David L. Pauls and Adria M. DiBenedetto
Child Study Center, Yale University School of Medicine, New Haven, Connecticut 06510
The occurrence of anxiety symptoms with episodes of major depression is veil documented (2,8-11,16). When individuals with either an anxiety disorder or major depression are followed over time, a particularly strong association has been observed between panic anxiety and depression. Dealy et al. (8) compared patients who had anxiety neurosis alone with patients who had anxiety neurosis with major depression. They found a significantly elevated rate of panic attacks among patients with depression. Other studies have suggested that when the two disorders co-occur in the same individual, they may be temporally independent. Raskin et al. (17) reported that 14 of 17 patients with panic disorder had experienced major depression that did not coincide with an episode of panic disorder. Bowen and Kohout (1) found that 91% of patients with agoraphobia and panic attacks had had an episode of major depression that did not coincide with episodes of panic, whereas Munjack and Moss (14) found that 41% of agoraphobic patients with panic attacks had had previous episodes of depression. Results from other studies have also suggested an association between panic disorder and depression. However, none of these studies specifically addressed whether the two disorders occurred together or separately. Cloninger et al. (4) noted that 75% of patients who experienced panic attacks later developed depression and Crowe et al. (6) found that 13 of 41 patients with panic disorder had chronologically secondary depression. In a follow-up study of 29 patients with "cardiac phobia," Nutzinger and Zapotoczky (16) found that 82.8% showed depression in addition to suffering from anxiety symptoms. While it is clear that the onset of depression came after the onset of anxiety in these studies, the investigators did not report whether the episodes of depression occurred independently of the episodes of panic disorder experienced by the patients. In examining patients with major depression, Leckman et al. (12) found that 81 of 133 patients had anxiety disorders in addition to being depressed. Twentytwo of the 81 individuals with both anxiety and depression had panic disorder. of the total 133 depressed subjects, 16.5% experienced panic attacks with suf ficient frequency and severity to satisfy criteria for a diagnosis of probable or definite panic disorder. This frequency is higher than expected by chance alone. These results, taken with the data from studies of panic disorder patients, suggest that panic disorder (PD) and major depressive disorder (MDD) may share some etiological factors. Treatment studies have also provided data that support the hypothesis of a potential relationship between PD and MDD. It is well known that tricyclic antidepressants and monoamine oxidase inhibitors are effective in the treatment of MDD (7). Evidence from controlled studies indicates that these agents also are effective in the reduction of the frequency of panic attacks (3). This antipanic effect appears to be independent of clinical manifestations of depression (18,22) suggesting that there may be some common underlying etiological factors for both panic and major depressive disorders. Although results from the above studies imply that PD and MDD are strongly associated and may have common factors involved in their etiology, additional data are needed to examine more carefully the relationship between the two illnesses. Since both disorders have been shown to be familial, family data can be quite useful in elucidating whether they share etiological factors. If PD and MDD are etiologically related, then increased rates of one disorder should be observed in the relatives of patients with the other illness. Family data have given apparently contradictory results. Several studies (1,12,14,19) have suggested that there is a familial relationship between PD and MDD since an increased rate of depression was observed among the relatives of probands with either MDD and PD (MDD + PD) or with PD alone. However, other studies (6,15) did not find an increased rate of primary depression in the relatives of probands with PD or agoraphobia with panic attacks (Ag/PD). The discrepancy between these family study results may be accounted for by methodological differences (5,13). The purpose of the study reported here was to reexamine the reports of Leckman et al. (12), Crowe et al. (6), and Noyes et al. (15) to determine if methodological differences could account for the apparent discrepant findings. Methodological differences between the studies focused on the procedures used in assigning diagnoses. In the Crowe et al. and Noyes et al. studies, DSM- III convention was used in that a diagnosis of a specific psychiatric condition was precluded when it could have been secondary to another mental disorder. For example, a diagnosis of primary major depression would not have been made if it occurred after the onset of panic disorder. Thus, the morbid risk figures reported in these two studies reflect the rates of chronologically primary disorders. These rates, in effect, represent the frequency of ill individuals among the relatives since no one would be included more than once. On the other hand, Leckman and co-workers did not use the primary/secondary distinction in reporting diagnoses. If an individual had sufficient symptoms to meet DSM- Ш criteria for a specific diagnosis, that diagnosis was assigned and reported. An individual could, and often did, receive several diagnoses. Therefore, the rates reported by Leckman et al. were rates of diagnoses and not rates of ill individuals, дп individual was included in the computation of rates as many times as that person received a diagnosis. Because of this difference in how rates of illness were reported, it is difficult to compare the results of the studies.