^{1}

^{1, 2}

^{3}

^{4}

^{5}

^{6}

^{4}

^{7}

^{1}

^{1}

^{2}

^{3}

^{4}

^{5}

^{6}

^{7}

The purpose of this study was to test the validity of affective temperaments for predicting psychiatric morbidity and suicide risk, using a two-factor model to explain the relationships between temperament, anxiety, depression, and hopelessness. We investigated 210 high school students, 103 males and 107 females, 18-19 years old, who were administered self-report questionnaires to assess temperament (TEMPS-A), depression (BDI-II), anxiety (STAI) and hopelessness (BHS). The final structural model had a good fit with the data, with two factors significantly correlated, the first labeled unstable cyclothymic temperament including Dysthymic/Cyclothymic/Anxious temperament, Irritable temperament and Depression, and the second labeled Demoralization including Anxiety (State/Trait) and Hopelessness. Depression, anxiety and hopelessness are in a complex relationship partly mediated by temperament.

The widely accepted etiological hypothesis proposes a cooccurrence between depression and anxiety, since these conditions share several symptoms and causal factors. However, hopelessness appears to play a unique role in this cooccurrence [

It is of special importance to understand the mechanisms involved in the development of depression and anxiety in adolescents, not only because of the high rates of prevalence of these disorders all over the world [

Another important construct, demoralization, was described by Frank [

Akiskal [

In spite of the fact that all the above factors, depression, anxiety, hopelessness, and affective temperaments, are related to suicidal behavior, no study so far investigated the interrelationship between these phenomena in a single framework. Delineating the associations between these phenomena is crucial in understanding how they mediate and influence one another in the development of suicidal behavior, describing possibilities for intervention on multiple levels. The aim of our present study was to explore the association of the five affective temperaments in the Akiskal model (Depressive, Hyperthymic, Cyclothymic, Irritable, and Generalized Anxious) with anxiety, depression, and hopelessness in a sample of adolescents. Our study also extends the model of demoralization by introducing temperament dysregulation, and we propose a two-factor model that explains the relationships between temperament, anxiety, depression, and hopelessness.

In the second half of 2009, 210 adolescents from the south of Italy voluntarily participated in this study. The participants were high school students, aged 18 to 19 years, 103 males (mean age: 18.43, SD: .49) and 107 females (mean age: 18.49, SD: .50). All students were from the same grade school and were late adolescents. After obtaining the permission from administrators of the schools, we administered four psychometric instruments. All subjects were culturally homogeneous, although they came from families with various socioeconomic backgrounds (including farmers, merchants, professionals, and industry workers), mainly middle class to upper-middle class. Their sociodemographic characteristics are shown in Table

Socio-demographic characteristics of subjects.

Characteristics | Males | Females | Statistics | |
---|---|---|---|---|

Age (years) | .39 | |||

Sociocultural level of families | .05 | |||

Low (N) | 9 | 14 | ||

Middle (N) | 69 | 54 | ||

High (N) | 25 | 39 |

^{
a} Values shown as mean ± SD.

The first version of

The

The

The

Two-tailed

Structural Equation Modeling (SEM) relies on several statistical tests to determine the adequacy of model fit to the empirical data. Specifically, confirmatory factor analysis (CFA) allows testing a hypothesis concerning a relationship between the observed variables and their underlying latent constructs. In this way, it is possible to use the prior empirical research to postulate a relationship pattern

The chi-square test indicates the amount of difference between the expected and observed covariance matrices. A chi-square value close to zero indicates little difference between the expected and observed covariance matrices. In addition, the probability level must be greater than 0.05 when chi-square is close to zero. The Comparative Fit Index (CFI) is equal to the discrepancy function adjusted for sample size. The CFI ranges from 0 to 1 with a larger value indicating better model fit. Acceptable model fit is indicated by a CFI value of 0.90 or greater. Root Mean Square Error of Approximation (RMSEA) is related to the residual in the model. RMSEA values range from 0 to 1 with a smaller RMSEA value indicating better model fit. Acceptable model fit is indicated by an RMSEA value of 0.06 or less [

The descriptive statistics and zero-order correlations between the variables are presented in Table

Descriptive statistics and zero-order correlations.

Means | SD | 2 | 3 | 4 | 5 | 6 | |
---|---|---|---|---|---|---|---|

(1) TEMPS-A—Dys/Cyc/Anx | 18.89 | 7.97 | .605 | .579 | .748 | .716 | .736 |

(2) TEMPS-A—Irritable | 6.39 | 3.28 | .470 | .671 | .589 | .571 | |

(3) BHS | 6.23 | 4.99 | .634 | .544 | .717 | ||

(4) BDI-II | 15.80 | 10.96 | .648 | .719 | |||

(5) STAI-S | 48.55 | 13.30 | .752 | ||||

(6) STAI-T | 46.89 | 11.63 |

Two confirmatory factor models were specified using the sample covariance matrix and the estimated parameters using maximum likelihood method. Model 1 was a 1-factor model with all the six variables loading on a general factor. Model 2 represented a two-factor model with three variables loading on each of their respective factors, and these were as hypothesized. The first model was statistically overidentified and produced fit indices as follows:

It seemed likely, therefore, that a two-factor model is more appropriate to describe the relationships between observed variables and latent factors, and this model appears to fit the data substantially better than the single factor model. The second model produced fit indices as follows:

Structural Equation Model with two correlated factors. The curved arrow represents the relationship between the latent factors, while straight arrows from latent factors to observed variables represent factor loadings.

This study sought to test two models for explaining the relationship between anxiety, depression, hopelessness, and affective temperaments The novelty of the present study, compared to that of Cunningham and colleagues [

The unstable cyclothymic temperament factor included the Irritable temperament (52%) and the Dysthymic/Cyclothymic/Anxious temperament (75%), as well as the component of depression (76%), involving individuals’ inability to experience positive emotions, feeling tired and dissatisfied, having a sensory response that involves slowing down, and holding general critical thoughts about oneself. Depression was associated with a feeling of vulnerability and structurally characterized by specific temperament traits with negative consequences both on personal mood and on interpersonal relationships. Thus the Unstable cyclothymic temperament factor involved a psychological condition in which individuals perceive a state of disequilibrium, characterized by generally pessimistic and self-critical cognitions.

The second factor, Demoralization, involved the constructs of Trait Anxiety (84%), State Anxiety (67%) and Hopelessness (56%) and, as with the previous factor, there were sensory and cognitive components. In this case, the “feeling” has no direct consequence with the representation of themselves as people with a basic structure of personality that is able or unable to act or think, but shifts attention from the here and now (I am) to a representation of a future that is negative and unchangeable. Hopelessness and anxiety are associated with a temporal perspective of negative future expectations and suicidal ideation.

Incidentally, previous research has shown that the Dysthymic/Cyclothymic/Anxious temperament predicts hopelessness but, unexpectedly, depression is not a good predictor of the risk of suicide [

The strength of the present study is that it has revealed the relevance of what has been called Unstable cyclothymic temperament, a latent factor that involves affective temperaments and depression, in addition to Demoralization that consists of anxiety and hopelessness. It extends the theoretical framework and provides a new approach for clinical intervention in the treatment of symptoms of depression and anxiety in adolescents and for preventing suicidal behaviors.

The novelty of our study is the introduction of affective temperaments in the model of phenomena closely related to and possibly predicting suicidal behavior. Affective temperaments are a crucial part of this model because due to their spectrum nature they underlie normal healthy personality processes, present a latent form of affective pathology and predispose to affective illness, and have a pathoplastic role in case of major mood disorders as well. The overall finding of our study establishes a general framework of those psychologically relevant factors known to play a crucial role in the background of suicidality. Better understanding of how depression, anxiety, hopelessness, and affective temperaments relate to each may provide us with a model for understanding the evolution of suicidal behavior as well as helping us determine multiple points of intervention at multiple levels.

However, there were some limitations to this study, such as a cross-sectional study design, while longitudinal studies are more informative; moreover the limited sample size (making exploration of sex differences in these associations impossible) and the use of a nonclinical population hinder the generalization of the results.

Furthermore, psychometric instruments used in this paper were used in a non-clinical population rather than a clinical population whereas they were involved in many comprising clinical populations. Therefore, our study lacks information related to lifetime depression or hypomania; in addition it lacks self-rated scale for hypomania, as well as data concerning the psychological role of hyperthymic temperament, and scales exploring the attitudes, beliefs, and cognitions.

Despite these caveats, this paper offers an original perspective that may be of help in understanding complex relationship between constructs involved in our study. Further research is needed on how temperament and hopelessness may be related to the cooccurrence of depressive and anxiety symptoms. There may be other pathways involved in these links, such as emotional distress, mood disorders, and insecure attachment. Research should also investigate the role of cognitions and expectations of others in interpersonal contexts on these associations.