ANTIDEPRESSANT THERAPY IS now the subject of a nation-wide controversy in Japan. Japanese psychiatrists are being accused of dubious diagnoses, excessive prescriptions and insufficient understanding of personal mentalities. What has happened to us?

Japanese psychiatrists had preserved the endogenous–neurotic dichotomy of depression and had regarded intensive pharmacotherapy (with tricyclic antidepressants) as treatment for endogenous depression until around the turn of the century when selective serotonin re-uptake inhibitors were introduced to Japan.[1, 2] We then came to know that effects of newer antidepressants were established through randomized controlled trials that recruited patients with major depression. Thereafter, results of numerous clinical trials were introduced and the language of the psychiatrist changed drastically, while we hardly grasped the gestalt of major depression. Is this a problem inherent in the concept of the disease, or have we failed to import anything attached to it? Apart from the cultural context, here we try to analyze what in the concept of major depression confuses us.

When major depressive disorder appeared in the Research Diagnostic Criteria, it was described as a category encompassing patients of various subtypes, including ‘some cases that would be categorized as neurotic depression, and virtually all that would be classified as involutional depression, psychotic depression, and manic depressive illness, depressed type.’[3] Although several subtypes of depressive disorder were provided, the concept of major depression seemed to reject a clear-cut separation of these types. Such criticism of the classification of depression was developed by British psychiatrists early in the last century through meticulous clinical observation,[4] and it is still echoed in several recent epidemiological studies that demonstrate the entangled effects of environmental and genetic factors on the onset of major depression.[5] Empirically, we appreciate the importance of this critical attitude toward simplistic attribution; the more closely a patient is studied, the harder it becomes to justify simple qualitative distinctions. Detailed knowledge about the environmental and constitutional factors of each patient is crucial for treatment. For example, a patient who shows a seemingly reactive onset may actually have considerable genetic risk, whereas another patient with endogenomorphic features may be affected by an unstable relationship with his or her spouse. When we disentangle the threads, we have to draw them one after another. Similarly, for each patient, effective treatment (whether it is pharmacotherapy, cognitive modification or assistance with realistic problem solutions) may vary at each stage.

However, the effectiveness of such personalized methods of treatment is difficult to demonstrate in clinical trials. The evidence level in the treatment of such a heterogeneous disease may reflect methodological difficulties rather than actual importance in clinical practice. Most clinical trials recruit numerous patients with major depression irrespective of such diversity, resulting in generalizable findings with modest significance. When clinicians other than the specialists are informed of these results, they understand only the generalizable relationship between major depression and its treatment response. The critical attitude toward simplification, which established the concept of the disease, is virtually lost, and there remains only a broad category that is susceptible to medicalization. Ironically, what was once a criticism of simplification now serves as the grounds for further simplification.

Recognizing this process may be helpful in analyzing the ongoing conflict over bereavement exclusion criteria. While Kendler argues for the elimination of the criteria relying on the prudence of ‘good’ psychiatrists,[6] Horwitz and Wakefield address the potential harm of over-diagnosing depressive disorders by general physicians in the primary-care setting[7] who have been informed about simplified concepts and generalizable research findings. So goes the simplification process of psychiatry during its propagation from specialists to general physicians, from one generation to another[8] and, in our case, from one culture to another.

An additional problem with major depression is its threshold. Based on the current diagnostic criteria, the category of major depressive disorder shows a poor correlation with treatment response, contrary to the expectations of most non-specialist clinicians. With the required number of symptoms, the threshold may be too low to predict responses to antidepressants and too high to judge the requirement for clinical attention.[9]

We suggest that recognition of these two points, the simplification process and the inappropriate threshold, will contribute to the understanding of how various problems related to major depression have emerged and how to minimize confusion related to these problems.

Shigenobu Kanba received: grant/research support from Pfizer, Ono, GlaxoSmithKline, Astellas, Janssen, Yoshitomiyakuhin, Eli Lilly, Otsuka, Dainippon Sumitomo, Meiji Seika Pharma, Kyowa Hakko Kirin and Shionogi; and honoraria from Pfizer, Janssen, GlaxoSmithKline, Eli Lilly, Eisai, Meiji Seika Pharma, Taisho Toyama, Astellas, Otsuka, Shionogi, Dainippon Sumitomo, Kyowa Hakko Kirin, Yoshitomiyakuhin, MSD and Wyeth. Keisuke Motomura received honoraria from GlaxoSmithKline, Mochida and Eli Lilly.


  1. Top of page
  2. References