Circadian type and mood seasonality in adolescents

Authors


Lorenzo Tonetti, MS, PsyD, Department of Psychology, University of Bologna, Viale Berti Pichat 5, 40127 Bologna, Italy. Email: lorenzo.tonetti2@unibo.it

Abstract

The aim of the present study was to investigate, for the first time, the relationship between circadian preference and mood seasonality in adolescents. To this end, 1539 participants (881 female; 658 male) completed the Morningness–Eveningness Questionnaire for Children and Adolescents and the Seasonal Pattern Assessment Questionnaire for Children and Adolescents to determine circadian preference and mood seasonality, respectively. Evening types achieved significantly higher mood seasonality scores than intermediate and morning types, as intermediates did than morning types. Agreeing with previous studies on young adults and adults, the present data indicate a significant relationship between eveningness and higher mood seasonality in adolescents.

IN THE LAST two decades the relationship between circadian preference and mood disorders has been explored. As regards depression, it has been shown that there is a link between eveningness and mood levels both in clinical and healthy samples.1,2 In line with these data, there is a higher prevalence of evening types among people with seasonal affective disorder (SAD),3 as well as a positive relationship between eveningness and mood seasonality in healthy young adults and adults.4,5 It has been found that circadian type is linked to circadian clock-related polymorphisms and it has been suggested that these polymorphisms could also be involved in the susceptibility to SAD.6

A variable that could modulate the association between circadian preference and mood seasonality is age, because both change over the lifespan, with an increase of both eveningness preference and seasonality of mood during adolescence.7,8 The aim of the present study was to explore the relationship between circadian type and mood seasonality in adolescents. To the best of our knowledge, no previous studies on adolescence have been performed. Because it has been proposed that there is a possible common genetic background for circadian type and seasonality of mood,6 we could expect to see a steady relationship across the lifespan, that is, higher mood seasonality in evening-type adolescents.

METHOD

Participants

The sample consisted of 1539 adolescents (881 female; 658 male). The mean age of the overall sample was 13.75 ± 2.12 years, while the age range was between 10 and 17 years. The age mode and median were 16 years and 14 years, respectively. The mean age of the boys (13.37 ± 1.99 years) was significantly lower than that of the girls (14.03 ± 2.16 years; t1537 = −6.15; P < 0.000001). Only the questionnaires fully filled out (n = 1539) were analyzed, and 148 questionnaires were discarded for this reason.

The parents provided informed consent prior to the participation of their sons/daughters in the research project; only 2% of parents did not provide informed consent. The ethics committee of the Department of Psychology of University of Bologna approved this study, which complied with the tenets of the declaration of Helsinki.

Procedure

Participants completed the Italian version of the Morningness–Eveningness Questionnaire for Children and Adolescents (MEQ-CA), which is a 19-item mixed-format scale in which participants are requested to indicate their own life rhythms and habits as far as going to sleep and waking up are concerned,9,10 and to supply further useful information to find the most suitable rhythm. The MEQ-CA score ranges from 16 to 86, with scores >58 identifying participants as morning types, and scores <42, as evening types.

Moreover the Italian version of the Seasonal Pattern Assessment Questionnaire for Children and Adolescents (SPAQ-CA), which is a self-reporting questionnaire for normal and clinical seasonal mood variations,11,12 was used. Eleven items on a 5-point Likert scale (score for each item: 0–4) evaluate the seasonal changes in the following areas: sleep length, mood, social activity, weight, energy, appetite, school performance, irritability and conduct problems. Summing the scores of those items, we obtain the Global Seasonality Score (GSS), with higher values corresponding to higher mood seasonality (total score range, 0–44).

Approximately half of the questionnaires were given in the first half of the scholastic year and the remaining part in the last half.

Statistical analysis

We performed a Kolmogorov–Smirnov test to verify the normal distribution of the dependent variable, the GSS. Afterwards, we carried out an ancova with gender and circadian type as independent variables, age as a covariate and GSS as the dependent variable. When ancova was significant, Tukey's post-hoc test for unequal samples was performed. Moreover we carried out a multiple regression analysis with gender, age and circadian type as independent variables and GSS as the dependent variable. Significance was set at P < .05.

RESULTS

The present subjects consisted of 279 morning types (18.13%; 176 girls; 103 boys), 1105 intermediate types (71.80%; 612 girls; 493 boys) and 155 evening types (10.07%; 93 girls; 62 boys). The distribution of boys and girls in the three circadian types was not significantly different.

Carrying out the Kolmogorov–Smirnov test, the frequency distribution of GSS was normal (d = 0.03; P > 0.05; skewness = −0.04; kurtosis = −0.41). Performing ancova, circadian preference was found to have a significant effect on GSS (F2,1532 = 10.81; P < 0.00001; Fig. 1). Performing Tukey's post-hoc test for unequal samples, all comparisons were significant, with evening types (23.13 ± 7.79) scoring higher than intermediate types (19.93 ± 7.46; P < 0.0005) and morning types (17.66 ± 8.21; P < 0.00005). Furthermore, intermediate types had higher mood seasonality scores than morning types (P < 0.005). Gender and the interaction between the two independent variables did not reach significance.

Figure 1.

Mean Global Seasonality Score by circadian type, according to gender (inline image, male; inline image, female) and (inline image) total sample.

Carrying out multiple regression analysis, age (β = 0.30; t1535 = 12.08; P < 0.000001) and circadian preference (β = 0.11; t1535 = 4.68; P < 0.000005) proved to be significant GSS predictors. The model with two significant predictors accounted for 12% of variance.

DISCUSSION

The present study has shown for the first time a significant relationship between circadian preference and mood seasonality in adolescents, with evening types gaining the highest GSS, followed by intermediate types and morning types. These data seem to confirm the expectations based on a supposed common genetic background between circadian type and mood seasonality.6 Because recent studies on genetic factors in extreme chronotype, depression including SAD and circadian rhythm sleep disorders found similar or different polymorphisms in these pathological situations,13 future genetic studies could provide further information.

The present data are in line with a previous study carried out in a similar age group, although that work was not focused on mood seasonality, only mood.14

Similarly to non-seasonal depression, it has been thought that higher eveningness in seasonal depressed patients might not be a feature of the depressive state, but reflect a premorbid trait or susceptibility.4 Because the present data are of a cross-sectional nature, we were not able to draw any conclusions about the causative relationship between eveningness and higher seasonality of mood. It should be noted that a limit of the present study was that data were based on subjective evaluations alone, although this weakness may be partially offset by the large sample size.

The present results suggest that a significant association between circadian preference and seasonality of mood begins during adolescence, and continues at least into young adulthood and adulthood.4,5 Nevertheless, studies on different age groups such as childhood and old age are lacking, and thus further studies on these age groups are required to fully understand the pathway of this relationship.

Finally the present data could have relevant implications in the clinical field, both for prevention and treatment. Indeed if the association between circadian type and seasonality of mood receive further confirmation, the assessment of circadian preference could lead to the early identification of people at higher risk for seasonal depression. Furthermore, supplementary therapeutic interventions aimed at manipulating circadian phase could facilitate relief of seasonal depressive symptomatology.

Ancillary