Depression. Aaron T. Beck, M.D.

psychotic factor.” The bipolar factor closely resembled that which was extracted in the study by Kiloh and Garside. Among features with high positive loadings on the first factor, and thus corresponding to a diagnosis of endogenous depression, were adequate premorbid personality; absence of adequate psychogenic factors in relation to illness; a distinct quality to the depression; weight loss; pyknic body build; occurrence of previous depressive episode; early morning awakening; depressive psychomotor activity; nihilistic, somatic, and paranoid delusions; and ideas of guilt. Among features with a negative loading, corresponding to a diagnosis of neurotic depression, were anxiety; aggravation of symptoms in the evening; self-pity; a tendency to blame others; and hysterical features.

      By means of multiple regression analysis, three series of 18 weighted coefficients for the differential diagnosis between the two varieties of depression and for the prediction of ECT response at three and six months were calculated. The multiple correlations between the summed features, on the one hand, and diagnosis and outcome at three and six months, on the other, were 0.91, 0.72, and 0.74 respectively. It was found that ECT response could be better predicted by the direct use of the weights for ECT response than from the diagnostic weights alone. The weights based on the 18 clinical features were complex, and therefore a table was constructed giving simplified weights based on ten features of diagnosis. When the weighted scores for each patient were computed, it was found that, of the patients with a score of six or higher, 52 had been diagnosed clinically as endogenous and three as neurotic. Those patients scoring below six included one endogenous and 60 neurotic depressives. The amount of overlap, consequently, was small, and the findings supported the two-type hypothesis.

Methodological Problems

      Several methodological questions may be raised in connection with these studies. First, the reliability of the ratings of the clinical material was not reported. As has been pointed out in many papers, interjudge agreement tends to be relatively low when applied to clinical material; low reliability automatically imposes a limit on the validity of any findings based on these ratings. In addition, since the psychiatrists making the ratings were cognizant of the underlying hypotheses, the possibility of bias in making their judgments cannot be excluded.

      The second methodological problem concerned the differences between the two groups studied with respect to uncontrolled variables of importance, such as age and sex. For example, relative sleeplessness and loss of appetite are characteristic of older patients. (We found a relatively high correlation between age and loss of appetite among our psychiatric patients.) There was also evidence that females and males reacted differently to stress. Since these studies did not control adequately for either age or sex (or for other demographic variables), we cannot be certain that the salient differences between the two groups are explained by the dualistic hypothesis.

      There is also a problem in the interpretation of the factor analysis. The authors extracted a bipolar factor that seemed to indicate a division of the patient sample into two independent groupings. In order to prove that these groupings apply to different kinds of patients rather than simply to different clusters of signs and symptoms, it is necessary to show that there is a clear-cut splitting of the patient sample into two independent groups. Kiloh and Garside²⁷ did not present any information regarding the distribution of the cases. In the study by Carney, Roth and Garside,²⁸ however, a separation of the endogenous and neurotic groups was achieved by weighting items based on the statistical analysis.

Studies of Symptomatology

      Hamilton and White²⁹ performed a factor analysis on data obtained from 64 severely depressed patients who had been evaluated with the use of Hamilton’s rating scale.³⁰ The first of the four factors obtained included such clinical features as depressed mood, guilt, retardation, loss of insight, suicidal attempt, and loss of interest. It proved, according to the authors, to be correlated with a clinical diagnosis of retarded depression. Significantly different mean scores between the endogenous and reactive groups were obtained for this first factor. It should be emphasized, however, that this finding does not specify whether the difference is qualitative or merely quantitative.

      Unfortunately, the so-called precipitating factors given to justify the diagnosis of reactive depression seemed unconvincing. In the three cases of reactive depression presented, the authors refer to the following as the psychological precipitating factors: one patient was left alone for prolonged periods while his wife went to look after their sick daughter; another was put in charge of a program that was beyond his capabilities; the third learned that the pulmonary tuberculosis he had had for nine years was bilateral.

      Findings contradictory to those reported by Hamilton and White were contained in a study by Rose.³¹ This investigator used the same clinical rating scale in studying 50 depressed patients. The patients were divided into endogenous, reactive, and doubtful groups. In contrast to Hamilton and White, Rose found no significant differences in symptoms among the three groups.

Physiological Responses and Tests

      Kiloh and Garside referred to the work on sedation threshold by Shagass and Jones,³² which indicated that cases of endogenous depression had lower sedation thresholds than those of neurotic depression. They also cited the work of Ackner and Pampiglione³³ and Roberts,³⁴ which failed to confirm Shagass’s results. The work of Shagass and Schwartz³⁵ on cortical excitability following electrical stimulation of the ulnar nerve was also cited. They found that in 21 patients with psychotic depression, the mean recovery time was significantly increased. However, controls for age were not included in these early studies, thus confounding the variables of interest.

      The Funkenstein test was cited by Sloan et al. as additional evidence supporting the distinction between these two types of depression.³⁶ Better designed studies failed to substantiate these findings (see Chapter 9). More recent research supports an interactive perspective on the development of depression, including cognitive vulnerability, stress, early experiences, and genetic components (see Chapter 13).

Body Build

      Kiloh and Garside quoted a study by Rees³⁷ that demonstrated an association between neurotic depression and leptomorphic physique, and between eurymorphic physique and manic-depressive disorder. Here again, the mean age of the manic-depressive group was significantly higher than that of the neurotic-depressive group. As will be indicated in Chapter 9, body build becomes more eurymorphic with increasing age.

Response to Treatment

      Kiloh and Garside cited several studies suggesting that exogenous depression reacts poorly to electroconvulsive therapy but endogenous depression reacts favorably. Some evidence to support this claim is contained in a study by Rose,³¹ who found that better response was obtained only for the women who had endogenous depression. There was no difference in treatment response among the men in this study. (The study by Carney et al.²⁸ described in detail above helped substantiate the claim of a differential response.)

Depressive Equivalents

      Many writers attempted to spread the umbrella of depression to cover cases showing clinical symptoms or behaviors different from those generally indicative of depression. The term depressive equivalents was introduced by Kennedy and Wiesel³⁸ to describe patients who had various somatic complaints but did not show any apparent mood depression. They reported three cases characterized by somatic pain, sleep disturbance, and weight loss, all of whom recovered completely after a course of ECT.

      A number of other terms have been applied at various times to designate such cases of concealed depression. These include incomplete depression, latent depression, atypical depression, and masked depression. Various psychosomatic disorders, hypochondriacal reactions, anxiety reactions, phobic reactions, and obsessive-compulsive reactions have also been implicated as masking the typical picture of depressive reactions.³⁹

TABLE