Dr. Klein Replies

To the Editor: Patients with dysthymic disorder differ from patients with episodic major depressive disorder in a number of respects, including experiencing higher rates of axis I and II comorbidity (1, 2). Dr. Bourgeois raises the important question of whether the differences in patients with comorbid personality and substance use disorders explain why patients with dysthymic disorder had a poorer course and outcome than patients with episodic major depressive disorder in our 5-year follow-up study.

To address this question, we conducted hierarchical multiple linear regression analyses to determine whether baseline diagnosis (dysthymic disorder versus episodic major depressive disorder) predicted mean level of depression across the follow-up period (3) and scores on the Modified Hamilton Rating Scale for Depression (4) at the 5-year follow-up evaluation after control for the presence of a personality disorder diagnosis (5) and lifetime alcohol and drug use disorders (few patients were currently abusing substances at entry into the study). Baseline scores on the modified Hamilton scale were included as a covariate in the analysis of 5-year scores on the modified Hamilton scale.

After control for personality disorder and lifetime substance use disorder, a diagnosis of dysthymic disorder (as opposed to episodic major depressive disorder) significantly predicted a greater mean level of depression across the follow-up period (change: R²=0.18; F=33.73, df=1, 121, p<0.001). Similarly, after control for baseline score on the modified Hamilton scale, personality disorder, and lifetime substance use disorder, dysthymic disorder was found to be significantly associated with a higher score on the modified Hamilton depression scale at follow-up (change: R²=0.09; F=13.89, df=1, 107, p<0.001). Thus, although differences in patients with comorbid personality and substance use disorders may be a contributing factor, they do not account for the differences in course between patients with dysthymic disorder and patients with episodic major depressive disorder in our study. Nonetheless, the impact of comorbidity on the course of dysthymic disorder is an important issue that we plan to explore in greater detail in future analyses of our data.