Mapping the knowledge of international Chinese medicines treatment on type 2diabetes: a biblimetrical study Jiahui Hu, Kangle Shi, Qinggang Meng DOI: 10.1016/j.jtcms.2016.12.002 To appea
Trang 1Mapping the knowledge of international Chinese medicines treatment on type 2
diabetes: a biblimetrical study
Jiahui Hu, Kangle Shi, Qinggang Meng
DOI: 10.1016/j.jtcms.2016.12.002
To appear in: Journal of Traditional Chinese Medical Sciences
Received Date: 28 December 2016
Accepted Date: 28 December 2016
Please cite this article as: Hu J, Shi K, Meng Q, Mapping the knowledge of international Chinese
medicines treatment on type 2 diabetes: a biblimetrical study, Journal of Traditional Chinese Medical Sciences (2017), doi: 10.1016/j.jtcms.2016.12.002.
This is a PDF file of an unedited manuscript that has been accepted for publication As a service to our customers we are providing this early version of the manuscript The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Trang 2M AN
US CR
IP T
AC CE
PT ED
ACCEPTED MANUSCRIPT
Mapping the knowledge of international Chinese medicines treatment on type 2
diabetes: a biblimetrical study
JiahuiHu, Kangle Shi, Qinggang Meng*
School of Basic Medical Science, Beijing University of Chinese Medicine Beijing 100029, China
*
Corresponding author
E-mail address: mqgang@126.com (Q Meng)
Abstract
Objective: To uncover and identify the hot topic and frontier of Chinese medicine
treatment of type 2 diabetes mellitus (T2DM)
Methods: Web of Science TM was searched for published articles for Chinese medicines treatment of T2DM ranging from January 1st, 2002 to July 6th, 2016 Knowledge maps of the international Chinese medicine treatment of T2DM are visualized by using document co-occurrence analysis and word frequency analysis (Institution and Journal),co-citation clustering analysis (Co-reference), keyword co-occurrence clustering analysis with CiteSpac III, a tool of scientometrics
Results: Universidad Nacional Autonoma de Mexico is the institution with the highest
number of published papers that had been cited in this field, while China has four institutions among the top 10 The journal of the highest frequency of co-cited journal
was Diabetes Care, a core one in the field Keywords co-occurrence network was
composed of 185 nodes, 541 lines, and divided into 10 clusters Co-citation network
of co-reference was composed of 407 nodes, 1199 lines, and divided into 20 clusters Using Chinese medicine to improve insulin resistance and Chinese medicine research
on blood glucose control are the hot topics The frontier contains two aspects: new drugs development and application of intestinal insulin treatment and development and use of traditional Chinese medicine antidiabetic plants
Conclusion: Institutions from China still plays a major role in TCM-focused T2DM
studies The effect of TCM herbs on insulin resistance is the hot topic of the domain Developing new TCM herbal medicine that regulates incretin effect is the domain frontier Research on the Chinese medicines treatment of T2DM needs more high-quality evidence to support, and its mechanism requires further exploration
KEYWORDS
Chinese medicine; CiteSpace; Knowledge map; Type 2 diabetes mellitus
1.0 Introduction
Diabetes is a growing problem worldwide that incurs an estimated 5 million deaths in 2015.1 Type 2 diabetes mellitus (T2DM) takes the majority of people with diabetes around the world and can be controlled through lifestyle regulations and medication.2 The disease belongs to consumptive thirst (xiao ke) and has been treated
Trang 3M AN
US CR
IP T
AC CE
PT ED
with empirical herb combinations in traditional Chinese medicine (TCM) since 26
BC.3 However, thousands of years of experience in dealing with T2DM have not been converted to clinical evidence due to the limited technologies and methodologies until recent 50 years, when greater efforts have been put forth to bring these TCM therapies under rigor scrutiny of science-based medicine.4,5 Followed with these efforts are the escalating amounts of TCM-focused or related T2DM literatures, as poses ever-increasing challenges for the newcomers to get a quick and exact picture of the area
Based on the bibliometrical methodologies of citation and co-citation analysis, 6 knowledge can be mapped to display and thus evaluate the structural and dynamic aspects of scientific research.7 Its current applications include medical science,8 management science,9,10 and information science,11 etc CiteSpace III is a well-validated, open source tool which enables analysts to realize knowledge mapping
in an intuitive way.12
In this paper, by CiteSpace III, we mapped the TCM-focused T2DM literatures retrieved from Web of Science TM toidentify the hot topics and frontier, in order to present a simple and concise picture of the field
2.0 Materials and methods
With a concerted search query, we retrieved 2348 studies related to TCM treatment on T2DM They were then separately screened by two investigators (Jiahui
Hu and Kangle Shi) for eligibility based on the inclusion and exclusion criteria A third investigator (Qinggang Meng will settle dispute between the initial two investigators Eligible literatures will be extracted as the format executable in CiteSpace III and imported into the tool for further treatment
2.1 Search strategy
The data ranging from January 1st, 2002 to July 6th, 2016 were downloaded from the Web of Science TM (WOS TM)
Articles were searched in title, abstract or keywords using the following queries: (“type 2 diabetic mellitus” or “type 2 diabetes” or “T2DM” or “type II diabetic mellitus” or “DM II” or “type II diabetes”) and (“Chinese medicine” or “TCM”
or “Chinese herb” or “Chinese herbal medicine” or “traditional medicine” or
“alternative medicine” or “complementary medicine” or “plant” or “herbal” or “herb”
or “herbal medicine” or “plant medicine” or “plant drugs” or “phytotherapy”
or “plant extracts” or “prescription” or “formula” or “Chinese patent medicine”
“integrated traditional Chinese and Western medicine”
“integrated tcm wm” )
2.2 Inclusion criteria
The articles related to Chinese medicines treatment on T2DM were selected
Trang 4M AN
US CR
IP T
AC CE
PT ED
ACCEPTED MANUSCRIPT
2.3 Exclusion criteria
The articles excluded were as follows: type 1 diabetes only; Western medicine treatment only; sports; diagnosis; policies; detection; care therapy; risk prediction, and others (massage, extraction process of plant components, and patient compliance, and the like)
The procedure of screening the articles was displayed (Fig 1)
Web of Science TM
Initially selected n=2348
Initially selected n=1706
Finally included n=1518
Excluded by reading titles,
abstracts and key words : n=642
In which:
T1DM only
Western medicine treatment only
Sports
Diagnosis
Policies
Detection
Care therapy
Risk prediction
Others
Excluded by reading full texts : n=188 In which: T1DM only
Western medicine treatment only
Sports
Diagnosis
Policies
Detection
Care therapy
Risk prediction
Others
n=4 n=54 n=11 n=4 n=12 n=4 n=7 n=17 n=75
n=95 n=153 n=55 n=10 n=16 n=20 n=17 n=109 n=167
Figure 1 Flow chart of screening the articles
2.4 Data treatment
The data were then imported into CiteSpace III to carry out document co-occurrence analysis, word frequency analysis (Institution, Journal), co-citation clustering analysis (Co-reference) and keyword co-occurrence clustering analysis Each analysis was reported as a series of statistics and a visualized map for manual interpretation and exploration
We visualized the knowledge maps of TCM-focused T2DM domain using the document type of “Institution”, “Cited Journal”, “Keyword”, and “Cited Reference”, respectively (this could be fulfilled by selecting the corresponding items in CiteSpace III parameter setting window) Based on the recommendations and examples from the tool developer, 8,12-14 other parameters were set as: the unit of analysis (year) =1; top
Trang 5M AN
US CR
IP T
AC CE
PT ED
N per slice = 50 ; pruning algorithm=pathfinder
3.0 Results and analysis
The retrieved articles (n=1518) were published from 2002 to 2016
3.1 Distribution
3.1.1 Analysis of institutions
CiteSpace III visualizes a merged network based on several networks corresponding to snapshots of consecutive years In this section, the nodes are institutions Lines that connect nodes are collaboration links Color indicates cited documents issued time, which corresponds to the top time bar The direction of time points to the right Cooler colors indicate older time (Fig 2)
By institution analysis, global distribution of institutions within a TCM-focused T2DM domain was demonstrated and quantified base on how frequently they had been cited, and this might help investigators to build research partnerships The statistics displayed there were 134 institutions in all from 1518 studies The top 10 most frequently cited institutions and their countries were listed (Table 1) It was
found that Universidad Nacional Autonoma de Mexico (English: National
Autonomous University of Mexico) was ranked first position with citation counts of
21, and this might be somehow inconsistent with most investigators’ impression that Mainland China; Taiwan, China; South Korea and Japan, where TCM were most frequently practiced, and should dominate the domain of TCM study However, China still took up four seats among the top 10, demonstrating its important role in the field Through the sparse lines of connection, collaboration between different institutions was believed to be not universal
Trang 6M AN
US CR
IP T
AC CE
PT ED
ACCEPTED MANUSCRIPT
Figure 2 International Chinese medicine treatment of T2DM document source institutions
Table 1 Top 10 institutions of international Chinese medicine treatment on T2DM
document source
Citation
3.1.2 Statistics of cited journals
The statistics of publication sources can be helpful in finding the core journals
in the field, which possesses a major position of the research By CiteSpace III (Table 2), it was indicated that the journal with the highest co-cited frequency was
DIABETES CARE, a monthly journal founded by the American Diabetes Association
in 1978 The aim of the journal is for the health care practitioner that is intended to
Trang 7M AN
US CR
IP T
AC CE
PT ED
increase knowledge, stimulate research, and promote better management of people with diabetes.15
Table 2 Top 10 cited journals
3.2 Analysis of co-occurrence keywords
Parameter settings only changed one place: pruning (minimum spanning tree, MST) The rest remained the same as listed in common selections
In this section, citation tree rings indicated the co-occurred keyword of an article The signature (fundamental setting information and statistical metrics) of the network was shown on the upper left corner of the visualization As Fig.3 displayed, keyword co-occurrence network was composed of 185 nodes, and 541 lines Two reported metrics, the modularity Q and the mean silhouette scores, depicted the overall structural properties of the network Here, the modularity Q value was 0.427, higher than 0.3 (reference value),12 which indicated the clustering was effective, and the mean silhouette score of 0.5981 suggested that the clusters were credibly well-defined
Trang 8M AN
US CR
IP T
AC CE
PT ED
ACCEPTED MANUSCRIPT
Figure 3 Knowledge map of co-occurrence keywords
3.2.1 Major clusters
The network was divided into 10 co-citation clusters, which were labeled by index terms from their own citers Clustering number was inversely proportional to the cluster size
The Silhouette column indicated the homogeneity of a cluster The higher the silhouette score obtained, the more consistent the cluster members became According
to previous studies,13,14 a silhouette score more than 0.5 suggested that we could have sufficient confidence to believe the cluster existed and the cluster deserved further exploration we listed four largest clusters with silhouette score higher than 0.5 (Table 3) The mean (cited year) of a cluster indicated whether it was formed by generally recent papers or old papers
Table 3 Largest clusters of Keywords
Cluster ID Size Silhouette Label Mean (cited
year)
Trang 9M AN
US CR
IP T
AC CE
PT ED
3.2.2 Citation frequency
Removing meaningless key words such as diabetes, T2DM, we obtained the highest frequency co-citation keywords as follows
According to Fig.3, Table 3 and Table 4 insulin resistance (IR) was the most frequently mentioned keyword IR causes healthy people to develop glucose intolerance and finally T2DM.16 TCM emphasizes syndrome differentiation and treatment, so the clinical significance of T2DM, which corresponds to clinical indicators, must be clear The number of T2DM complications entails a wide range of TCM therapies that have been explored by investigators On the specific clinical treatment methods, the use of phytotherapy includes single herb, prescriptions and plant extracts In Western countries , TCM is usually considered as complementary and alternative medicine, indicating it is applied as adjuvant approaches when some modern therapy exists for a disease in question (complementary) and sometimes as major therapy when it does not (alternative) Oxidative stress can cause β-cell function decline and lead to IR.17 While IR and β-cell dysfunction play important roles in the occurrence and development of T2DM
Table 4 Most citation frequencies of Keywords
Citation
frequency Keywords Cluster ID Centrality Year
3.3 Analysis of co-cited references
In this section, we identified co-citation references network in Chinese medicine treatment of T2DM domain This could be seen in Fig.4 by clustering and Fig.5 by transforming the clusters into a timeline view with a built-in “Fisheye” function A citation tree ring represented an article citation history Its overall size reflected the number of citations, and its color referred to citation frequency The thickness of the rings was proportional to the number of citations in the corresponding time partition
In a timeline view, each cluster was arranged on a horizontal timeline The direction
of time points to the right, facilitated the identification of field growth
Trang 10M AN
US CR
IP T
AC CE
PT ED
ACCEPTED MANUSCRIPT
Figure 4 Knowledge map of co-cited references
Figure 5 Mapping on co-citation references by time line view
According to Fig.5, the network was composed of 407 nodes, and 1199 lines The modularity Q value was f 0.7247, whichpassed the threshold of credibility, indicating the network has been reasonably divided into loosely coupled clusters The mean