PDF version

M. Afifi,¹ A. Al Riyami,¹ M. Morsi¹ and H. Al Kharusil¹

ABSTRACT We used the self-reported 27-item Child Depression Inventory to investigate the rate and correlates of depressive symptoms among 5409 secondary school adolescents in Oman. Being female was a significant predictor of adolescent depression in bivariate analysis, but adjusting to other predictors in multivariate analysis, it was no longer a significant independent risk factor. History of mental illness, high score in negative health locus of control, low score in positive health locus of control, not taking breakfast, poor relationship with family members, friends and teachers and physical abuse during adolescence significantly predicted depressive symptoms in the logistic regression model. Having a hobby and never dropping a class were protective variables.

Symptômes dépressifs chez des adolescents du secondaire à Oman

RÉSUMÉ Nous avons utilisé l’inventaire de dépression chez l’enfant à 27 items auto-administré pour étudier le taux et les corrélats des symptômes dépressifs chez 5409 adolescents du secondaire à Oman. Le fait d’être de sexe féminin constituait un facteur prédictif significatif de dépression de l’adolescent dans l’analyse bivariée, mais après ajustement sur d’autres facteurs prédictifs dans l’analyse multivariée, cela ne constituait plus un facteur de risque indépendant significatif. Des antécédents de maladie mentale, un score élevé pour les croyances négatives en fonction du lieu de contrôle de la santé, un score faible pour les croyances positives en fonction du lieu de contrôle de la santé, le fait de ne pas prendre de petit déjeuner, de mauvaises relations avec les membres de la famille, les amis et les enseignants et des violences physiques pendant l’adolescence prédisaient significativement les symptômes dépressifs dans le modèle de régression logistique. Avoir un passe-temps et ne jamais avoir abandonné de matière à l’école étaient des variables constituant des facteurs de protection.

¹Department of Research and Studies, Ministry of Health, Muscat, Oman (Correspondence to Mustafa Afifi: This e-mail address is being protected from spambots. You need JavaScript enabled to view it ).
EMHJ, 2006, 12(Supplement 2): 126-137

---

Introduction

A significant body of research has shown that major depression is one of the most common psychiatric disorders of adolescence, and has also indicated the comorbidity of depression in adolescents and several health risk behaviours or mental conditions, including tobacco use [1], substance use [2], sexual activity [3], obesity [4,5], intention to use violence [6], conduct disorder, anxiety and attention deficit/hyperactivity disorder [7]. However, a 2004 study also demonstrated the comorbidity of sub-threshold depression with other mental disorders [8]. Other studies have concluded that diagnosable clinical depression exists on a continuum with sub-threshold depressive symptoms [9,10]. Lewinsohn et al. pointed that the clinical significance of depressive symptoms does not depend on crossing the major diagnostic threshold of depression [9]. They also concluded that the results of their studies suggest that clinical depression is not categorically distinct from other degrees and patterns of depressive symptoms. Their data also showed that the greater the level of depressive symptoms among adolescents, the greater the risk of developing problematic patterns of substance abuse. Lewinsohn et al. [10] stated “the fact that sub-threshold depressive symptoms predicted future major depressive disorders is consistent with findings reported in previous studies” [11,12].

Hence, the investigation of depressive symptoms among Omani adolescents is as important as studying major depression among this dynamic age group. Our aim in this study was to estimate the rate and investigate the correlates of depressive symptoms among high school adolescents in a national representative sample of 5409 students in Oman.

Methods

Sample

The sample size was determined for males and females separately using EpiInfo, assuming a prevalence rate of 1% for depressive disorders among adolescents and a confidence interval of 99%. Accordingly, 2885 male adolescents and 2731 female adolescents, 5616 students in total were randomly selected (about 5% of the 111 849 secondary school students for the academic year 2004); these covered 43 wilayat (districts) (73% of the 59 wilayat in Oman). Of these, 5409 (2739 boys and 2670 girls) were screened and entered in statistical analysis (response rate 93% overall, 95% for boys, 97% for girls). The 207 missing students were absent at the time of screening, had dropped classes or had transferred to another school.

We used a multi-stage, stratified random sampling technique. All the regions of the country were selected at the first stage and sample size for each sex was determined in proportion to the total number of adolescent students aged 14–20 years. Then, in each region, male and female sub-samples were selected in proportion to the secondary school adolescent population in each of the 5 grades (I; II and III Arts and II and III Science) in Omani secondary schools. In the next stage, ≥ 1 schools were chosen randomly from each region, and then from each school ≥ 1 classes were randomly selected to cover the number of adolescents needed for the sample.

Questionnaire and measurements

The total sample completed the 27-item Child Depression Inventory and a questionnaire including demographic data, health locus of control and other factors associated with depression. The self-reported questionnaire covered sociodemographic data, chronic physical illness (e.g. diabetes, cardiovascular disease) and mental disorders (e.g. depression, schizophrenia) diagnosed by a doctor. Relationships with father, mother, siblings, friends and schoolteachers were assessed on a 5-item scale with a score ranging from 1–5 where 5 represents an excellent relationship. It took around 40–50 minutes for each student to complete the whole questionnaire (around 1 class session).

The questionnaire also included a group of questions adapted from Takakura and Sakihara on health practices such as sleeping habits and hours of sleep, eating breakfast regularly, smoking and physical activities [13]. Each of the following was considered a positive health practice: sleeping 7–8 hours/night; eating breakfast every day; not currently smoking; and doing physical activity outside school premises ≥ 3 times per week.

The students were also asked about their health locus of control beliefs, whether internal or chance external, adapted from the multi-dimensional Health Locus of Control form A [14]. The first 2 subscales: internal health locus of control and chance health locus of control were translated into Arabic by the author who also calculated Cronbach’s α to assess the internal consistency as reliability in a previous study. Both subscales showed an accepted level of internal consistency i.e. > 0.4 (0.5111 and 0.4636 for the 2 subscales respectively). Each subscale contained 6 questions and for each question study participants chose 1 of 6 answers ranging from strongly agree (= 6) to strongly disagree (= 1). The scores for each subscale ranged from 6–36 [14].

The Arabic translation of the 27-item Children’s Depression Inventory [15] has been used in previous studies [16,17]. Each of the 27 items of the inventory assesses 1 symptom by presenting 3 choices ranging from 0–2 in the direction of increasing psychopathology and total score ranges from 0–54. Those who scored ≥ 20 were considered as having at least mild depressive symptoms. The cut-off score of 20 is suggested for screening in a general population, such as in a school setting, in which the prevalence of depression is likely to be low [18]. Reliability of Cronbach’s α was 0.8.

Training and piloting

Three days training was tailored to the school health doctors who would undertake the screening process. In addition, a medical officer (regional coordinator) was recruited in each health region of the country to manage the administration and logistics of the survey and to ensure the implementation of all phases of the survey according to plan. During training a pilot study was conducted in Muscat by school health doctors on 400 secondary school students from both sexes in 2 randomly selected secondary schools (1 boys and 1 girls) not included in the study sample.

Data processing and statistical analysis

Data coding, entry and management were done using EpiInfo followed by data analysis using SPSS for Windows, version 9. Data are given as counts, percentages and means. Group means were compared using analysis of variance and chi-squared test was used to examine the distribution of data using the likelihood ratio χ2. After doing bivariate analysis between the dependent variable and its predictors, all variables were then entered in a logistic regression model to determine the most significant variables adjusted for the others. The dependent dichotomous variable was coded to 0 = normal and 1 = having depressive symptoms (Child Depression Inventory score ≥ 20). The odds ratio which shows the change in the odds of dependent variable(s) when the independent variable changed from 0 to 1 in the case of binary variables, or the next category or score in the case of categorical or continuous variables adjusted for age, sex and all other variables in the model. P-value ≤ 0.05 was considered significant in all statistical tests.

Ethical issues

The study was approved by the Ethical Committee in the Ministry of Health before commencing the training and fieldwork. Confidentiality for the study participants was maintained as no direct or indirect identification was used. Verbal consent was obtained from the adolescent participants as well as the school headmasters.

Results

The age of the study group ranged between 14 and 20 years with about 75% aged 16–18 years (Table 1). Less than 8% reported having had a chronic physical illness (diagnosed by a doctor) and 2.3% reported having had a mental disorder (diagnosed by a doctor). The majority had parents whose education was below secondary level (84.1% for fathers and 92.2% for mothers). The mean [standard deviation (SD)] for birth order was 4.87 (SD 2.9). Around 25% of the sample was subjected to physical abuse during adolescence by their parents or those who raised them and 7.5% the sample were also physically abused during their childhood (Table 1).

The mean (SD) score for the Child Depression Inventory was 13.16 (6.69).

In bivariate analysis, all 25 independent variables were significantly associated with the dependant variable, having depressive symptoms, except for 3 variables: age, birth order and mother’s education. Overall, 17.0% of our sample had depressive symptoms, 19.4% of girls and 14.7% of boys (χ2 21.58; P < 0.01). Those with a personal history of mental illness were more likely to have depressive symptoms (39.2%) than those who did not (16.5%) (χ2 = 35.56; P < 0.01). Similarly, those with a personal history of chronic physical illness were more likely to have depressive symptoms (27.0%) than those with no history (16.2%).

Having a hobby, taking breakfast regularly, not smoking, enough hours night sleeping, doing physical exercises ≥ 3 times/week and having a high score in relationships with social contacts protected against having depressive symptoms (Tables 2 and 3).

All variables, including the 3 non-significant variables, were entered into a forward stepwise logistic regression model. This was done because of the importance of adjusting for age and because education of parents is a proxy for social class. Birth order was also introduced in the model due to its demonstrated importance as a predictor of adolescent mental health [19]. Only 13 variables were significantly found to predict having depressive symptoms in the multivariate analysis. These could be grouped into a protective variables group and a risky variables group. The protective variables group comprised those with healthy practices: taking breakfast daily; sleeping 7–8 hours a night; having a hobby; and having good relationships with social contacts. The risky variables group comprised those who scored high for external health locus of control, those who scored low for internal health locus of control, those with a positive personal history of mental illness, those who ever dropped a class, and those who were abused by their parents during adolescence (Table 4).

Discussion

The current study spotlights the rates and correlates of adolescent depressive symptoms in Oman; 17.0% of the adolescents we surveyed had depressive symptoms indicating that depressive symptoms is a public health problem among Omani adolescents.

Fergusson et al. investigated the association between extent of depression (asymptomatic, sub-threshold, and major depression) and rates of subsequent mental health problems [20]. They found that adolescents with sub-threshold depression had elevated risk of later depression and suicidal behaviours. Depressive symptoms range from none to severe, and those meeting diagnostic criteria for major depression represent the extreme of a continuum rather than a distinct group of individuals suffering from a specific disorder [20]. Lewinsohn et al. argued whether to treat persistent sub-threshold depression or not by mentioning the possible iatrogenic risks such as stigmatization and financial hardship. They also mentioned the “stepped care” model in which clinicians should strive to provide the least aggressive and least costly treatment to resolve clinical symptoms and any underlying vulnerability to depression [8]. Investigating the factors associated with depressive symptoms would help in early management of adolescents with such symptoms and would reduce future risk of undesirable mental health outcomes.

The rate of depressive symptoms in adolescents in this study was not comparable with rates in previous studies in Oman [16] or other Arab countries. Shaaban and Baashar [21] in Sudan used the Beck Depression Inventory (BDI) to screen for depressive symptoms in 1107 girls aged 12–19 years; 11% reported severe depression. Daradkeh et al. used the Composite International Diagnostic Interview on 1394 participants from Al Ain in the United Arab Emirates: the lifetime prevalence of major depression was significantly higher among females than males, 10.3% and 2.8% respectively [22]. Using the 13-item Beck Depression Inventory for screening, a similar depression rate was found in a Finnish study: 17.2% of 16 464 adolescents aged 14–16 years had at least mild depression [23].

Some studies have shown that frequency of depression in children and adolescents increased with age [24]. Our study did not show the same results. Female preponderance has been found in a previous study as well as in our study [23], although in multivariate analysis we did not find being female was a significant predictor. There could be a number of explanations for this; it could be related to the weak (or lack of) association of sex with depressive symptoms, as indicated in a previous study in Oman [25]. It could also be attributed to the significant independent variables in the model which explain or mediate sex differences in adolescent depression. Depression among adolescents results in impaired cognitive, interpersonal and academic functioning [26] and that could explain the significant association with both poor relations with the teacher and ever dropping a class with having depressive symptoms in this study.

There was a strong positive association between current smoking and having depressive symptoms in our study as well as other studies. Depressed adolescents are more likely to begin smoking, to smoke more and to continue smoking as young adults. Smokers with mild or major depression find it hard to quit smoking [27,28].

The association between health locus of control and depressive symptoms was significant. Adolescents who were oriented toward chance locus of control were more likely to report having depressive symptoms in the bivariate and multivariate analysis. The same has been found in other studies. A high score for external locus significantly increased the risk for behavioural problems in general [29,30].

Adolescents with past history of physical abuse during childhood have also been found to be more likely to score higher on external chance locus of control [30]. Both externality and child abuse are risk factors for depression. Psychopathology coexisting with history of physical abuse in adolescents has been encountered in a number of studies. Abused adolescents showed significantly higher prevalence rate of depression and conduct disorder [31]. It has also been shown that they have significantly greater exposure to risk factors for adolescent suicide, including family disintegration, diagnosis of depression, disruptive behaviour disorders and substance abuse and dependence [32].

The negative association of depressive symptoms with healthy practices and social support, evidenced by good relationships with social contacts in our study, has also been demonstrated [13]. It seems that social support and healthy practices could have a buffering effect on depressive symptoms.

Finally, we would like to mention the study limitations. The dearth of research on adolescent depression in the Arab world published in PubMed-indexed journals and the diversity of tools used to screen depressive symptoms and its correlates limited the chances for comparison of results. In addition, the use of self-reported questionnaires might elicit inflated or false responses, especially in such a sensitive age group. Self-reported questionnaire are easy to apply, however, and it is difficult logistically to conduct structured interviews for such a big sample. Another limitation was the cross-sectional design of our study where causality and/or temporal association could not be established and hence we could not determine why adolescents present with sub-threshold depressive symptoms.

Another limitation was the difficulty in demonstrating how representative this student sample in Oman was to Omani adolescents in general. Although education is universal in the country the possibility that some adolescents with depressive symptoms may have already dropped out of school should be always be taken into consideration. However, to the best of our knowledge, this was the first large, national study to investigate adolescent depressive symptoms and their correlates in Oman.

Conclusion

To conclude, depressive symptoms in children can be subtle and varied. Diagnosis in adolescents is often complicated by behavioural manifestations associated with hormonal changes, but evidence suggests secondary prevention can significantly reduce future psychosocial problems [33].

We recommend using the findings of the current study by taking into consideration the protective factors as well as the risk factors of adolescent depression in a future prevention programme, along with strengthening the mental health component of the school health programme.

Acknowledgements

We would like to thank the Ministry of Health and the World Health Organization Regional Office for the Eastern Mediterranean for their financial support in conducting this study. We would also like to thank the Minister of Health, the undersecretaries of planning, health affairs and financial affairs, the directors-general of health services in the regions and a great number of other people who were involved in data collection within the Ministry of Health for their support.

References

1. Goodman E, Capitman J. Depressive symptoms and cigarette smoking among teens. Paediatrics, 2000, 106(4):748–55.
2. Christie KA et al. Epidemiologic evidence for early onset of mental disorders and higher risk of drug abuse in young adults. American journal of psychiatry, 1988, 145(8):971–5.
3. DiClemente RJ et al. A prospective study of psychological distress and sexual behavior among black adolescent females. Pediatrics, 2001, 108(5):E85.
4. Pine DS et al. The association between childhood depression and adulthood body mass index. Pediatrics, 2001, 107(5):1049–56.
5. Goodman E, Whitaker RC. A prospective study on the role of depression on the development and persistence of adolescent obesity. Pediatrics, 2002, 110(3):497–504.
6. Durant RH, Treiber F, Goodman E, Woods ER. Intentions to use violence among young adolescents. Pediatrics, 1996, 98(6 pt 1):1104–8.
7. Angold A, Costello EG. Depressive comorbidity in children and adolescents: empirical, theoretical and methodological issues. American journal of psychiatry, 1993, 150(12):1779–91.
8. Lewinsohn PM et al. The prevalence and co-morbidity of subthreshold psychiatric conditions. Psychological medicine, 2004, 34(4):613–22.
9. Lewinsohn PM et al. Clinical implications of “subthreshold” depressive symptoms. Journal of abnormal psychology, 2000, 109(2):345–51.
10. Lewinsohn PM et al. Family study of subthreshold depressive symptoms: risk factor for MDD? Journal of affective disorders, 2003, 77(2):149–57.
11. Angst J, Merikangas K. The depressive spectrum: diagnostic classification and course. Journal of affective disorders, 1997, 45(1–2):31–40.
12. Angst J, Merikangas KR, Preisig M. Subthreshold syndromes of depression and anxiety in the community. Journal of clinical psychiatry, 1997:58(suppl. 8):6–10.
13. Takakura M, Sakihara S. Psychological correlates of depressive symptoms among Japanese high school students. Journal of adolescent health, 2001, 28(1):82–9.
14. Wallston KA, Wallston BS. Development of the Multidimensional Health Locus of Control (MHLC) scales. Health education monographs, 1978, 6(2):160–70.
15. Kovacs M. Rating scales to assess depression in school-aged children. Acta paedopsychiatrica, 1981, 46(5–6):305–15.
16. Afifi M. Study of school adolescent depression in the South Sharqiya region, Oman. Journal of the Bahrain Medical Society, 2000, 12(1):27–30.
17. Afifi M. Adolescent use of health services in Alexandria, Egypt: association with mental health problems. Eastern Mediterranean health journal, 2003, 10(1–2):64–71.
18. Mattey S, Petrovski P. The Children’s Depression Inventory: error in cutoff scores for screening purposes. Psychological assessment, 2000. 14(2):146–9.
19. Hirshfeld-Becker et al. Lack of association between behavioral inhibition and psychosocial adversity factors in children at risk for anxiety disorders. American journal of psychiatry, 2004, 161(3):547–55.
20. Fergusson DM et al. Subthreshold depression in adolescence and mental health outcomes in adulthood. Archives of general psychiatry, 2005, 62(1):66–72.
21. Shaaban KM, Baashar TA. A community study of depression in adolescent girls: prevalence and its relation to age. Medical principles and practice, 2003, 12(4):256–9.
22. Daradkeh TK, Ghubash R, Abou-Saleh MT. Al Ain community survey of psychiatric morbidity II. Sex differences in the prevalence of depressive disorders. Journal of affective disorders, 2002, 72(2):167–76.
23. Kaltiala-Heino R et al. Adolescent depression: the role of discontinuities in life course and social support. Journal of affective disorders, 2001, 64(2–3):155–66.
24. Pataki CS, Carlson GA. Childhood and adolescent depression: a review. Harvard review of psychiatry, 1995, 3(3):140–51.
25. Afifi M. Depression in adolescents: gender differences in Oman and Egypt. Eastern Mediterranean health journal, 2006, 12(1–2):61–71.
26. Garland EJ. Adolescent depression. Part 1. Diagnosis. Canadian family physician, 1994, 40:1583–7.
27. Glass RM. Blue mood, blackened lungs. Depression and smoking. Journal of the American Medical Association, 1990, 264(12):1583–4.
28. Hughes JR et al. Prevalence of smoking among psychiatric out-patients. American journal of psychiatry, 1986, 143(8):993–7.
29. Liu X et al. Life events, locus of control, and behavioral problems among Chinese adolescents. Journal of clinical psychology, 2000, 56(12):1565–77.
30. Afifi M. Health locus of control and depressive symptoms among adolescents in Alexandria, Egypt. Eastern Mediterranean health journal, (in press).
31. Pelcovitz D et al. Post-traumatic stress disorder in physically abused adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 1994, 33(3):305–12.
32. Kaplan SJ et al. Adolescent physical abuse and suicide attempts. Journal of the American Academy of Child and Adolescent Psychiatry, 1997, 36(6):799–808.
33. Lamarine RJ. Child and adolescent depression. Journal of school health, 1995, 65(9):390–3.