Current status of research on cognitive therapy/cognitive behavior therapy in Japan


Yutaka Ono, MD, Center for Stress Management, Keio University Health Center, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. Email:


Cognitive therapy/cognitive behavior therapy was introduced into the field of psychiatry in the late 1980s in Japan, and the Japanese Association for Cognitive Therapy (JACT), founded in 2004, now has more than 1500 members. Along with such progress, awareness of the effectiveness of cognitive therapy/cognitive behavioral therapy has spread, not only among professionals and academics but also to the public. The Study Group of the Procedures and Effectiveness of Psychotherapy, funded by the Ministry of Health, Labor and Welfare, has conducted a series of studies on the effectiveness of cognitive therapy/cognitive behavior therapy since 2006 and shown that it is feasible for Japanese patients. As a result, in April 2010 cognitive therapy/cognitive behavior therapy for mood disorders was added to the national health insurance scheme in Japan. This marked a milestone in Japan's psychiatric care, where pharmacotherapy has historically been more common. In this article the authors review research on cognitive therapy/cognitive behavior therapy in Japan.

COGNITIVE THERAPY WAS developed by a US psychiatrist, Aaron T. Beck, to treat depressive disorders and other mental disorders, and has been practiced for over 30–40 years in Western countries. In Japan it was only in the late 1980s that cognitive therapy/cognitive behavior therapy (cognitive therapy/CBT) was introduced into the field of psychiatry. In 1989 the inaugural National Conference of Cognitive Therapy was held and, around 1998, the first issue of its quarterly newsletter, Cognitive Therapy News, was published, and has continued regularly since then.

In 2001 the Japanese Association for Cognitive Therapy (JACT) was founded, and their first conference was held, followed by annual meetings thereafter. In July 2004 the World Congress of Behavioral and Cognitive Therapies (WCBCT) was held in Kobe, and a great number of people gathered together for this international meeting. From then on, interest in cognitive therapy/CBT has expanded, and JACT now has more than 1500 members. Since 2008 the Journal of Cognitive Therapy, the Japanese official journal of JACT, has been issued annually. Along with such progress, awareness of the effectiveness of cognitive therapy/CBT has spread, not only among professionals and academics but also to the public.

Nonetheless, the number of medical facilities capable of providing cognitive therapy/CBT is still limited. According to the data released in 2006 by the Study Group of the Procedures and Effectiveness of Psychotherapy, funded by the Ministry of Health, Labor and Welfare, only 28% of medical facilities reported being capable of conducting any form of psychotherapy in a satisfactory manner.1 The reasons for this include a shortage of personnel who are competent to conduct psychotherapy, insufficient evidence of remuneration for medical services, and so forth. The Study Group indicated that, among various forms of psychotherapies, cognitive therapy/CBT was ranked first as the treatment method that medical facilities would like to provide. The same study group also conducted a pilot study of the cost-effectiveness of depression treatment, and reported that although combination treatment required more direct expenses compared to pharmacotherapy, it is highly effective and thus considered sufficiently cost-effective.

As a result, in April, 2010, cognitive therapy/CBT for mood disorders was added to the national health insurance scheme in Japan. This marked a milestone in Japan's psychiatric care, where pharmacotherapy has historically been more common. It is now being used not only to boost research using randomized controlled trials (RCT), but also training and practice in the field.

In this article the authors have reviewed research on cognitive therapy/CBT in Japan, most of which has been published in Japanese.



In order to get a grasp of the current picture of cognitive therapy/CBT for depressive disorders, Web searches were performed using the Web database of Japanese medical journals as well as the Health and Labour Sciences Research database using ‘cognitive therapy or cognitive behavior therapy’ and ‘depression’ in Japanese as key words. A total of 485 titles of original articles, review articles, tutorial articles, and case reports from 1983 to 2009 (392 titles after excluding conference minutes) were obtained. Additionally, a different search was conducted on Medline using ‘cognitive therapy’, ‘Japan’, and ‘depression’ as key words.

After first excluding conference minutes then reviewing the papers' contents, it was determined that 13 were either case series studies or case–control studies, and 27 were case reports.

In those case series or case–control studies, the main effects of the treatment were found to be improvement in the subjective/objective depressive symptoms, quality of life, wellbeing, social functioning (e.g. return to work), and dysfunctional thoughts. Furthermore, because the therapy settings in those studies varied from university hospitals, general hospitals (inpatient, outpatient), psychiatric hospitals, and psychiatric clinics to day-care programs in health-care centers and assistance to return to work services, these overall results regarding a variety of clinical settings suggest that the effectiveness of cognitive therapy is relatively universal.

The first randomized controlled trial using this subject group in Japan was carried out by the Study Group for the Procedures and Effectiveness of Psychotherapy chaired by Yutaka Ono, funded by the Ministry of Health, Labor and Welfare, in 2004–2006. An open trial was conducted of individual CBT following the manual, with depressive patients. The results (Table 1) suggested that the manualized treatment, which consists of a 16-week individual cognitive therapy/CBT program for major depression, appears feasible and may achieve favorable treatment outcomes among Japanese patients with major depression.2 Furthermore, a new RCT (UMIN000001218) has been under way since 2007, and the outcomes are anticipated.

Table 1.  Outcome measures of major depression2
Outcome measuresBaselineEnd-point% improvementCohen's d
  • *

    P < 0.001.

  • Intent-to-treat sample, last-observation-carried forward analysis.

  • BDI-II, Beck Depression Inventory-II score; DAS, Dysfunctional Attitude Scale; GAF, Global Assessment of Functioning of DSM-IV; HAMD, Hamilton Depression Rating Scale; QIDS-SR, Quick Inventory of Depressive Symptomatology–Self Rated; SUBI, WHO Subjective Wellbeing Inventory.

SUBI – health subscale32.69.538.3*
SUBI – fatigue subscale38.46.446.4*5.420.81.35

The same study group also performed functional brain imaging tests before and after the group CBT for drug-treatment-resistant depression, and found that the effects of treatment may be related to the medial prefrontal cortex and the ventral anterior cingulate cortex during self-referential processing of negative stimuli.3–7

In addition, Watanabe et al. also investigated the value of adding brief behavioral therapy to usual clinical care for insomnia over treatment as usual for residual depression and refractory insomnia, and found that it produced statistically significant and clinically substantive added benefits.8

Anxiety disorder

Panic disorder

Randomized controlled trials for panic disorder have not yet been done in Japan, but Kobayashi et al. conducted group CBT (83 of 90 completed) using the Japanese version of the self-reported Panic Disorder Severity Scale (PDSS), and found that 43.3% of the subjects reported that the severity decreased by more than 40%.9 Among the subjects with active panic disorder (whose PDSS scores were >1.1), more than 50% reported that the severity decreased by more than 40%. Sakai et al. found that, using individual CBT, 12 of 14 patients reported that the severity on PDSS decreased by more than 50%.10 Similarly, Nakano et al. found that 44.7% of the subjects who underwent group CBT (56 of 70 completed) reported that the severity on PDSS decreased by more than 40% (Table 2).11Table 2 also shows that the pre-treatment scores were significantly higher than, but the post-treatment and 3-month follow-up scores were not significantly different from, the 1-year follow-up data for the Fear Questionnaire Agoraphobia subscale, Mobility Inventory Accompanied, Mobility Inventory Alone, Agoraphobia Cognitions Questionnaire, Body Sensations Questionnaire, and Work, Home and Leisure scale.

Table 2.  Outcome measures of panic disorder11
 Before treatmentAfter treatment3-month follow up1-year follow up
  • n = 39.

  • n = 37.

  • ACQ, Agoraphobia Cognitions Questionnaire; BSQ, Body Sensations Questionnaire; FQ-A, Fear Questionnaire Agoraphobia subscale; MI AA, Mobility Inventory Alone; MI ACC, Mobility Inventory Accompanied; WHL, Work, Home and Leisure scale.


Nakano et al. suggested that group CBT for panic disorder, originally developed in Western countries, is equally as applicable to Japanese patients with panic disorder as to Western patients.11 For example, the dropout rate in their study was 20%, while that in the Wade et al., Martinsen et al., Hahlweg et al., and Barlow et al. studies was 26%, 14%, 22%, and 38%, respectively.12–15 Furthermore, as Table 2 shows, their program reduced the FQ-agoraphobia subscale by 53%, while the program at a community mental health center in the USA reduced the subscale score by 47%. Their sample had a mean 47% reduction in the Agoraphobia Cognitions Questionnaire score, while the group CBT program in Norway and that in Germany reported similar reduction rates of 41% and 46%, respectively.13,14 Nakano et al. also reported effect sizes of 0.76 and 0.81, respectively, while those in the Perini et al. study in Sydney were 0.78 and 0.76.16

Given these findings, CBT for panic disorder in Japan yields comparable results to that reported in other countries. Cognitive therapy performed in these research settings consisted of 10–15 sessions in the form of either individual therapy or group therapy, following the Practice Guideline (June 2009) of the American Psychiatric Association (APA). During treatment, trained therapists carried out psychoeducation, self-monitoring, breathing retraining, cognitive restructuring of anxiety-provoking automatic thoughts, exposure to phobic situations and internal sensations, modification of anxiety-sustaining behaviors, CBT package with prevention of recurrence, and so on. Those results indicate that the outcomes of treatment in Japan are comparative to the top-level results in other countries.

Moreover, among the study groups of the same ministry, Furukawa and his group conducted manualized group CBT for 187 outpatients with panic disorder (with and without agoraphobia).17–26 In the Nakano et al.11 study 14 patients (20%) out of 70 patients dropped out of the program. For the remainder, the total scores on PDSS were found to decrease from 12.8 to 7.1 (a 44.7% decrease) on average. This effect had been maintained >12 months after the completion of treatment (Table 2).

When prognostic factors were explored on multiple regression analysis after controlling for the severity, the duration of disorder and social dysfunction were found to be predictors of this disorder.

These studies suggest that group CBT programs for panic disorder for Japanese patients have comparable outcomes to those of patients in Western countries.

Social anxiety disorder

Again, RCT of CBT for social anxiety disorder have not yet been done in Japan. Furukawa and his group have been conducting a cohort study of group CBT (uncontrolled open trial) at Nagoya City University since 2003.25–27 From these studies, Chen et al. suggested that, irrespective of cultural background, the group CBT program appears to be equally acceptable to Japanese patients with social anxiety disorder as to Western patients.25

In one of those studies, 57 outpatients with social anxiety disorder were evaluated to measure various aspects of social anxiety disorder symptomatology at the beginning and at the end of the program of group CBT.25 The dropout rate was 12.3%, which was lower than that reported in the Heimberg et al. (22.2%) and Stangier et al. studies (15.4%).28,29

Furthermore, as Table 3 shows, for the intention-to-treat (ITT) sample, the percentage reduction in questionnaire scores changed from 20% to 30% from before to after treatment, and the pre–post-treatment effect sizes ranged from 0.37 to 1.01. Among the completers, the respective figures were from 20% to 33% and 0.41 to 1.19. These data on effectiveness compare favorably with those of Western studies.28–30

Table 3.  Effect sizes for group CBT for social anxiety disorder25
StudyCompleters analysisITT
Heimberg et al. (1998)28Stangier et al. (2003)29Chen et al. (2007)25Mortberg (2007)30Chen et al. (2010)27
Group12 weeks15 weeks16 weeks3 weeks16 weeks
Dropout (%)22.215.412.325.712.3
  1. CBT, cognitive behavior therapy; ES, effect size; FNE, Fear of Negative Evaluation Scale; FQ-sp, Fear Questionnaire Social Phobia Subscale; ITT, intention to treat; LSAS, Liebowitz Social Anxiety Scale; SIAS, Social Interaction Anxiety Scale; SPS, Social Phobia Scale.

Social fear15.112.00.44
 SD7.04.9    6.87.2    6.97.2 
Social avoidance13.19.90.42   18.414.20.53
 SD7.74.8    7.88.5    7.98.5 
Performance fear17.013.40.65   20.915.90.83   20.816.30.78
 SD5.63.8    6.17.2    5.86.9 
Performance avoidance13.810.90.47   17.311.50.71
 SD6.04.3    8.38.0    8.27.9 
 SD7.24.0    6.28.1 6.77.1 6.27.8 
 SD17.010.0 15.315.3 15.415.3 13.811.5 15.415.2 
 SD18.413.2 14.012.4 14.815.6 16.114.7 15.016.0 
FNE22.321.40.12   24.320.80.6822.521.50.1524.622.10.49
 SD7.64.3    5.26.5 6.57.2 5.16.5 

Obsessive–compulsive disorder

There are, similarly, only a few controlled studies for obsessive–compulsive disorder in Japan. Only eight titles were found between 1983 and 2009 using key word searches combining ‘obsessive–compulsive disorder (neurosis)’, ‘(cognitive) behavioral therapy’, ‘effects of treatment (outcome)’ of the Web database of Japanese medical journals. First, regarding RCT, the only research published was the one by a behavior therapy study group at Kyushu University. In that study comparing the efficacy of behavior therapy and fluvoxamine, the subjects were randomly assigned to one of three groups: (i) behavior therapy + pill placebo; (ii) fluvoxamine + autogenic training (a psychological placebo); and (iii) autogenic training + pill placebo. The authors reported that 28 patients completed the study. Patients in the behavior therapy and fluvoxamine groups had significantly more improvement than those in the control group on mean total Y-BOCS score. Moreover, the behavior therapy group had significantly more reduction in total Y-BOCS score at the end of treatment than the fluvoxamine group (behavior therapy > fluvoxamine, P < 0.01). Patients with lower baseline total Y-BOCS, past history of a major depressive episode and absence of cleaning compulsion improved more on fluvoxamine (Table 4).31,32 There is no other controlled study, but the team at National Sanatorium Kikuchi Keifuen conducted a longitudinal study for obsessive–compulsive disorder and found that for the prognosis of obsessive–compulsive disorder, behavior therapy is the strongest indicator of remission.33 In a survey by the team at Hizen Psychiatric Center of the effects of treatment for inpatients with obsessive–compulsive disorder, in most cases behavior therapy and pharmacotherapy were combined, and improvement was seen in approximately 90% of cases.34 The rest of the search results were all individual case reports.

Table 4.  Outcome measures of obsessive–compulsive disorder31
 BT (n = 10)FLV (n = 10)Control (n = 8)
12 weeks before treatment12 weeks after treatmentt12 weeks before treatment12 weeks after treatmentt12 weeks before treatment12 weeks after treatmentt
  • **

    P < 0.01,

  • ****

    P < 0.0001 (paired t-test).

  • BT, behavioral therapy; FLV, Fluvoxamine; Y-BOCS: Yale–Brown Obsessive Compulsive Scale.


Post-Traumatic Stress Disorder

In psychiatry in Japan, much attention was given to post-traumatic stress disorder (PTSD) after the Great Hanshin-Awaji Earthquake and the Tokyo subway sarin gas attack, both of which took place in 1995. For psychotherapy in Japan, searches on the largest web database of Japanese academic journals using ‘PTSD, psychotherapy’ as key words for the range 1983–2008, found 590 titles. Narrowing the results down to articles and conference minutes describing the effects of before and after treatments based on objective assessment, seven titles were found (five on eye movement desensitization and reprocessing, and two on prolonged exposure).

The articles on eye movement desensitization and reprocessing were all case reports using one patient. Regarding research on prolonged exposure, Yoshida et al. and Asukai et al. both carried out uncontrolled open trials.35,36 Both studies found significant decreases in Clinician-Administered PTSD Scale for DSM-IV (CAPS) and Impact of Events Scale–Revised (IES-R) scores. The average rates of decrease in the CAPS and IES-R scores were 42.7% and 50%, respectively, in the Yoshida et al. study, and 77.9% and 74.0%, respectively, in the Asukai et al. study, indicating prominent improvement in PTSD. Moreover, the rates of recovery were 42.9% in the Yoshida et al. study, and 80% immediately after the termination of therapy and 100% at 6 months after therapy in the Asukai et al. study. Given that the average rates of decrease among the subjects who completed the treatment in the Foa et al. study was 72.0% (PTSD symptom scale interview of the subjects who completed),37 and 31.8% in the Schnurr et al. study (CAPS, ITT),38 despite the differences in the methods of measurement and analysis, the effects of treatment in Japan are considered equivalent to the ones overseas.

Regarding long-term prognosis, Asukai et al. found decreases in both the CAPS and IES-R scores immediately after the termination of therapy as well as at 3 months later and 6 months later, suggesting that the outcomes are also similar to the results of overseas studies. The dropout rates in two Japanese studies were 7.1%35 and 16.7%,36 while in the Foa et al. study37 it was 34.2%, and in the Schnurr et al. study38 it was 37.6%. At this point the dropout rates in Japan are relatively low but, because the sample sizes were very small in both the Japanese studies, further studies with larger sample sizes are needed. Overall, the findings of the open trials in Japan, although obtained with small samples, imply the possible effectiveness of prolonged exposure in this patient group in Japan.

Personality disorders

Ishii et al. conducted an open trial of dialectical behavior therapy (DBT) at a psychiatric hospital of 22 clinically referred women (mean age, 29.9 ± 5.9 years) with recent suicidal and self-injurious behaviors (e.g. strangling, hanging, and jumping from a high place) meeting DSM-IV criteria.39,40 The patients were assigned to either the 3-month program (13 patients) or the 6-month program (9 patients). The effectiveness of both programs was not significantly different. In that pilot study 22 female patients were compared at admission and at 1 month after discharge with respect to psychopathology and frequency of self-injuries, and a highly significant improvement in depressive symptoms was found on the Beck Depression Inventory-II (P < 0.000).39 With regard to improvement of overall mental health, significant to highly significant results were seen on the 90-item Short Checklist–Revised (SCL-90-R; P < 0.005). A highly significant decrease in the number of suicidal and self-injurious behaviors was also reported (F1,21 = 56.5, P < 0.001, ES = 1.25). DBT appears to be effective in reducing suicidal and self-injurious behaviors.


In Western countries the effectiveness of psychotherapy including cognitive therapy/CBT has been clearly shown, and it is suggested that the same applies to Japan with regard to depression and anxiety disorders following uniform procedures. The findings described here, however, are mostly from case reports or open trials, and in order to prove the effectiveness more well-controlled RCT with larger sample sizes are needed.

The quantity of outcome research in cognitive therapy, particularly RCT, is exceptionally low in Japan. This is not surprising when we consider the sociocultural aspects of Japan's psychiatric care. First, the language barrier makes it difficult for Japanese people to master Western psychotherapy, and this hinders the introduction of not only cognitive therapy/CBT, but psychotherapy in general. Cognitive therapy/CBT is used in the area of depressive disorders, but other forms of psychotherapy, such as interpersonal psychotherapy, have been rarely introduced into practice despite evidence of effectiveness reported in many parts of the world.

Second, an even greater hindrance comes from the conservativeness of Japanese society and the particular style of interpersonal relationships among Japanese people. Japanese psychiatry has its roots in descriptive psychiatry from Germany, thus psychiatrists were interested only in psychopathology, and psychotherapy did not generate much interest from the beginning. Further, as aforementioned, the language poses an obstacle and makes it a considerable challenge for Japanese practitioners when attaining clinical credentials in the USA, and the number of psychiatrists who have received clinical training abroad is very limited. This explains why the importation of foreign knowledge has considerably lagged in clinical medicine, and especially in psychotherapy, when compared to basic medicine.

Psychotherapy is not alone in the lack of RCT research. These types of clinical studies have seldom been conducted even in the field of pharmacotherapy, except for clinical trials for new drugs in the pharmaceutical industry. With regard to pharmacotherapy for depression, a large epidemiological research project was started only in 2010 to examine its effectiveness. In this sense, the Nakatani et al. study, in which the efficacy of behavioral therapy for obsessive–compulsive disorder was evaluated against control groups,31 was groundbreaking work in the Japanese psychiatric field, and the launch of RCT research in cognitive therapy/CBT for depression is also worth noting. Paradoxically, it would appear that the slow dissemination of psychotherapy encouraged professionals to prove its efficacy, and led to initiation of empirical research to scientifically demonstrate its effectiveness.

Such a substantial lag in clinical research is also related to paternalistic, dependent interpersonal relations that characterize Japanese society. In other words, the relationship style in which the specialists assume absolute authority and the patients take them at their word has affected the functions of psychiatric practice, which in part has impeded its progress. Fortunately, this relationship style has greatly shifted in this advanced information society, in which users come to recognize the importance of empirical evidence, while the mass media raises public awareness. As people learn more about evidence in cognitive therapy/CBT abroad, the demand for this treatment has increased. In turn this growing demand further enhances the specialists' focus on outcome studies. The study group of the Ministry of Health, Labor and Welfare, mentioned earlier, has led research efforts in this area since 2006.

It is particularly noteworthy that this line of study has demonstrated the effects of cognitive therapy/CBT in Japan in the face of its sociocultural distinctiveness. The usefulness of psychotherapy in the socioculturally different contexts have been much debated, and yet its effects had never been empirically examined. Meanwhile, the reason why cognitive therapy/CBT was able to take this critical step ahead of others is related to the fact that cognitive therapy is evidence-based psychotherapy utilizing reality-oriented techniques. From the recent findings, this therapeutic approach, which understands and deals with pathological conditions from an information-processing perspective, is confirmed to be universal. Yet the sociocultural elements inevitably influence the styles and functions of information processing as well as the interpersonal relationships, and we should continuously seek ways to adjust clinical techniques and interventions in the light of diversity.


This review was supported by a Grant-in-Aid for Scientific Research from the Japanese Ministry of Health, Labor and Welfare to YO. The authors thank the members of the Study Group of the Procedures and Effectiveness of Psychotherapy.