Anxiety symptoms in regular school students in Mumbai City, IndiaS Karande1, NJ Gogtay2, N Bala1, H Sant1, A Thakkar1, R Sholapurwala1
1 Department of Pediatrics, Seth G.S. Medical College and K.E.M. Hospital, Mumbai, Maharashtra, India
2 Clinical Pharmacology, Seth G.S. Medical College and K.E.M. Hospital, Mumbai, Maharashtra, India
Correspondence Address: Source of Support: None, Conflict of Interest: None DOI: 10.4103/jpgm.JPGM_445_17
Source of Support: None, Conflict of Interest: None
Keywords: Adolescents, children, questionnaire.based study, stress disorders, variables
Anxiety disorders are one of the most common psychological disorders in school students world-wide., Recent community studies among school students in the UK and US indicate a period prevalence rate of them having at least one anxiety disorder between 9 and 32% and 15 and 20%,, respectively. Anxiety disorders have been found to negatively affect students in terms of academic, social, and personal development.,,,,,
A PubMed search using the medical subject headings (MeSH) words “anxiety,” “students,” and “India” did not reveal any study evaluating anxiety in school students in Mumbai city. Hence, we conducted the present cross-sectional questionnaire-based study with the primary objective of evaluating the prevalence of anxiety symptoms in school students in our city. A secondary objective was to assess the impact of variables on anxiety symptoms.
The present study was approved by the institutional ethics committee. The study protocol is registered with the clinical trials registry of India (CTRI/2017/08/009512). An accompanying parent or legal guardian signed an informed consent form permitting the participation of his/her offspring. Additionally, all school students signed an assent form prior to enrolment. Confidentiality was maintained using unique identifiers.
Design, setting, and sample size
A questionnaire-based (pre-validated) cross-sectional study was conducted from January to August 2013. A total of 493 students in the age group of 8–15 years were recruited (nonprobability sampling) from four English medium private schools which were in the vicinity of our institute. Within each school every student who met the selection criteria and gave assent was included. Three schools' curriculums were under the State Government Board [Secondary School Certificate (SSC)], while one was under the National Board [Indian Certificate of Secondary Education (ICSE)]. Three of the schools were co-educational and one was an only girls' school.
“Regular” students of either gender, between 8 and 15 years of age, studying in class standards III to X, conversant in English, and with consistently “average” academic performance (obtaining 60-90% marks or Grade B) for the past 2 years were included in the study. Exclusion criteria were students with chronic health conditions (such as epilepsy, asthma, allergic rhinitis, and obesity), students with symptoms of inattentiveness, hyperactivity or disruptive behavior in the classroom, those with poor academic performance (obtaining less than 60% marks or Grade C and below), and those excelling in studies (achieving more than 90% marks or Grade A and above).
Anxiety symptoms were measured using the Spence Children's Anxiety Scale (SCAS)-child self-report version questionnaire. All students were explained how to complete the questionnaire by a Special Educator (RS), following which they individually completed it in a quiet room in their respective schools. At least two trained research assistants (NB, HS, or AT) were always present to ensure independent responding and to provide clarification to any student, if necessary. Each student was given enough time to complete the questionnaire. Data related to five variables: age, gender, absence or presence of sibling(s), the type of school board curriculum, and type of school attended were noted using a supplementary questionnaire.
Measuring anxiety symptoms
The SCAS is a self-report questionnaire designed to measure total (or “overall”) anxiety symptoms in school students. This questionnaire also assesses their perceptions of the frequency with which they experience symptoms relating to six anxiety disorders (“subscales”), namely: (i) separation anxiety, (ii) social anxiety, (iii) obsessions/compulsions, (iv) panic/agoraphobia, (v) fears of physical injury, and (vi) generalized anxiety, and is based on the DSM-IV diagnostic categories of anxiety disorders [Table 1]. Over the last two decades, this questionnaire has been used worldwide to measure anxiety symptoms in school students.,,,,,, The SCAS has been shown to have good test-retest reliability and internal consistency.,,,
The SCAS is a Likert-type scale and consists of 44 items. Of these, 38 items relate to specific symptoms of anxiety and are used to measure the six subscale scores. The remaining six items are positive filler items that have been included to reduce negative response bias. The response to each item can be one of the following: never, sometimes, often, or always. Each item is scored as 0–3, and therefore, the maximum possible SCAS total score is 114. Each item question is simple and easy to understand. However, as recommended by Spence, a student was permitted to take help of one of the trained research assistants to read out the question and explain before marking his/her response. The SCAS scores (total and individual subscales) of the students were calculated as recommended.
Data analyses were done using the Statistical Package for Social Sciences, version 17.0 for Windows (Chicago, Illinois, US). First, to investigate the reliability of the SCAS, internal consistencies (as measured by Cronbach's alpha) were calculated for the total SCAS score and for each of the SCAS subscale scores. Second, the correlation coefficients (as measured by Pearson's r) between the SCAS total and subscale scores of the study students were computed. Third, the SCAS scores (total and individual subscales) were converted into T-scores having a mean of 50 and a standard deviation of 10. A cut-off T-score of 10 above the mean of 50, i.e., above 60 on any dimension represents a value of 1 standard deviation above the mean and is indicative of elevated levels of anxiety and justifies further investigation to confirm diagnostic status using a clinical interview. T-scores also take into account younger and older age groups [8–11-year-olds and 12–15-year-olds, respectively] and gender of the student. This method also considers that high anxiety status may be reflected in both the total score and/or an elevated subscale score.
In the present study, participants were divided into two groups, namely, (i) those having anxiety symptoms (T-score > 60) and (ii) those having no anxiety symptoms (T-score ≤ 60). Using the Fisher's exact test, subgroup analyses were carried out to find out whether the odds of having overall anxiety/anxiety disorder(s) symptoms as per total/individual subscale(s) scores were influenced by any of the five variables. Wherever appropriate, the odds ratio (OR) [with 95% confidence interval (CI)] were to be calculated. P values of < 0.05 (two-tailed) were considered statistically significant.
Characteristics of enrolled students
Four hundred ninety-three students were enrolled, of which 200 (40.6%) were boys and 293 (59.4%) were girls. In the younger age group, there were 271 (55.0%) students, of which 116 were boys and 155 were girls. In the older age group, there were 222 (45.0%) students, of which 84 were boys and 138 were girls. The boys to girls' ratios were 0.7:1 and 0.6:1 in these two groups, respectively. No parent/guardian declined consent and no student declined assent for participation. Time taken by the students to fill the SCAS questionnaire ranged from 30 to 45 minutes. There were no missing data for the SCAS items.
Reliability of SCAS scores (total and individual subscales)
Testing for reliability (“internal consistency”) involves estimating how consistently individuals respond to the items within a scale. In the current study sample, internal consistency for the total SCAS score was high (i.e., alpha = 0.83); and for four subscale scores was moderate to high (ranging from separation anxiety, social anxiety, panic/agoraphobia, alpha = 0.60 each; generalized anxiety, alpha = 0.54 and obsessions/compulsions, alpha = 0.53). The fears of physical injury subscale had a lower internal consistency (alpha = 0.45).
Correlations between SCAS total and subscales' scores
There was a highly strong relationship (Pearson's r = 0.5–1.0) between the total SCAS score and each of the SCAS subscales' scores (ranging from separation anxiety, r = 0.65; social anxiety, r = 0.67; obsessions/compulsions, r = 0.73; panic/agoraphobia, r = 0.69; fears of physical injury, r = 0.56; and generalized anxiety, r = 0.71), indicating a good convergent validity for all these constructs.
Overall anxiety symptoms in study participants
Fifty-three (10.8%) of the 493 participants had overall anxiety symptoms based on their total SCAS T-scores being >60 [Table 2].
Anxiety disorder(s) symptoms in study participants
Symptoms of all six anxiety disorder types were detected, with the highest number of students (146, 29.6%) having obsessions/compulsions, followed by fears of physical injury in 134 (27.2%), separation anxiety in 59 (12.0%), generalized anxiety in 47 (9.5%), panic/agoraphobia in 39 (7.9%), and social anxiety in 38 (7.7%) [Table 3].
Impact of variables on overall anxiety symptoms in study participants
Older students had greater odds of having overall anxiety symptoms [40/222 (18.0%) vs. 13/271 (4.8%), crude OR = 4.36, 95% CI 2.27 to 8.39, P < 0.0001].
Impact of variables on anxiety disorder symptoms in study participants
Boys and students studying in a SSC board curriculum school had greater odds of having symptoms of separation anxiety [31/200 (15.5%) vs. 28/293 (9.6%), crude OR = 1.74, 95% CI 1.01 to 3.00, P= 0.049; and 56/413 (13.6%) vs. 3/80 (3.8%), crude OR = 4.03, 95% CI 1.23 to 13.20, P= 0.013, respectively].
Older students and boys had greater odds of having symptoms of social anxiety [26/222 (11.7%) vs. 12/271 (4.43%), crude OR = 2.86, 95% CI 1.41 to 5.82, P= 0.003; and 26/200 (13.0%) vs. 12/293 (4.10%), crude OR = 3.50, 95% CI 1.72 to 7.11, P < 0.001, respectively].
Older students and boys had greater odds of having obsessions/compulsions [87/222 (39.2%) vs. 59/271 (21.8%), crude OR = 2.32, 95% CI 1.56 to 3.44, P < 0.0001; and 70/200 (35.0%) vs. 76/293 (25.9%), crude OR = 1.54, 95% CI 1.04 to 2.27, P= 0.035, respectively].
Boys had greater odds of having panic/agoraphobia [22/200 (11.0%) vs. 17/293 (5.8%), crude OR = 2.01, 95% CI 1.04 to 3.88, P= 0.042].
Fears of physical injury
Students who had sibling(s) had greater odds of having fears of physical injury [107/343 (31.2%) vs. 27/150 (18.0%), crude OR = 0.48, 95% CI 0.30 to 0.78, P= 0.003].
Older students had greater odds of having symptoms of generalized anxiety [33/222 (14.9%) vs. 14/271 (5.17%), crude OR = 3.21, 95% CI 1.67 to 6.16, P < 0.001].
It is important to identify anxiety disorders in school students as these predict not just adult anxiety disorders but also other psychopathologies such as depression, substance use problems and suicide attempts.,, The present study has documented that, in the city of Mumbai, approximately 11% (53/493) of regular school students studying in class standards III to X had symptoms of overall anxiety. Bakhla et al. also utilized the SCAS questionnaire and documented that 11% (16/146) students studying in class standard VIII in Jamshedpur (India) had symptoms of overall anxiety. In the present study, as in the study by Bakhla et al. school students had symptoms of all six anxiety disorders with obsessions/compulsions, fears of physical injury and separation anxiety being the most common disorders in both studies; namely 29.6%, 27.2%, and 12.0% in the present study; and in 17.9%, 8.2%, and 9.6% in the other Indian study, respectively. Utilizing another questionnaire, namely the State-Trait Anxiety Inventory (STAI), Deb and Walsh  documented that 19% (88/460) of students studying in class standards VIII to XII in Kolkata (India) had high anxiety. Utilizing the SCAS, studies done in other countries, namely, Australia, the Netherlands, South Africa, Japan,, Germany,, Cyprus, the UK, Sweden, Italy, and China  also documented that school students have symptoms of anxiety.,,,,,, However, none of these studies ,,,,,, reported any prevalence data.
Studies done in Australia, the Netherlands, and Japan  have reported that younger students are more prone to develop symptoms of overall anxiety. In contrast in the present study, a significantly higher proportion of older students had overall anxiety symptoms, which has been reported earlier only in Chinese students.
For reasons that still remain unclear overall anxiety usually occur more frequently among girls than among boys., Studies done in Jamshedpur (India), Australia, the Netherlands, Germany, Japan,, and China  have reported that girls were significantly more prone to develop symptoms of overall anxiety. In contrast in the present study, gender was not a risk factor for developing overall anxiety symptoms. Bhasin et al. also found no effect of gender on developing anxiety symptoms in their study which utilized the Depression, Anxiety and Stress Scale (DASS) questionnaire in assessing 242 affluent adolescent school students in Delhi. In contrast, Deb and Walsh  reported that boys in Kolkata city were significantly more prone to develop anxiety symptoms.
Significantly higher proportions of older Chinese students  have been reported to have significantly higher symptoms of all six anxiety disorders. In the present study, older students had significantly higher symptoms of three out of six anxiety disorders; namely, social anxiety, obsessions/compulsions, and generalized anxiety. Older Australian, German, and Japanese  students have been reported to have significantly higher symptoms of only one disorder; namely, social anxiety. In contrast, younger Dutch students  have been reported to have significantly higher symptoms of all six anxiety disorders; Australian students  of five disorders (all except social anxiety); Japanese students  of three disorders (separation anxiety, obsessions/compulsions, and panic/agoraphobia); and German  and Japanese  students of two disorders (separation anxiety and panic/agoraphobia).
All anxiety disorders also usually occur more frequently in girls., Dutch, German, and Japanese , girls have been reported to have significantly higher symptoms of all six anxiety disorders; and Australian, Chinese, and Indian  girls of five disorders (all except obsessions/compulsions). In the present study, boys had significantly higher symptoms of obsessions/compulsions, which has been reported earlier in Australian, Chinese, and Indian  boys. To our knowledge, boys having significantly higher symptoms of separation anxiety, social anxiety and panic/agoraphobia has not been reported earlier.
What is the importance of the present study?First, anxiety disorders usually remain undiagnosed in school students owing to the internalized nature of its symptoms. The results of the present study suggest that there is an urgent need to start screening school students, especially older students, in our city for anxiety disorders. Anxiety can have a ruinous effect on academic performance. Further, untreated anxiety over a period of time can contribute to school refusing behavior and eventually to early school leaving. We are unable to explain why a significantly higher proportion of older students in our city were having symptoms of overall anxiety, social anxiety, obsessions/compulsions, and generalized anxiety. This finding could be related to their experiencing increasing parental expectations for good school grades, school curriculum getting tougher, and unusually fast pace of life in this mega-city, resulting in inadequate social mingling. Second, the present study's data will help guide the treating counselor/psychiatrist in alleviating symptoms in school students. There is now convincing evidence that with cognitive behavior therapy, a majority of school students with anxiety disorders show favorable outcomes. Rarely, pharmacological treatments such as selective serotonin reuptake inhibitors may be needed. Moreover, anxiety disorder(s) in adolescents can be a comorbid condition with prodromal psychosis., Effectively managing the anxiety disorder(s) along with implementing simple lifestyle changes such as stress reduction and sleep hygiene may help delay onset of psychosis or lessen its symptoms., Third, the present study has identified two new inexplicable findings, namely students having greater odds of having symptoms of: (i) fears of physical injury if they had sibling(s); and (ii) separation anxiety if they were studying in a SSC board curriculum school.
The strengths of the present study include use of a validated pediatric instrument, namely, the SCAS questionnaire to measure anxiety symptoms and high participation and response rates. Also, the reliability for the total SCAS score and for five subscale scores (all except fears of physical injury subscale) was sufficient. The convergent validity for all the constructs of the SCAS questionnaire was good.
The present study has its limitations. First, reliability was assessed only by means of internal consistency. Other aspects of reliability such as test-retest stability and inter-rater reliability were not examined. Second, the study relied solely on students self-report. Parent or teacher ratings of students' anxiety symptoms might have provided important additional information. Also no behavioral observations or clinical indices were used to confirm these self-report measures and this could have biased some of our results. Future researchers should simultaneously obtain data from both students and parents (utilizing SCAS-parent-report version) to address this shortcoming. Third, although we used a validated instrument, further assessment by a psychiatrist would have helped in confirming the diagnosis of overall anxiety/anxiety disorder(s) in students having T-scores above 60; but this was beyond the scope of the present study. It is possible that some of the students identified with high scores may not have had clinically significant overall anxiety/anxiety disorder(s) on psychiatric evaluation. Fourth, certain variables such as socioeconomic status, parenting style, neighborhood environment, social support, and type of social support which may influence development of anxiety symptoms were not investigated. Because parents were not interviewed we could not collect data on these variables. Fifth, because non-English-speaking students were excluded from the study there may be a potential language bias in our findings. Although Hindi, Bengali, and Urdu versions of the SCAS questionnaire were available, they could not be used for lack of availability of these language-medium schools in the vicinity of our institution. Although Marathi-medium schools were accessible, the Marathi version of the SCAS questionnaire was not available. Sixth, the nonprobability sampling of the present study may have led to a recruitment bias in our findings. However, we do not believe that these limitations adversely affected the utility of our results. Due to the limitations outlined above and the general paucity of data on anxiety symptoms in Indian students, future researchers should investigate whether the present study's results can be generalized to the population level.
There is an urgent need to start screening school students in our city for anxiety disorders and the SCAS questionnaire (including its Hindi, Bengali, and Urdu versions) can help in this process. Early diagnosis of anxiety symptoms would help to optimize management of students and may lead to favorable long-term academic and social outcomes.
We thank Professor Susan H. Spence, Department of Psychology, University of Queensland, Brisbane, Australia, for permitting us to use the SCAS manual free of cost and granting us permission to use the SCAS instrument; Dr. D.P. Singh, Department of Science Methodology, Tata Institute of Social Sciences, Mumbai, Maharashtra, India for his help in the statistical analysis of the data; all the students who participated in the present study, and their parents/legal guardians and school authorities.
Financial support and sponsorship
The Learning Disability Clinic at our institute is partially funded by a research grant from Tata Interactive Systems, Mumbai, Maharashtra, India.
Conflict of interest
Sunil Karande is the Editor of the Journal of Postgraduate Medicine.
[Table 1], [Table 2], [Table 3]