|Home | About | Journals | Submit | Contact Us | Français|
Alexithymia, a clinical condition typified by a reported inability to identify or describe one's emotions, is associated with various forms of psychopathology, including depression. Using cluster analyses, Vanheule et al.  identified two groups of depressed outpatients, i.e. highly and moderately alexithymic. Highly alexithymic (HA) outpatients were more likely to be female, less likely to have children and were characterized by more somatic-affective symptoms of depression and interpersonal aloofness.
We extended these findings by examining personality traits and childhood sexual abuse history. In light of prior research, we expected that higher levels of neuroticism, lower levels of extra-version and openness to experience [2, 3] and a greater prevalence of childhood sexual abuse [4, 5] would characterize the HA group. Participants were 94 depressed patients [57.45% with recurrent major depressive disorder (MDD), 37.23% with single-episode MDD, 5.32% wit h depressive disorder not otherwise specified] 50 years of age and older recruited from psychiatric treatment facilities in Upstate New York.
Age-eligible patients who were thought to have an affective disorder were invited to participate. Individuals providing written informed consent completed questionnaires and the Structured Clinical Interview for DSM-IV Axis I disorders . Diagnoses were made in consensus conferences following reviews of the interview and medical chart data. The validity of this diagnostic method is well established .
Alexithymia was assessed with the 20-item self-report Toronto Alexithymia Scale . Its 3 subscales measure difficulty identifying feelings and distinguishing them from bodily sensations (DIF), difficulty describing and communicating feelings (DDF) and externally oriented thinking (EOT), the latter being a tendency to focus on concrete details of external events rather than on aspects of inner experience. Depressive symptom severity was assessed with the Beck Depression Inventory-II (BDI-II) . Three BDI-II subscales measure the cognitive, affective and somatic symptoms of depression [10, 11]. Five personality domains, i.e. neuroticism, extraversion, openness to experience, agreeable-ness and conscientiousness, were assessed with the NEO Personality Inventory . History of childhood sexual abuse was assessed using the Childhood Sexual Abuse subscale of the Childhood Trauma Questionnaire .
A latent class cluster analysis (M-Plus 4.20)  was performed on the DIF, DDF and EOT subscales. Cluster solutions were compared based on the Bayesian information criterion, Akaike's information criterion and entropy. A better model fit is indicated by lower values of the Bayesian information criterion and Akaike's information criterion, and a better classification is indicated by entropy values closer to 1. All 3 indicators favored a 3–cluster solution. This solution identified 3 groups, i.e. low alexithymia (LA; n = 11, 63.64% women), moderate alexithymia (MA; n = 40, 60% women) and HA (n = 43, 60.47% women). Table Table11 presents descriptive statistics for our study population. The distribution of mood diagnoses, single-episode MDD, recurrent MDD and depressive disorder not otherwise specified was not significantly different among the 3 alexithymic clusters (χ2 test 5.89, p > 0.05).
We conducted 10 separate multivariate generalized logit regressions; odds ratios were calculated for LA versus HA and MA versus HA class membership. Putative predictors were total BDI-II and the 3 subscales, childhood sexual abuse and the 5 personality domains. Covariates were age, gender and education. The 3 BDI-II subscales as well as the total score significantly distinguished LA and MA from HA. Lower levels of depressive symptoms significantly decreased the odds of HA membership. Odds ratios ranged between 0.32 and 0.94 (p < 0.05).
Childhood sexual abuse distinguished MA from HA; lower levels of childhood sexual abuse decreased the odds of HA membership (odds ratio 0.60, confidence interval 0.38–0.96; p < 0.05). Neuroticism, openness and conscientiousness also distinguished the 3 groups. Low neuroticism decreased the odds of HA membership, with odds ratios ranging between 0.95 and 0.97 (p < 0.01). Low openness and conscientiousness increased the odds of HA membership, with odds ratios ranging between 1.02 and 1.07 (p < 0.05).
The present study is limited by the small sample size. Findings cannot be generalized to other demographic groups. The cross-sectional design precludes causal conclusions.
Our findings partially replicated the results of Vanheule et al. ; however, while 3 groups emerged in our cluster analyses, 2 groups emerged in theirs. However, they generated 3 clusters in an initial solution based on the DIF, DDF and EOT subscales, and a 2–cluster solution emerged when they excluded EOT from the analysis because of lack of differentiation. In our analysis, we selected the cluster solution based on all 3 subscales because EOT did distinguish the groups. Like Vanheule et al. , we found that the HA group was characterized by more depressive symptoms. In contrast to their results, we did not find gender differences in group composition.
The HA group in our study was characterized by higher neuroticism and lower openness to experience and conscientiousness, a profile that reflects a distressed personality type [15,16,17], which has been related to poorer health outcomes and general functioning and more psychological distress . A similar personality profile was observed for the MA group, with the notable exception being their above-average levels of agreeableness and openness to experience.
Interestingly, the HA group was characterized by more childhood sexual abuse compared to the MA patients. These findings add to the mounting evidence for a relationship between childhood sexual abuse and alexithymia [4, 5, 19]; individuals with a history of childhood sexual abuse may have a reduced capacity to experience emotion in relation to their trauma, and this phenomenon may generalize to experiencing all emotions. Childhood sexual abuse has been found to interfere with the development of emotion regulation  and to be related to attachment disturbance . The combination of childhood sexual abuse and alexithymia must be considered in the design and implementation of treatment studies, as these patients are more resistant to treatment (the attachment disturbance makes it more difficult for these clients to engage in a therapeutic relationship) and have slower recovery rates and poorer outcomes [4, 22, 23]
These findings emphasize the importance of considering whether a patients is alexithymic. Identifying a patient as alexithymic may suggest a history of early traumatic events which increases the likelihood and severity of depression. Treatment should be tailored to address the depressive symptoms along with the affective experience (identification, differentiation, labeling and management of feelings).
This study was supported by grants R01MH60285, K24MH072712 and R24AG031089 from the National Institutes of Health, USA.