents, and staff of schools/colleges. All participants were provided contact numbers of the investigators for clari cations and were given an option to consent for an online interview. On a web link, participants got a set of questions in a sequential manner, which they had to answer by clicking on the given choices.

Measures
Semi-structured proforma: A self-explanatory web-based semi-structured proforma was developed, which was linked to a unique web link. This proforma collected sociodemographic information related to the participants [age, gender, education, occupation, and domicile]. For those who reported the presence of mental health symptoms, there were 3 items on whether 1-did not report any change in their symptoms and/or level of symptoms experienced during COVID-19 2-. the symptoms worsened during the pandemic 3-. the symptoms started during the COVID-19 outbreak, DSM-5 Self-Rated Level 1 Cross-Cutting Symptom Measure-9: [14] This tool was developed by the DSM-5 Task Force and Work Groups to review the mental systems in each patient. We used this measure assesses the presence and severity of mental health domains (depression, anxiety, anger, repetitive thoughts and actions, somatic complaints, personality disturbances, memory, sleep problems, dissociation, substance, suicidal ideation and substance use). The respondents are asked how much or how often they have been disturbed problems in the last 2 weeks (0 = none/never, 1 = slight/rare, 2 = mild/several days, 3 = moderate/more than half the days, 4 = severe/almost daily). Research evidence shows that the measure has good to excellent test-rested reliability, [15] and good validity. [16] Perceived Stress Scale (PSS): [17] PSS measures the extent to which one appraises the situations of one's life as stressful in the last month. It can be used in samples with minimum educational quali cation of junior high school. It is a 10 item 5-point Likert scale (0 = never, 1 = almost never, 2 = sometimes, 3 = fairly often, 4 = very often). PSS-10 shows a strong Cronbach's alpha = 0.84, with test-retest -reliability (ICC) to be 0.935 after three weeks and Kappa's Coe cient ranging from 0.74 to 0.89. [18,19] Brief COPE: [20] Brief COPE is a self-report questionnaire with 28 items designed to assess individuals' primary ways to cope with a stressful life event. The coping strategies were categorized as "adaptive" (active coping, use of emotional support, use of instrumental support, positive reframing, planning, selfdistraction, and acceptance) and "maladaptive" (behavioral disengagement, venting, denial, substance use, and self-blame) for this study. After taking due cognizance of the sociocultural backdrop of India, an item on religion was included. The items are rated on a four-point Likert scale (1 = I haven't been doing this at all, 2 = I have been doing this a little bit, 3 = I have been doing this a medium amount, 4 = I've been doing this a lot). The individual scores for each subscale is, then, computed by adding scores on all the items separately. The scale has shown high Cronbach's alpha values ranging from 0.69 to 0.8. [21] Data analysis The results were analyzed using descriptive statistics and making comparisons among various groups. The analyses were performed using SPSS version 23.0 (SPSS Inc., Chicago, IL, USA). Categorical data were summarized as proportions and percentages (%) and discrete as mean and standard deviations (SD). Given the non-parametric nature of data, the difference between groups was analyzed using the Chi-Square Test of Independence and correlational analysis was done using Spearman Rank Correlation. Multiple regression analyses were performed for the mental health facets along with bootstrapping method.

Results
A total of 1,027 participants have taken part in the study including 456 adolescents (M = 200; F = 256); 347 young adults (M = 124; F = 223) and 224 middleaged adults (M = 103; F = 121). The mean age of male females was found to be comparable among all across the groups. In the adolescent age group majority were students of intermediate (Table 1-a).    The correlational analysis showed 'age' was negatively associated with depression anxiety, repetitive thoughts, personality changes, suicidal ideations and memory, and positively associated with 'substance use' at low levels. There was moderate positive correlations between perceived stress and all the mental health domains. All mental health domains, except substance use showed moderate-to-low correlations with 'Impact of COVID'. Moreover, substance use itself showed moderate positive correlations with other mental health domains. Maladaptive coping was moderately positively correlated with all the mental health domains whereas adaptive coping showed only low positive correlation with anxiety.

Regression Analysis
The regression analyses show that coping strategies, perceived stress, age, gender and impact of COVID provide a signi cant model for prediction of depression, anxiety, suicidal ideation and substance use. Perceived stress, impact of COVID, self-distraction, self-blaming were signi cant independent positive predictors for all the mental health domains except substance use. Behavioural disengagement also showed positive association with depression, anxiety, suicidal ideation, anger, somatic concern, sleep, repetitive thoughts, personality disturbances and dissociation. Emotional support and Religion were negative independent predictors for depression, anxiety, suicidal ideation, repetitive thoughts, personality disturbances, somatic concern and dissociation; and Anger was also predicted by Emotional Support. Denial was a signi cant independent predictor for depression and repetitive thoughts and actions, and gender was an independent predictor for anger and somatic concern. Suicidal ideation also had active coping with negative association. Positive reframing was in a negative relation with substance use and denial, perceived stress, substance use as coping were signi cant independent predictors for maladaptive substance use. In addition, age was a signi cant independent predictor for depression, anger and personality disturbances with negative association.
The regression analyses show that coping strategies, perceived stress, age, gender and impact of COVID provide a signi cant model for prediction of depression, anxiety, suicidal ideation and substance use. Perceived stress, impact of COVID, self-distraction, self-blaming, religion, age, behavioural disengagement, emotional support and denial were signi cant independent predictors for depression and anxiety, where emotional support and religion were negatively associated. In addition, age and denial were signi cant independent predictors for depression, where age was negatively associated. Suicidal ideation also had self-blaming, behavioural disengagement, perceived stress, active coping, denial, self-distraction, impact of COVID as signi cant independent predictors, where active coping was negatively associated. Positive reframing was in a negative relation with substance use and denial, perceived stress, substance use as coping were signi cant independent predictors for maladaptive substance use.

Discussion
The current study was conducted to compare comprehensive facets of mental health reactions and their association with age, stress and coping among adolescents and adults amid the "New Normal Pandemic era". Mental Health status across all age groups and genders revealed that the pandemic played a signi cant role in the mental health of participants. The feelings of depression and sadness [mild to severe] were the most prevalent complaints among Adolescents amid the pandemic. Moreover, a signi cantly large proportion of the sample reported suicidal ideations, which is consistent with the literature, suggesting statistically signi cant relations between pandemic stress and suicidal ideations. [22] This detrimental effect was speci cally suggested by the high percentage of adolescent population reporting an absence of the symptoms before the COVID-owing to the virus and its impact on individuals' social, occupational, and academic life led to more individuals reporting depression, anxiety, somatic concerns, obsessive thoughts and drives, suicidal ideations, sleep problems, memory problems, and personality issues.

Comparisons among adolescents and adults:
Adolescence is marked by a period of "storm and stress", social sensitivity, and a need for independence. In our study, a general decreasing trend was observed with increase in age across most of the mental health indicators i.e. Ad > Y > MA for Depression, Somatic Complaints, Suicidal Tendencies and Personality Issues, Ad &Y > MA for Anger and Anxiety, and Ad > Y&MA for Repetitive Thoughts and Actions and Dissociation, suggesting that Adolescents had more signi cant symptoms with a need for further professional assessment and treatment. This trend is the opposite of the overall prevalence indicated in the National Mental Health Survey, which estimated higher mental morbidity in middle-aged individuals (40-49 years age group). [23,24] Adolescents' depression point prevalence at 0.1-6.94% in community studies and 1.2%-21% in the clinic-based studies, [25] and a crude prevalence of depressive disorders and anxiety disorders in adults at around 3.3%. [26] During the pandemic, adolescents were particularly distressed with being socially disconnected from their friends, and illness-related worries along with educational di culties can be signi cantly associated with depression, anxiety, and life satisfaction, [27,28] making adolescents speci cally vulnerable.
While pre-pandemic suicide risk, according to the National Mental Health Survey, increased from the age group of 18-29 years to 50-59 years. [23] During the pandemic, adolescents reported more suicidal tendencies. This could be attributed to the current pandemic being a major stressful life event and related experiences (such as a fatal threat for life and health and the effect of social distancing policies), which can be a signi cant risk factor for adolescents suicidality. [29] Mid adults report lesser deterioration of the sleep quality than adolescents as the pandemic had a more signi cant impact on adolescents' social interactions and education and caused them more signi cant anxiety, which led to disturbances in sleep such as di culties initiating and maintaining sleep. [28,30] More adolescents with personality issues were found in our sample, which is explainable given that adolescence is the age for resolving Eriksonian con ict of "identity vs role confusion". This resolution involves the exploration of roles, which was restricted during the pandemic. Hence, the emotional detriment caused during the lockdown leads to lower self-concept values than before the restrictions. [31] Higher somatic complaints in adolescents can be understood as stemming from three types of concerns -the threat to life and health posed by COVID-19, daily life necessities, and e cacy of prevention and control measures. [32] These concerns may be lesser in the older population due to higher perceived self-e cacy in times of crisis. [33] In our sample, substance abuse (drinking, smoking, and other drugs) was reported by more Adults than Adolescents, which is aligned with various studies which suggest that the overall rate of alcohol drinking and smoking increased only marginally during the COVID-19 pandemic, however, the frequency of use increased substantially. Hence, substance use did not increase in the form of initiation by Adolescents but in regular drinkers and regular smokers, and as a relapse in ex-drinkers and ex-smokers. [34,35] Comparisons among males and females: Similar to previous studies, [7,9] our ndings posit that there is no signi cant gender difference in most of the mental health correlates, except substance abuse -where more Males reported a high intake of substance compared to Females. This is an interesting nding as the National Mental Health Survey delineated that Males had a higher prevalence of mental morbidity in the pre-pandemic era. [23] Therefore, the lack of signi cant difference between genders highlights the importance of the COVID-19 pandemic as a signi cant stressor. However, substance use trends are similar to the pre-Covid mental health ndings since externalizing disorders such as alcohol, drug abuse, and antisocial behavior are more predominant in men than existing cultural script. [36] Higher levels of substance use are also related to the Asian cultural value of masculinity and are, therefore, widely used as a form of coping in Asian men. [37,38] Within age groups, our study found signi cant differences in Suicidal Tendencies between Male and Female Adolescents, where more Females reported suicidal tendencies, and between Young Adult Males and Females, where more Males reported suicidal tendencies. Even pre-pandemic females had a higher risk of suicide than males due to gender discrimination, social exclusion, gender disadvantage. [23,39] However, the higher number of Young Adult Males expressing suicidal tendencies is a notable change observed, which could be attributed to the externalization of their social and individual turmoil. The stress of probable unemployment due to the uncertainty during the pandemic could be a substantial risk factor for Young Adult Males due to the social meanings attached to work and the association between hegemonic masculinity and working. Moreover, our study also found that more Adolescent Females reported sleep disturbances that could be understood as a product of catastrophic worrying that Adolescent Females are susceptible to sleep problems compared to Adolescent Males.
A disaster or crisis, such as the COVID-19 pandemic, is potentially traumatic and imposes massive collective stress. Moreover, an individual's work environment, school environment, and family environment play important roles in the pathogenesis of mental disorders. [39] Our study indicates a speci c vulnerability of the younger population while adjusting to environmental stressors and highlights the pivotal role of the pandemic for individual mental health.
Associations of Mental health facets: The correlational analysis showed age to be negatively associated with depression, anxiety, and other mental health facets, and positively associated with substance use at lower levels. In our study, a general decreasing trend was observed with increase in age across most of the mental health indicators i.e. Ad > Y > MA. Literature during the pandemic has posited poorer resilience and emotional regulation in adolescents in times of crisis. [4,5,6,8,9,27] There was moderate positive relations of mental health domains with perceived stress which is consistent with the literature that suggests that, regardless of the current level of adjustment for internal resources and social support, stress is consistently the strongest correlate of mental health. [40] The speci cally high correlation between substance abuse and stress can be explained by the tension-reduction hypothesis that suggests that exposure to stressors induces substance use as a means of mitigating experienced tension and strain. [41] All mental health domains, except substance use showed moderate-to-low correlations with 'Impact of COVID 19'. [4,5,6,8,9,27] As expected, maladaptive coping was positively related to mental health domains and showed a moderate correlation with repetitive thoughts, personality disturbances as well as suicidal ideation. It has been observed that ruminative coping acts as a signi cant maladaptive strategy and perpetuated repetitive thoughts. [42] Moreover, coping styles can mediate between a stressful situation and a negative outcome, like suicidal behavior. [43] A vital and interesting nding was the lack of negative correlation between adaptive coping and mental health domains and a signi cant positive correlation between adaptive coping and anxiety. This can be associated with the high uncertainty during the pandemic and further explains the low positive correlation between adaptive coping and anxiety. This highlights the limitations of coping styles and re-emphasizes that various factors, like the nature of pandemic as a 'stressor' and the degree of stress experienced, can in uence the relation between coping and mental health outcomes, which were particularly important during the pandemic. [44] Being a problem-focused coping strategy, planning was included in 'adaptive coping style' in this study. However, planning can be associated with more severe anxiety symptoms during COVID-19. [45] Predictors of Mental Health facets: Our results show that coping strategies, perceived stress, age, gender and impact of COVID-19 provided a signi cant predictive model for depression, anxiety, suicidal ideation and maladaptive substance use. The independent predictors provide further clarity on the role of each variable.
Perceived stress was a signi cant and strong independent predictor across the four mental health facets. This is consistent with the literature that suggests that a stressful life event is a signi cant causal factor for the onset of depression and anxiety, and a stressful environment leads to more substance use. [46] To a stressful situation, humans and animals produce a coping response, [46] hence coping becomes extremely important while predicting mental illnesses. In our study, maladaptive coping strategies signi cantly positively predicted depression, anxiety, suicidal ideation and substance use, whereas adaptive coping strategies signi cantly negatively predicted the mental health facets.
Denial was a signi cant independent predictor for all four mental health facets, while self-blaming, behavioural disengagement and self-distraction positively predicted depression, anxiety and suicidal ideation but not substance use. Denial provides a form of escapism from a stressful situation, and can therefore lead to further worsening of mental health. Substance use particularly shows denial as an underlying form of cognitive process where an individual has a poor ability to modify the way of approaching a problem or identify and use a concept. [47] Self-blame is primary feature of depression, according to A. T. Beck's cognitive theory. [48] While self-blame can stimulate active coping, it can lead to further guilt and even depression [49] as well as anxious rumination. Hence, self-blame can predict depression, anxiety and suicidal ideation.
Self-distraction and behavioural disengagement, both, provide an avoidant form of coping and are signi cant predictors of depression, anxiety and suicidal ideation. This result is in tandem with past researches [50] and can be explained by the possibility that reliance on avoidance coping could generate a broad range of stressors where cognitive avoidance could permit growth of incipient stressors, and behavioural avoidance could actively promote new stressors, [51] therefore resulting in poor mental health outcomes.
While research has found an association between coping through emotional support seeking and higher quality of life and general mood, [52] interestingly, while emotional support was a signi cant negative predictor for depression and anxiety, it was not signi cant for suicidal ideation. Suicidal ideation, therefore, poses a special speci c problem where an individual might be more sensitive to the potential rejection while seeking emotional support as is explained by the Interpersonal-Psychological Theory of suicide. [50] The negative predictive association of religion as coping with depression and anxiety is consistent with past researches which suggest that positive religious coping is associated with positive psychological adjustment. [53] This positive and adaptive involvement in religiosity can assist and equip individuals to understand the world, and allows effective appraisal and management of stressful situations, which was particularly important during a pandemic [53] . As mentioned above, the impact of COVID-19 on mental health has been severe, hence was also found a signi cant predictor for depression, anxiety and suicidal ideation.

Conclusion
The DSM-5 Cross-Cutting Measure provides the severity of symptoms, where levels mild and above are clinically signi cant for all disorders except substance use, suicidal ideation, and psychosis, where even a minimal level of severity is important for clinical decision-making through research. Based on our results, it is vital to note that around half of the Adolescents and Young Adults show clinically signi cant levels of Depression, Anger, and Somatic Complaints. 35%-45% of Adolescents and Young Adults reported anxiety, and 25%-35% reported suicidal ideations and sleep disturbances at clinically signi cant levels. A lower percentage (15%-30%) reported memory disturbances, repetitive thoughts, and dissociation. Among Adolescents and Young Adults, more severe symptoms were reported in Adolescents. Moreover, Middle-Aged Adults showed low clinically meaningful symptoms across domains (15%-25%). However, the clinical severity of substance use was more pronounced in Middle-Aged Adults compared to Adolescents and Young Adults.
Therefore, it can be observed Adolescents are more clinically vulnerable in most domains. A signi cant amount of individuals from our sample require further clinical assessment and intervention, where Middle-Aged Adults should further be seen as a clinically vulnerable population for substance use during the pandemic. Hence, this study provides a comprehensive analysis for clinical decision-making and policy development to combat the mental health problems arising and exacerbating due to the COVID-19 pandemic.

Limitations:
The cross-sectional design of the study acts as a signi cant limitation as it provided data only within a single time frame and not throughout the pandemic.
Most of the participants were residents of Uttar Pradesh, which could have affected the results from a cultural point of view. However, the large sample size, population representation, and weighting by age, sex, and region could reduce this limitation to an extent.

Conclusion:
The DSM-5 Cross-Cutting Measure provides the severity of symptoms, where levels mild and above are signi cant for all disorders except substance use, suicidal ideation, and psychosis. Even a minimal level of severity is essential for clinical decision-making through research. Based on our results, it is vital to note that around half of the Adolescents and Young Adults show signi cant levels of Depression, Anger, and Somatic Complaints. 35%-45% of Adolescents and Young Adults reported anxiety, and 25%-35% reported suicidal ideations and sleep disturbances at signi cant levels. A lower percentage of individuals (15%-30%) reported memory disturbances, repetitive thoughts, and Dissociation. Among Adolescents and Young Adults, more severe symptoms were reported in Adolescents.
Moreover, Middle-Aged Adults showed low signi cant symptoms across domains (15%-25%). However, the clinical severity of substance use was more