In recent years, several lines of evidence have emerged suggesting that eating disorders in general, and bulimia in particular, are in some way linked to the major affective disorder. ' Studies, however, on the phenomenological relationship between the two illnesses have shown inconsistent results, which might be partly due to the heterogeneity of eating disorders and shortcomings of a diagnostic assessment of affective disorders. We looked for depressive syndromes in subtypes of eating disorders using standardized assessment and DSM-111 diagnostic criteria.Characteristics of the study groups are depicted on TABLE 1. Diagnostic assessment was made using the National Institute of Mental Health Diagnostic Interview Schedule and the Eating Disorder Inventory. Sixty-three patients were interviewed as inpatients and 22 as outpatients, one year after hospital treatment.DSM-111 lifetime rates of affective disorders are depicted in FIGURE 1. The three bulimic groups differed significantly from the restrictors. The onset of the major affective disorder occurred significantly more frequently (in 72% of the patients) at least one year after onset of the eating disorder than before the eating disorder (9.4%). The rate of the current major affective disorder in the 63 inpatients with an active and severe eating disorder was significantly higher (31.7%) than in the outpatients whose eating disorder was in remission (4.5%).In some contrast to previous results,* our data on temporal relationship indicate that the affective disturbance was likely to be secondary to an eating disorder. This is further supported by our finding a very low rate of depression in patients after treatment of the eating disorder. The higher frequency of major affective disorders in bulimics than in restrictors could partly be explained by the longer duration of illness in the bulimics. Our data are also consistent, however, with recent suggestions that bulimia is related to a more severe unspecific psychopathology and has a poorer prognosis. This view is further supported by our finding that in most patients developing bulimia after anorexia nervosa, the onset of major depression occurs only when bulimia is present. REFERENCES

DSM-III lifetime diagnoses were assessed in 52 patients with a lifetime history of anorexia nervosa or bulimia by means of a standardised diagnostic interview. It was found that 44.2% had a lifetime diagnosis of DSM-III major affective disorder, with abstaining anorectics having a lower rate of depression than those with bulimic symptoms. In the great majority of cases, the onset of affective disorder post-dated the onset of the eating disorder by at least one year. In patients whose eating disorder was in remission, the rate of depressive symptoms was lower than in those in the acute stage of their illness. These findings, combined with recent studies on biological changes in eating disorders, and psychological theories of depression, suggest that in most cases in which the two conditions are associated, the depression is secondary to the eating disorder.

Proceeding from a cognitive view of depression, the relationship between depression and cognitive characteristics referring to shape and weight was investigated in 99 patients with anorexia nervosa or bulimia. Significant correlations between depression and cognitive schemata were found in both patient groups. Multiple regression analysis revealed that four scales measuring cognitive features explained between 34% and 45% of the variance in the depression score. The strongest single predictor was the variable “negative body attitudes.” In conjunction with observations in cognitive‐behavioral treatments, the data indicate that the role of cognitive distortions characteristic of patients with eating disorders is crucial for the development and maintenance of depressive symptoms in these patients.

Influence of diet composition on mood during weight-reducing diets was studied in healthy young women of normal weight. A broad range of macronutrient intake was achieved by means of divergent dietary instructions for the composition of a 1,000 kcal per day diet adhered to for six weeks. Global mood during the last three weeks of the diet was significantly better in the "vegetarian" than in the "mixed" diet group. During this time a significant correlation was observed between relative carbohydrate intake and global mood (r = -0.74; p less than 0.01) and between the ratio of plasma tryptophan to other large neutral amino acids (a predictor of tryptophan flow into brain) and global mood (r = -0.52; p less than 0.05). Results suggest that group differences are related to differences in carbohydrate intake. It is hypothesized that impairment of central serotonergic function due to reduced tryptophan availability can prompt mood deterioration in situations of relatively low carbohydrate intake.