[Study of the CES-D on a sample of 1,953 adolescent students]

To evaluate the psychometric characteristics of the CES-D (Center for Epidemiological Studies - Depression Scale) in adolescents and to estimate the prevalence of major depression in French high-school students.

A random sample of 2 583 high-school students from the departments of Haute-Garonne and Pyrénées-Orientales, France, were asked to complete the CES-D. Of these subjects, 1 953 (75.6%) (966 boys and 987 girls, mean age=17 1.4) completed satisfactorily the questionnaire. A sub-sample of 60 adolescents volunteered for an interview and were assessed using the major depressive episode module of the MINI (Mini International Neuropsychiatric Interview).

The internal consistency of the CES-D was satisfactory (Cronbach alpha=0.85). A principal component analysis produced consistent sub-scales. The factor structure was similar to that observed in previous studies in adults. Four factors were extracted which accounted for 49.6% of the variance: a depressive cognition factor (items 1, 9, 10, 15, 19); a positive affect factor (items 4, 8, 12, 16); a factor reflecting slowing of thought and reduction of energy (items 5, 7, 20); and a depressive affect factor (items 2, 3, 6, 11, 14, 17, 18). Among the 1 953 subjects, the mean score for girls was significantly higher than for boys (18,8 10,2 versus 14,4 7,7). Of the 60 subjects who were assessed with the MINI, 28 received the diagnosis of major depressive episode. The mean scores of depressed girls and boys in this sub-sample (respectively 35.5 8.7 and 27.7 9.6) were significantly higher than the mean scores of girls and boys for the total sample. According to the results, the best cut-off score was a score of 24 or higher, providing a sensibility of 0.74 and a specificity of 0.73. According to this cut-off score, 9.9% of boys and 24.2% of girls had a probable major depressive episode.

This study showed a high level of depressive symptoms and a high frequency of probable depression in French adolescents. Two findings suggest the reliability of these results: first, the male-female ratio, close to 1:2 for probable depression, is in keeping with current epidemiological data; second, the high internal consistency and the factor structure are indicators of the reliability of CES-D in adolescents. Our study found a higher cut-off score than the previous French study of the CES-D in adolescents by Bailly et al. This discrepancy may be due to our small interview sample. However, the cut-off of 24, identified with our study, provided frequency rates more in agreement with epidemiological data than the cut-off of 21 proposed by Bailly et al. Mean CES-D scores of boys and girls in our study were respectively significantly higher than those obtained in boys (n=439, 12.4 7.5) and girls (n=305, 16.6 9.7) by Bailly et al. eleven years before. Given the similarities of mean age and representativeness of samples in both studies, this discrepancy may be due to an increase in depressive symptoms in French adolescents which may be linked to the parallel increase in substance use and particularly in cannabis use.

The high internal consistency and the factor structure of the CES-D suggest its reliability in adolescents. This study found a high frequency of probable major depression in French high-school students.