Depression. Aaron T. Beck, M.D.

for depressives was 36 times as high as for the general population and about three times as high as for either schizophrenics or alcoholics.

      The suicide rate among patients known to be suicidal risks is apparently high. Moss and Hamilton²⁸ conducted a follow-up study for periods of two months to 20 years of 50 patients who had been considered “seriously suicidal” during their previous hospitalization (average 4 years). Eleven (22 percent) of the 50 later committed suicide. In a retrospective study of 134 suicides, Robins et al.²⁹ found that 68 percent had previously communicated suicidal ideas and 41 percent had specifically stated they intended to commit suicide.

      The figures available at that time clearly indicated that the suicidal risk was greatest during weekend leaves from the hospital and shortly after discharge. Wheat,³⁰ surveying suicides among psychiatric hospital patients, found that 30 percent committed suicide during the period of hospitalization, and 63 percent of the suicides among the discharged patients occurred within one month after discharge. Temoche et al.²⁷ calculated that the suicidal risk in the first six months after discharge is 34 times greater than in the general population and in the second six months about nine times greater. About half of the suicides occurred within 11 months of release.

      Many earlier studies reported the observation that women depressives attempted suicide more frequently than men but that men were more often successful. Kraines¹⁸ reported that, in his series of manic-depressive patients, twice as many women as men attempted suicide and three times as many men as women were successful suicides.

      Although no data are available regarding the suicidal methods employed by depressives, recent statistics for the general population may be relevant. In 2001 intentional self-harm (suicide) by discharge of firearms was 16,869. By other and unspecified means, the number was 13,753. The ratio of male to female was 4.6 to 1; Black to White .5; Hispanic to non-Hispanic .5.³¹

      There is evidence that the number of suicides each year in the United States is greater than the official 2001 report of 30,622. Many accidental deaths actually represent concealed suicides. For instance, in 1962 MacDonald³² reported 37 cases of attempted suicide by automobile. Writers believed that the actual rate of suicide was three or four times as great as the official rate. The number of attempted suicides was believed to be seven or eight times the number of successful suicides.³³

      Homicide may occur in association with suicide among depressed patients.³⁴ Reports, for example, of parents killing their children and then themselves are not rare. One woman, convinced by her psychotherapist that her children needed her even though she believed herself worthless, decided to kill them as well as herself to “spare them the agony of growing up without a mother.” She subsequently followed through with her plan.

      Several factors contribute to the risk of attempted or completed suicide. Risk is especially high during a major depressive episode in those with psychotic symptoms, previous suicide attempts, a family history of completed suicides, and concurrent substance use.^35,36 The best indication of a suicidal risk is the communication of suicidal intent.²⁹ Stengel³³ pointed out that the notion that the person who talks about suicide will never carry it out is fallacious. Also, a previous unsuccessful suicidal attempt greatly increases the probability of a subsequent successful suicidal attempt.^36,37 Brown et al.³⁸ were able to reduce repeat suicide attempts by 50 percent through the application of cognitive therapy, compared to usual care of tracking and referral services. They also were able to reduce depression severity and hopelessness (see Chapter 15).

      Over a 5-year follow-up period, Klein et al.¹⁵ found that suicide attempts were made by 19 percent (16 of 84) of patients with chronic depression, and one of these resulted in actual suicide. In this study, there were no attempts among 37 patients with episodic disorder only. This suggests that the rate of suicide attempts increases in cases of chronic depressive illness (dysthymic disorder) compared to episodic major depression.

      In addition to trying to elicit suicidal wishes from the depressed patient, the clinician should look for signs of hopelessness. In our studies we found that suicidal wishes had a higher correlation with hopelessness than with any other symptom of depression. Furthermore, Pichot and Lempérière,³⁹ in a factor analysis of the Depression Inventory, extracted a factor containing only two variables, pessimism (hopelessness) and suicidal wishes.

Suicide Risk in Bipolar Disorder

      Fagiolini et al.⁴⁰ found suicidal thinking and behavior to be common in individuals with bipolar disorder. The people in the study were 175 patients with bipolar I disorder who were participating in a randomized controlled trial, the Pittsburgh Study of Maintenance Therapies in Bipolar Disorder. Suicide had been attempted by 29 percent of the patients prior to entering the study.

      The method used in this study was to compare clinical and demographic characteristics of those who had attempted suicide before entering the study to those who had not attempted suicide. Among the conclusions was that greater severity of bipolar disorder and higher body mass predicted a history of suicide attempts. Severity was defined as a greater number of previous depressive episodes, as well as higher scores on an evaluator-rated measure of depression (Hamilton Rating Scale—25 items).⁴⁰

Predictors of Chronic Depression

      Riso et al.⁴¹ reviewed the studies of determinants of chronic depression. They reported that such determinants have not been adequately elucidated, but that studies have considered six possible factors: (1) developmental factors such as childhood adversity (early trauma or maltreatment), (2) personality and personality disorders like neuroticism (emotional instability or vulnerability to stress) and stress reactivity, (3) psychological stressors, (4) comorbid disorders, (5) biological factors, and (6) cognitive factors. In what follows, we summarize their findings.

Developmental Factors

      Among the developmental factors, there is some evidence for the importance of early trauma or maltreatment but not for early separation and loss.

Personality Disorders

      In 11 studies comparing personality rates in dysthymia to major depression, patients with dysthymia were found to have higher rates of personality disorders. However, as of 2002 only one prospective study had been carried out. The two conditions may share causal factors, rather than dysthymia developing as a consequence of personality disorder.

Psychological Stressors

      Concerning psychological stressors, the duration of chronic depression makes it more difficult to disentangle stressors that may lead to prolongation of depression from the effects of depression itself in generating stressors. Riso et al.⁴¹ noted that the APA diagnostic manuals assert that dysthymic disorder is associated with chronic stress, but that it is possible that the two studies supportive of this may be confounded by patient perceptions of stressors rather than actual events. Supportive of this is that treatment with antidepressant medications modifies reports of daily hassles.

Comorbid Disorders

      Findings on comorbid disorders include one study suggesting that chronic illness in a spouse can lead to dysthymia. Also, dysthymia has been found associated with several psychiatric conditions, including anxiety and substance abuse, with social phobia the most common.

TABLE 3-7. Cognitive Variables in Chronic Depression (CD), Nonchronic Major Depressive Disorder (NCMDD), and Never Psychiatrically Ill Controls (NPI)

      Adapted from Riso et al. 2003.

Biological Factors

      Biological factors are considered more fully in Chapter 9. With respect to predicting a chronic course of depression, neuroendocrinology studies have