This is an important and worthwhile effort, but its credibility nonetheless falters as it strains to achieve generalizations, frequently combining proverbial apples and oranges to offer fruit salad conclusions. For example, in reviewing old and disparate studies, the authors confound several issues into a notion of "complexity/chronicity" relating to the severity, chronicity, and multiple types of comorbidity someone with a depressive disorder might have. In addition, in addressing the important question of how valuable psychotropic medication may be for patients with mild depressive episodes, they principally draw conclusions from one study of 184 patients, 20 of whom received an antidepressant (desipramine); details of dose, duration, compliance, and comorbidity are not mentioned. There is a general bias against psychopharmacology, and only an idiosyncratic, selective review could result in assertions such as that depression fails to be a specific entity because it lacks a specific treatment, citing two studies allegedly demonstrating that "nonantidepressant medications and a variety of psychotherapies are all, on the average, similarly effective among patients diagnosed with Major Depression." The fact that selective serotonin reuptake inhibitors may successfully treat individuals with depression, generalized anxiety disorder, posttraumatic stress disorder, obsessive-compulsive disorder, social phobia, panic disorder, and premature ejaculation does not mean that these disorders are not usefully separated, any more than the fact that propranolol is useful in treating hypertension, angina, essential tremor, and performance anxiety means that these disorders are not validly separable. There is poor understanding of the psychopharmacological literature, an atavistic belief in the validity of "endogenous depression," and an opinion that psychotropic medication should not be a first-line treatment even for moderately severe depression.