3.1. Document type and language of publication
There were 6,004 publications with 13 document types indexed in the SCI-Expanded, which include 3,975 articles. The article, as the most popular document type, comprises 66% of the total pro-duction and was followed distantly by reviews (508; 8.5%), letters (458; 7.6%), meeting abstracts (457; 7.6%), editorial materials (286;
4.8%), proceedings papers (213; 3.5%), and the remainder having less significance were news items (38), corrections (32), notes (24), reprints (6), book reviews (4), biographical-items (2), and addition correction (1). Journal articles represented the majority of doc-ument types, therefore only the 3,975 articles were analyzed in further study.
There were 15 languages in use. English, as the most popular language, comprises 93% of the total articles and was followed by German (168; 4.2%). Some other languages that were less used were Russian (50) and French (22). Still some other languages gener-ally less appeared were Portuguese (10), Chinese (8), Spanish (8), Japanese (3), Hungarian (2), Korean (2), Turkish (2), Polish (2), and one for Czech, Slovene, and Ukrainian, respectively.
3.2. Publication patterns
The total amounts of SCI-Expanded journals’ publications including searching words “electroacupuncture”, “electro-acupuncture”, “acupoint*”, “acupunctur*”, and “percutaneous electrical nerve stimulation” in titles only from 1900 were counted and displayed inFig. 1. As a whole, the development of acupuncture literature can be divided into two phases in terms of the increase in the number of SCI-Expanded journal papers. The first phase is between 1973 and 1997, in which number of articles fluctuated to form a plateau, averaged 85 per year. The number of articles showed a dramatic increase of 40% in 1998. The extrapolation from Fig. 1suggests that the development in this field will continue in the fast growing tract after 2009. According to a limitation of the SCI-Expanded database, searching keywords in titles, keywords, and abstracts can only obtain for publications since 1991. A development trend is also found for all documents and articles, searched by titles, keywords, and abstracts (Fig. 1). There is an increase in all documents from 2 in 1921 to 703 in 2009 and an increase of articles from 1 in 1921 to 451 in 2009.
Fig. 1. Number of SCI-Expanded journals’ papers referring to “electroacupuncture”,
“electro-acupuncture”, “acupoint*”, “acupunctur*”, and “percutaneous electrical nerve stimulation” since 1900.
3.3. Subject categories and journals
Based on the classification of subject categories in JCR in 2009, the article output data of acupuncture research was distributed in 117 SCI subject categories. Top four main subject categories were integrative & complementary medicine (976; 25%), neurosciences (750; 19%), general & internal medicine (555; 14%), and clinical neurology (504; 13%).
In total, 3,975 articles were published in 927 journals includ-ing specialty journals and journals of other disciplines. Out of the 927 journals, 487 (53%) journals contained only 1 article and 151 (16%) contained two. One-third of all articles were published in 14 core journals (Table 1). Journal of Alternative and Complementary Medicine published the most acupuncture articles (258), followed by American Journal of Chinese Medicine (222), Acupuncture &
Electro-Therapeutics Research (176), American Journal of Acupuncture (141), and Neuroscience Letters (111). However, American Journal of Acupuncture was not listed in JCR since 1996. The CA-A Cancer Jour-nal for Clinicians, which ranked not only the top one jourJour-nal in the ISI category of oncology but also in all 7,347 journals listed in JCR in 2009, with the IF as 87.925, was the leading journal on acupuncture research.
3.4. Countries/territories’ research performances
The contribution of different countries/territories is estimated by the location of the affiliation of at least one author of the published articles. There were 110 (1.8%) articles without author address information in the ISI. For this reason, not all of the articles were included in this analysis. Total number of articles, independent, and collaborative articles have been applied to be indicators to evaluate countries/territories research performance (Chiu and Ho, 2007). In recent years, five indicators including the total, independent, collaborative, first author, and corresponding author articles, were also used for countries/territories research performance (Malarvizhi et al., 2010).Table 2shows the top 11 countries/territories which published at least 100 articles ranked by the number of total articles with the five indicators. Among the 3,865 articles with author address information, international col-laborations accounted for 14% of the articles compared to 86% from
Fig. 2. Comparison of the publication growth trends of top seven countries.
single countries/territories. Single-country articles were published by 59 different countries/territories, most of which originated from the USA (819; 25%) followed by China (mainland) (428), the UK (232), South Korea (222), and Germany (222). The USA had the most-frequent partners accounting for 51% of the internation-ally collaborative articles, followed by China (mainland) with 28%.
Moreover, the USA also had high first author articles (24%) and cor-responding author articles (25%) followed by China (mainland) with 13%. Canada (42%) had a relative high percentage of internationally collaborative articles (%C), while Taiwan (14%), Sweden (16%), and Japan (17%) were ranged as the lower ones. The time series analysis among top seven countries/territories with more than 200 articles was displayed inFig. 2. An obvious sharp rise could be found in the number of articles related to acupuncture research of the USA and China (mainland) in last 6 years. In Asian region, China (main-land), South Korea, Taiwan, and Japan ranked among the top seven.
However, Japan did not increase in number of articles in the study period.
3.5. Institutes’ research performances
The contribution of different institutes was estimated by the institute of the affiliation of at least one author of the published arti-cles. Of the 3,865 articles with author addresses, 1,821 (47%) were single institute articles and 2,044 (53%) were inter-institutionally collaborative articles. The top 10 institutes were ranked by the total number of articles, including the number and percentage of single institute articles and inter-institutionally collaborative articles, as well as first author articles and corresponding author articles (Table 3). However, a bias appeared because some institutes changed their status, for example, from a college to a university; and two universities were merged to be one. Only data for these insti-tutes appeared inTable 3were grouped together under the new name. For instance, in 2000, Fudan University and Shanghai Medical University were merged to be Fudan University, and Peking Univer-sity and Beijing Medical UniverUniver-sity were also merged to be Peking University. Whereas China Medical College changed its name to China Medical University, articles published under both institu-tional names were grouped together under the new university’s name. Among the top 10 institutes, 3 of them were derived from the USA, followed by South Korea and China (mainland) with two
Table 1
The 14 most productive journals with the number of articles, IFs, SCI categories of journals, and the position of the journal in its category.
Journal TP (%) IF SCI subject category Position
Journal of Alternative and Complementary Medicine 258 (6.5) 1.685 Integrative & Complementary Medicine 5/17
American Journal of Chinese Medicine 222 (5.6) 1.422 Integrative & Complementary Medicine 7/17
General & Internal Medicine 57/132 Acupuncture & Electro-Therapeutics Research 176 (4.4) 0.250 Integrative & Complementary Medicine 15/17
Neurosciences 221/230
American Journal of Acupuncture 141 (3.5) N/A Clinical Neurology –
Neuroscience Letters 111 (2.8) 1.925 Neurosciences 165/230
Complementary Therapies in Medicine 95 (2.4) 1.950 Integrative & Complementary Medicine 4/17
Pain 63 (1.6) 5.371 Anesthesiology 1/25
Clinical Neurology 16/167
Neurosciences 34/230
Brain Research 52 (1.3) 2.463 Neurosciences 132/230
Anesthesia and Analgesia 47 (1.2) 3.083 Anesthesiology 6/25
Clinical Journal of Pain 37 (0.93) 3.005 Anesthesiology 7/25
Clinical Neurology 47/167
Alternative Therapies in Health and Medicine 35 (0.88) 2.250a Integrative & Complementary Medicine –
Brain Research Bulletin 31 (0.78) 2.184 Neurosciences 149/230
Neural Regeneration Research 30 (0.75) N/A Cell Biology –
Neurosciences –
Neurological Research 27 (0.68) 1.277 Clinical Neurology 118/167
Neurosciences 191/230
TP: number of article; IF: impact factor; N/A: not available.
aImpact factor in 2008 JCR.
Table 2
Eleven most productive countries/territories between 1991 and 2009.
Country/territory TP TPR (%) SPR (%) CPR (%) FAR (%) RPR (%) %C
USA 1,097 1 (28) 1 (25) 1 (51) 1 (24) 1 (25) 25
China (mainland) 583 2 (15) 2 (13) 2 (28) 2 (13) 2 (13) 27
South Korea 316 3 (8.2) 4 (6.7) 3 (17) 3 (7.5) 3 (7.7) 30
UK 313 4 (8.1) 3 (7.0) 5 (15) 5 (6.9) 5 (7.1) 26
Germany 310 5 (8.0) 4 (6.7) 4 (16) 4 (7.3) 4 (7.3) 28
Japan 232 6 (6.0) 6 (5.8) 7 (7.3) 7 (5.3) 7 (5.1) 17
Taiwan 224 7 (5.8) 6 (5.8) 9 (5.9) 6 (5.6) 6 (5.6) 14
Sweden 170 8 (4.4) 8 (4.3) 12 (4.9) 8 (4.1) 8 (3.9) 16
Canada 124 9 (3.2) 10 (2.2) 6 (9.5) 9 (2.3) 9 (2.6) 42
Australia 103 10 (2.7) 9 (2.3) 13 (4.8) 10 (2.3) 10 (2.3) 25
Austria 100 11 (2.6) 11 (2.1) 10 (5.7) 11 (2.2) 11 (2.1) 31
TP: the number of total articles; TPR (%), SPR (%), CPR (%), FAR (%), RPR (%): the rank and percentage of total articles, the rank and percentage of single-country articles, internationally collaborative articles, first author articles, corresponding author articles in total articles; %C: country collaboration ratio, the percentage of collaborative articles in total articles for each country.
institutes, and one institute for each of the UK, Taiwan, and Sweden (Fig. 3).
Concerning the trends of top four institutes with more than 80 articles, Peking University and Fudan University started to pub-lish acupuncture-related articles since 1970s and remained a mild increase across the four decades. Harvard University began to pub-lish acupuncture-related articles in 1993 and 1997, and increased quickly since then. Most dramatic change occurred in South Korea where the Kyun Hee University started to step into this field in the
late-1990s, beginning to march in the 21st century and became the most published institution in 10 years.
3.6. The most frequently cited article
The most frequently cited article in the field of acupuncture was “unconventional medicine in the United States—prevalence, costs, and patterns of use” published by Harvard University, where all unconventional medicine, including massage,
chiroprac-Table 3
Top 10 most productive institutes of articles during 1991–2009.
Institute R (TP) R (SP) R (CP) R (FA) R (RP) %C
Kyung Hee University, South Korea 1 (149) 3 (37) 1 (112) 2 (108) 2 (103) 75
Fudan University, China (mainland)a 2 (132) 1 (91) 7 (41) 1 (115) 1 (107) 31
Harvard University, USA 3 (98) 12 (13) 2 (85) 5 (42) 6 (38) 87
Peking University, China (mainland)a 4 (88) 2 (41) 4 (47) 3 (73) 3 (55) 53
University of Exeter, UK 5 (66) 4 (28) 9 (38) 4 (46) 4 (44) 58
China Medical University, Taiwana 6 (64) 18 (10) 3 (54) 8 (35) 10 (27) 84
Seoul National University, South Korea 7 (60) 9 (18) 6 (42) 6 (41) 5 (41) 70
Karolinska Institutet, Sweden 8 (59) 12 (13) 5 (46) 10 (26) 8 (29) 78
University Maryland, USA 9 (54) 10 (15) 8 (39) 7 (37) 7 (32) 72
University of California, Los Angeles, USA 10 (46) 8 (19) 15 (27) 9 (31) 8 (29) 59
TP: the number of total articles; SP, CP, FA, RP: the number of single institute articles, inter-institutionally collaborative articles, first author articles, corresponding author articles; R: rank; %C: the percentage of collaborative articles in total articles for each institute.
aFudan University includes Shanghai Medical University; Peking University includes Beijing Medical University; China Medical University includes China Medical College.
Fig. 3. Comparison of the publication growth trends of top four institutes.
tic, acupuncture, were analyzed (Eisenberg et al., 1993). It has been cited 2,095 times since its publication to 2009. The 3 most cited papers related to acupuncture mechanism, written byPomeranz and Chiu (1976),Mayer et al. (1977)andSjolund et al. (1977)were all about the evidence that the analgesic effect of acupuncture is mediated by opioid substances via opioid receptors. In addition, the most frequently cited review paper in early years was “neu-rochemical basis of acupuncture analgesia” in the Annual Review of Pharmacology and Toxicology in 1982 prepared byHan and Terenius (1982). It has been cited 270 times since its publication to 2009.
The paper prepared byRamsay et al. (1998)on “acupuncture” pub-lished in JAMA-Journal of the American Medical Association in 1998 constituted a mile stone of acupuncture history in the western world where the efficacy of acupuncture was formally endorsed by National Institute of Health of the United States. It summarized the consensus that reached in the NIH sponsored “Consensus Develop-ment Conference on Acupuncture” which was held on November 3–5, 1997 in Bethesda, MD. The great impact of this paper was shown in the fact that after its publication, the insurance company started to pay the costs of acupuncture treatment for certain dis-eases, such as pain and nausea/vomiting. In later years, the most frequently cited review was “postoperative nausea and vomiting:
its etiology, treatment, and prevention” published in Anesthesiol-ogy (Watcha and White, 1992). It has been cited 706 times since its publication to 2009.
3.7. The general trends of acupuncture research
Analysis of the 3,975 articles related with acupuncture from 1991 to 2009 showed that 1,647 (41%) were dealt with pain and analgesia. Among the pain-related articles, 366 (22%) were dealt with opioid mechanisms. Furthermore, analysis of the most cited 100 articles concerning acupuncture showed a similar trend.
Acupuncture treatment for pain conditions and the opioid mech-anisms for acupuncture-induced analgesia were among the most prevalent topics.
One of the recent trends is the use of brain imaging technique to trace the brain matrix mediating acupuncture effect. This is one of the most successful non-invasive approaches to explore the sites and the characteristics of the functional changes occurred in human brain in response to acupuncture intervention.
In 1995, the first attempt of using fMRI to evaluate brain response to acupuncture was presented (Yoshida et al., 1995). Since then to 2009, 133 related papers, including 99 articles were pub-lished in SCI-Expanded journals. Thirty-two articles were pubpub-lished in journals with impact factor > 5.
Another hot topic is the discussion on how important is the role played by placebo (psychological) effect as compared to the physiological effect induced by acupuncture. In order to assess the placebo effect related to acupuncture, a placebo needle (touch-ing but not pierc(touch-ing the skin) was designed byStreitberger and Kleinhenz (1998), and being widely used in clinical trials.