The research questions in controlled trials involve interventions and controls. This report included a few studies comparing Kampo medicines, using several Kampo medicine groups as intervention or evaluating the system of Kampo medicine itself.

Further studies of this kind using Kampo formulations are desired.

In Kampo medicine in Japan, not only Kampo formulations but also decoctions are sometimes used. RCTs using decoctions are expected in future, but were listed as excluded references in this report. When a certain number of references are accumulated, their structured abstracts will be prepared.

For RCTs of crude drug therapies not based on Kampo formulae, or non-pharmacological traditional medicines such as acupuncture, preparation of structured abstracts and accessible forms of providing them are desired. Cooperation with other related organizations will also be a future challenge. In Japan, although not limited to traditional medicine, Minds (http://minds.jcqhc.or.jp/) has prepared and published structured abstracts free of charge.

As traditional medicines have already been on the international market, Japanese Kampo formulations are slightly lagging behind this worldwide trend. If high-quality structured abstracts of Chinese and Korean traditional medicine formulations are prepared, controversy could be reduced in the medical guidelines to be prepared by the WHO Regional Office for the Western Pacific (WPRO) and other situations.

This is also the case with abovementioned acupuncture, etc., and the FACT has already launched such a project.

The use of universal style and quality standardization would be key to preparation of structured abstracts of RCTs in the area of traditional medicine.

With the aim of improving the quality of references of RCTs, the CONSORT statement was published in 1996 and revised in 2001 (http://www.consort-statement.org/).

This consists of a total of 22 items, and authors are requested to attach a checklist showing on which page the information on each item can be found. They are also requested to attach a flowchart representing subject disposition in case results might differ depending on the size of the analysis population. These requirements not only control the quality of RCT articles but also improve the quality RCT themselves. The JSOM has also added the statement that “RCT papers shall conform to revised CONSORT statement (2001)” to the contribution rule for its journal since the March 2008 revision (Nihon Toyo Igaku Zasshi [Kampo Medicine] 2008; 59: 580-89).

Two herbal extensions of the CONSORT statement have been published:

Gagnier JJ, Boon H, Rochon P, Moher D, Barnes J, et al. Reporting randomized controlled trials of herbal interventions: An elaborated CONSORT statement. Annals of Internal Medicine 2006; 144(5): 364-7. (Available in Japanese: Okabe T, Tsutani K [trans], Habu kainyu no randamuka hikakusiken hokoku: shosai na CONSORT seimei. In: Nakayama T, Tsutani K [supervise-eds] Rinshokenkyu to Ekigakukenkyu notameno Kokusairuru-shu (International Rulebook for Clinical and Epidemiological Studies. Tokyo: Life Science Publishing Co., Ltd.; 2008: 156-63）

Bian ZX, Moher D, Dagenais S, Li YP, Wu TX, et al. Improving the quality of randomized controlled trials in Chinese herbal medicine, Part IV: applying a revised CONSORT checklist to measure reporting quality. Journal of Chinese Integrative Medicine (Zhong Xi Yi Jie He Xue Bao) 2006; 4(3): 233-42. (Available in Japanese: Tsutani K, Araki S [trans]. Chuyaku no randamuka Hikakushiken no Hokoku ni kansuru CONSORT seimei. In: Nakayama T, Tsutani K [supervise-eds], Rinshokenkyu to Ekigakukenkyu notameno Kokusairuru-shu [International Rulebook for Clinical and Epidemiological Studies]. Tokyo: Life Science Publishing Co., Ltd.; 2008: 164-9).

The former applies to plain herbs, while the latter to Chinese prescriptions. The former provides detailed explanation on how to describe “intervention” in consideration of the nature of crude drugs, and the latter gives consideration to the diagnosis system of Chinese traditional medicine and years of clinical experience. In the third phase of the EBM Special Committee, the Task Force for Kampo CONSORT (KC-TF) was established for preparation of the Kampo version of the CONSORT statement, which is now underway.

A survey of the references compiled as structured abstracts this time for compliance with the CONSORT statement demonstrated that few were good in quality. Particularly, the following defects were commonly noted: no indication that the study is an RCT in title/abstract; no indication of the study site or period; no indication of the name of the manufacturer of Kampo formulation or the daily dosing frequency; no indication of the methods and assurance of randomization; unclear numbers of enrolled patients, assigned patients, and analyzed patients; no indication of adverse events in the control group. In the future, Kampo RCTs will also be required to be reported in compliance with the CONSORT statement. These findings were presented in the following meeting and published in Nihon Toyo Igaku Zasshi (Kampo Medicine) as study results.

Okabe T, Arai I, Tsutani K. Ebidensu repoto purojekuto: autorain to Kampo RCT no shitsu hyoka (Evidence Report Project [1] Outline and evaluation of Kampo RCTs). The Japan Society for Oriental Medicine 60th Annual Meeting Forum “Kampo no ebidensu wo tsutaeru (Transfer Kampo evidence),” 21 Jun. 2009, Tokyo. Nihon Toyo Igaku Zasshi (Kampo Medicine) 2009; 60 suppl.: 160 (in Japanese).

The third version of the CONSORT statement was published in March 2010. The CONSORT statement originally limited to RCTs when it was published in 1996, but thereafter extended to include various study designs such as epidemiological studies and systematic reviews, and further studies on above-mentioned complementary and alternative medicine (CAM). In this context, the “equator network” was established in June 2008 to cover the publication guidelines for all these studies for enhanced accessibility and to help prepare such guidelines in future [http://www.equator-network.org ].

The Japanese version of these guidelines are included in the following:

Nakayama T, Tsutani K (eds). Rinsho kenkyu to ekigaku kenkyu notameno kokusai ruru-shu (International Rulebook for Clinical and Epidemiological Studies). Tokyo: Life Science Publishing Co., Ltd.; 2008 (in Japanese).

In the Declaration of Helsinki revised in October 2008, a sentence “Every clinical trial must be registered in a publicly accessible database before recruitment of the first subject” was added in section 19. However, this requirement is not well known. Therefore, the discussion here includes the historical background underlying the implementation of this requirement.

Awareness of the clinical trial registry (CTR) has increased since the 1990s, when evidence-based medicine (EBM) emerged. In particular, the problems became clearer after the Cochrane Collaboration, which plays a role in the EBM information infrastructure, was established in 1992 to fully implement systematic review (SR). SR is almost synonymous with meta-analysis (MA).

However, no matter how exhaustive the survey or search, how careful the assessment of quality, and how sophisticated the statistical method used for data consolidation, a problem of bias ("publication bias") arises when studies are not reported. This leads to flaws in decision making by health care providers, policy makers, medical consumers, etc. As a result, ineffective therapies, hazardous therapies, and therapies with poor cost-effectiveness are “used”.

Table 5 gives an example of publication bias in the area of acupuncture. Search for papers with an abstract published between 1966 and 1995 by the Medline identified 108 of 109 papers published in China (99%) that have shown favorable results, that is, demonstrated the efficacy of acupuncture compared with control. The efficacy rate was 75% in England and similar to that in China in other countries including Japan. This has been attributed to failure to publish studies that do not show efficacy.

This situation was widely acknowledged by those involved in systematic review of these studies. Although some measures might avoid this bias, such as encouraging researchers to publish all studies, passing legislation mandating registry of all trials, and establishing a website to register planned or ongoing clinical trials, no specific measures were fully established.

In 2003, the National Library of Medicine (NLM) of the National Institute of Health (NIH) established "Clinical Trials.gov" (http://clinicaltrials.gov/ct2/info/about) with the aim of encouraging patients to access information on clinical trials of new drugs (therapies) intended for life-threatening disease. This is also an aim of the 1997 US-FDA Modernization Act. This system was not intended to avoid publication bias, but partially did so for a limited number of diseases including cancer, AIDS, and Alzheimer's disease.

The Glaxo SmithKline fraud scandal reported on the front page of the New York Times dated 3 Jun. 2004 triggered a worldwide reaction. In a clinical trial of an anti-depressant in children, the company failed to properly report attempted suicide as an adverse event. The scandal prompted worldwide support for legislation requiring clinical trial registration and raised ethical issues such as the risk adverse events in similar clinical trials and abuse of the altruistic goodwill of participants.

In September 2004, the International Committee of Medical Journal Editors (ICMJE) issued a statement that no manuscript would be accepted in advance of registration of the clinical trial. Trials already in progress were given additional time to comply. The Cochran Colloquium in October 2004 issued the "Ottawa Statement." The WHO also supported these statements by holding the “WHO Technical Consultation on Clinical Trial Registration Standard Meeting” in its headquarters in Geneva in April 2005 to determine 20 items to be registered, etc.

This trend continued between 2004 and 2005 as indicated below:

(1) Rinshoshiken no toroku to kekka no kokai (Clinical trial registry and publication of the results [regardless of whether or not results are positive or negative]). The 25th Annual Meeting of the Japanese Society of Clinical Pharmacology and Therapeutics Symposium 12 (18 Sept. 2004, Shizuoka). Rinsho Iyaku (Journal of Clinical Therapeutics Medicine) 2005; 21(1): 3-62.

(2) Rinshoshiken toroku nikansuru “Otawa seimei” to “Junebu kaigi” no doko (“Trend toward clinical trial registry as advocated by the "Ottawa Statement" and "Geneva Conference."). Yakuri to Chiryo (Japanese Pharmacology & Therapeutics) 2005; 33(6): 543-66.
　　　http://www.lifescience.co.jp/yk/jpt_online/ottawa/index_ottawa.html

(3) UMIN Clinical Trial Registry System Symposium (2 Feb. 2005)
　　　http://www.umin.ac.jp/ctr/symposium20050202.htm

In Japan, the University hospital Medical Information Network Clinical Trials Registry (UMIN-CTR) (June 2005-), the Japan Pharmaceutical Information Center Clinical Trials Information (JAPIC-CTI) (July 2005-), and Japan Medical Association Center for Clinical Trials (JMA CCT) system (December 2005-) were launched. Subsequently, the Japan Primary Registries Network (JPRN) (October 2008-) of the National Health Science Institute consolidated these three systems.

In 2007, the WHO established the International Clinical Trials Registry Platform (ICTRP, http://www.who.int/ictrp/en/) and opened a search portal (http://apps.who.int/
trialsearch/ ) to facilitate access to all systems worldwide for searching.

Currently this site can be used to search for all registered trials, including trials of Kampo formulations in Japan.

In April 2007, interventional clinical studies supported by Health and Labour Sciences Research Grants were required to be registered. Also, in April 2009, the Ministry of Health, Labour, and Welfare “Ethical Guidelines for Clinical Research” required registry.

The intent of clinical trial registry has been discussed above from a historical perspective in some detail. Questions remain: how many RCTs and clinical studies are performed and what percentage of them is registered in Japan?

In a book summarizing the situation in various nations in the world released in 2006, it was estimated that 400 RCTs and 2,300 other clinical researches, for a total of 2,700, were performed at that time.

Matsuba H, Kiuchi T, Tsutani K, Uchida E, Ohashi Y. The Japanese perspective on registries and a review of clinical trial process in Japan. In: Foote M ed. Clinical trial registries: A practical guide for sponsors and researchers of medicinal product. Basel: Brikhäuser, 2006

According to the book, as of February 2006, the three systems combined had a total of 550 registries. UMIN-CTR was the first system in Japan, started in June 2005. A review in May 2010 found that the number of registries in 2006 was 259, 138 and 7 registries in UMIN-CTR, JAPIC-CTI and JMACCT, respectively. The total of 404 registries represented approx. 15% of all clinical studies (404/2,700).

A review showed that the number of registries in 2009 had risen to 1,311, 295 and 9, respectively, for a total of 1,615. Assuming that the the number of clinical studies remained the same in 2009, i.e., approx. 2,700, registered studies accounted for approx. 60% (1,615/2,700) of all clinical studies

The problem of publication bias was described above, citing the field of acupuncture in the period from 1966 to 1995 as an example. Is there currently publication bias in the reporting of studies on Kampo formulations?

Preliminary analysis of all 345 articles in this compilation of Evidence-based Reports revealed “favoring test treatment” results in 301 out of 308 trials (98%) (excluding those trials comparing only Kampo formulations with other Kampo formulation(s) as control). This result is consistent with the above-mentioned rate of publication bias in the reporting of acupuncture studies. In Japan, inasmuch as the percentage of registered clinical trials is increasing, it is hoped that clinical trials of Kampo formulations are also being registered so as to provide true efficacy and safety evidence.

The global trend is toward requiring not only registry of ongoing clinical trials but also publication of the results. NIH-funded trials must fulfill these requirements. The situation relating to the publication of results from trials of Kampo formulations registered in Japan is described in the following article:

Arai I, Tsutani K. Kampoyaku no rinshoshikentoroku to kekka no kohyojokyo (Situation of registry of clinical trials of Kampo medicine and publication of the results). The Japan Society for Oriental Medicine 61st Annual Meeting (5 Jun. 2010, Nagoya), Nihon Toyo Igaku Zasshi [Kampo Medicine] 2010; 61 suppl.: 244.

Publication bias is considered to be more of a problem in CAM than in conventional allopathic medicine. This is because CAM studies lack an administrative framework, as many of these studies are conducted by physicians who are unfamiliar with clinical trial registry trends. This is in contrast to clinical trials conducted by pharmaceutical companies.

The intent of this “Evidence Report” project, involving compilation of structured abstracts with comments about RCTs in Japan, is to “transfer” users and reduce bias. However, publication bias cannot be prevented if the results of RCTs are not published.

We strongly recommend registration of all RCTs of Kampo formulations in Japan in the future.