Asthma is one of the most common allergic respiratory diseases around the world and places great burden on medical payment. Chinese herbal medicine (CHM) is commonly used for Taiwanese children to control diseases. The aim of this study is to analyze the CHM prescriptions for asthmatic children by using a nationwide clinical database. The National Health Insurance Research Database (NHIRD) was used to perform this study. Medical records from 1997 to 2009 with diagnosis with asthma made for children aged 6 to 18 were included into the analysis. Association rule mining and social network analysis were used to analyze the prevalence of single CHM and its combinations. Ma-Xing-Gan-Shi-Tang (MXGST) was the most commonly used herbal formula (HF) (20.2% of all prescriptions), followed by Xiao-Qing-Long-Tang (13.1%) and Xing-Su-San (12.8%). Zhe Bei Mu is the most frequently used single herb (SH) (14.6%), followed by Xing Ren (10.7%). MXGST was commonly used with Zhe Bei Mu (3.5%) and other single herbs capable of dispelling phlegm. Besides, MXGST was the core formula to relieve asthma. Further studies about efficacy and drug safety are needed for the CHM commonly used for asthma based on the result of this study.
Asthma, characterized by a chronic inflammatory airway disease, is one of the most common allergic diseases among pediatric population around the world, affecting about 10%–20% of children in Europe and the United States and 11.9% in Taiwan [
The use of complementary and alternative medicine (CAM) for diagnosis and treatment of pediatric diseases is not uncommon, especially for asthma [
Nevertheless, the evidence of CHM used for asthma is still insufficient; only small amount of CHM was proved by randomized clinical trials [
The aim of this study is to analyze the CHM prescriptions used for asthma among children in Taiwan. Based on these epidemiological statistics, more attention could be focused on efficacy and safety of the CHM which is used more commonly, especially the core formula.
The National Health Insurance Research Database (NHIRD) was used to perform CHM prescription analysis in this study. The NHIRD was composed of every medical record reimbursed by the National Health Insurance (NHI) in Taiwan. The NHI in Taiwan had two unique features: high coverage of insured populations and total reimbursement of the usage of TCM. To date, Taiwan is the only country around the world which fully paid the medical expense of TCM, including Chinese herbal medicine, acupuncture, massage, and moxibustion. Therefore, people in Taiwan may freely choose western medicine (WM) or TCM treatments for diseases without bias. Additionally, from 1995, the year the NHI started, to the present day, the coverage of the NHI increased yearly and approached 100% of total populations. The high coverage of the NHI makes the NHIRD become a nationwide database since nearly all medical records of Taiwanese were included in this database. This database can be applied and analyzed by researchers as long as the study protocol is approved by the Institution Review Board (IRB). As a nationwide database, the NHIRD has been successfully used for numerous studies, including cohort or cross-sectional design, about WM or TCM utilization status in Taiwan [
Detailed clinical information of every medical record from ambulatory visits and hospitalization was included in the NHIRD, and thus it was feasible to perform large-scale statistical analysis on prescription among pediatric asthmatic patients. For ambulatory visits, contents of the NHIRD contained patients’ identification number, gender, birthday, visit date, diagnosis, faculty, doctor, management, and prescription, in which the dose, frequency, and duration of medication were provided. The International Classification of Diagnosis, 9th revision, with Clinical Modification (ICD-9-CM) was used to represent diagnosis of every visit, and three diagnoses could be used for a visit at most. Furthermore, the first diagnosis was requested to be the major reason for a visit.
To perform the prescription analysis on pediatric TCM users with asthma, a composite two million people sample dataset was used, and the prescriptions made by these patients during January 1, 1997, and December 31, 2009, were extracted from the whole NHIRD. The sample process was done randomly by the Bureau of National Health Insurance, the administration of the NHIRD. No differences of patients’ gender and age was found between sample and whole datasets; thus, the sample dataset could be used as the representative of general population. To identify the pediatric TCM users for asthma, a series process was used and illustrated in Figure
Flow diagram of enrollment of subjects.
Additionally, age restriction of six to eighteen years was used to acquire asthmatic children. It was reported that wheezing and low lung function in early childhood may be related to small airway but not asthma, and therefore, the pathogenesis may be quite different from the asthma diagnosed in later childhood [
Furthermore, only ambulatory visits with CHM prescriptions were included into the final analysis. Visits with acupuncture, massage, or moxibustion were excluded to avoid the potential influence on prescriptions.
Frequencies, durations, and dosages of commonly used CHM and its combinations were used to demonstrate the prescription patterns for asthmatic children. Two kinds of CHM were reimbursed by the NHI in Taiwan: single herb (SH) and herbal formula (HF). HF were the mixture of multiple SH in fixed proportion, and the proportion must be the same as the HF records in TCM classics. HF and SH were all made into concentrated powder by the Good Manufacturing Practice pharmaceutical factories. Additionally, the indication, composition, and TCM properties, such as four qi and five flavors, were provided by the Committee on Chinese Medicine and Pharmacy, the administration of CHM in Taiwan. The characteristics and role of HF or SH among prescriptions could be disclosed by linking the corresponding CHM information to the prescriptions in the NHIRD.
Descriptive statistics were used to present the frequencies of the most commonly used CHM, and association rule mining (ARM) was used to analyze the combinations of CHM for asthmatic children. ARM was one of the data mining techniques, and it was developed to explore significant combinations among huge database. While the amount of clinical data increased rapidly, ARM was successfully applied into various kinds of studies in both in WM and TCM categories, such as the analysis on the comorbidities of a disease and prescription analysis of CHM on diseases [
Furthermore, the network of CHM commonly used can be established by illustrating the significant combinations. The core formula can be disclosed on the center of all CHM, and the freeware NodeXL (
The identification numbers of all subjects in the NHIRD were well encrypted, and therefore it was impossible to trace any private information. The protocol of this study was closely examined by the institutional review board of the Chang Gung Memorial Foundation (IRB no.: 101-3032B).
A total of 108,362 patients had diagnosis of asthma at least once during 1997–2009, and a total of 401,149 ambulatory visits were made. 69,879 visits were made by the children aged from six to eighteen. Additionally, a total of 24,650 CHM prescriptions could be extracted for further analysis (Figure
Distribution of numbers of CHM per prescription.
Ma-Xing-Gan-Shi-Tang (MXGST) was the most commonly used HF for asthmatic children (20.2%), followed by Xiao-Qing-Long-Tang (XQLT) (13.1%), Xing-Su-San (12.8%), Mai-Men-Dong-Tang (10.7%), and Zhi-Sou-San (10.5%) (Table
Herbal formulae (HF) frequently used for asthmatic children in Taiwan (total prescriptions: 24,650).
Latin name | Composition | Indication | Average dose (gm/day) | Prevalence (%) |
---|---|---|---|---|
Ma-Xing-Gan-Shi-Tang (MXGST) | Ma Huang ( |
Heat evils congest the lung | 14.7 | 4,986 (20.2%) |
| ||||
Xiao-Qing-Long-Tang (XQLT) | Ma Huang ( |
Exterior wind-cold with internal accumulation of retained fluid | 13.5 | 3,225 (13.1%) |
| ||||
Xing-Su-San | Xing Ren ( |
Cool dryness attacks the lung | 15.9 | 3,159 (12.8%) |
| ||||
Mai-Men-Dong-Tang | Mai Men Dong ( |
Lung and stomach Yin deficiency with deficient fire flaming upward | 12.4 | 2,640 (10.7%) |
| ||||
Zhi-Sou-San | Jie Geng ( |
Lung Qi failing to diffuse due to external pathogens | 15.0 | 2,594 (10.5%) |
Additionally, Zhe Bei Mu was the most frequently used SH, and it could be found in around 15% of prescriptions. The markedly high prevalence of Zhe Bei Mu, even higher than most HF, symbolized its significance for asthma. As the second and third commonly used SH, both Xing Ren and Jie Geng had around 10% prevalence, followed by the Huang Qin (8.1%) and Yu Xing Cao (6.6%) (Table
Single herbs (SH) frequently used for asthmatic children in Taiwan (total prescriptions: 24,650).
English name | Latin name | Indication | Average dose (gm/day) | Prevalence (%) |
---|---|---|---|---|
Zhe Bei Mu |
|
Lung heat with phlegm | 4.2 | 3,605 (14.6%) |
Xing Ren |
|
Constipation | 3.3 | 2,633 (10.7%) |
Jie Geng |
|
Lung phlegm, pus | 3.6 | 2,629 (10.7%) |
Huang Qin |
|
Lung heat, phlegm | 3.7 | 1,991 (8.1%) |
Yu Xing Cao |
|
Lung heat, pus | 4.1 | 1,634 (6.6%) |
Gan Cao |
|
Cough, smooth muscle spasm | 2.1 | 1,413 (5.7%) |
Gua Lou Zi |
|
Lung heat, phlegm, constipation | 3.7 | 1,379 (5.6%) |
Sang Bai Pi |
|
Lung heat | 3.3 | 1,314 (5.3%) |
Qian Hu |
|
Lung heat, phlegm | 3.7 | 1,176 (4.8%) |
Kuan Dong Hua |
|
Lung phlegm, cough | 3.0 | 1,007 (4.1%) |
Ma-Xing-Gan-Shi-Tang with Zhe Bei Mu was the most commonly used two CHM in combination (3.5%) (Table
Important drug pairs for asthmatic children (total prescriptions: 24,650).
Combination of two CHM | Prescription counts (%) |
---|---|
Ma-Xing-Gan-Shi-Tang with Zhe Bei Mu | 871 (3.5%) |
Xing Ren with Zhe Bei Mu | 822 (3.3%) |
Jie Geng with Zhe Bei Mu | 720 (2.9%) |
Ma-Xing-Gan-Shi-Tang with Jie Geng | 649 (2.6%) |
Ma-Xing-Gan-Shi-Tang with Huang Qin | 552 (2.2%) |
Huang Qin with Zhe Bei Mu | 537 (2.2%) |
Ma-Xing-Gan-Shi-Tang with Sang-Ju-Yin | 502 (2.0%) |
Ma-Xing-Gan-Shi-Tang with Sang Bai Pi | 474 (1.9%) |
Ma-Xing-Gan-Shi-Tang with Yu Xing Cao | 474 (1.9%) |
Ma-Xing-Gan-Shi-Tang with Xin-Yi-Qing-Fei-Tang | 457 (1.8%) |
Network of CHM combinations for asthmatic children.
To explore possible pharmacological mechanisms of CHM as external validation, extensive literature search was done in PubMed (last assess on April 20, 2013). Search keywords were composed of every CHM name and asthma-related words, such as bronchitis, allergic rhinitis, allergy, and hyper responsive airway. CHM names in English, Latin, and Chinese (pin yin) were used as keywords when searching references, and, additionally, the corresponding names in Kampo medicine and Korean medicine were included in searching words to avoid omission. The immunomodulation effects on suppressing type 2 T-helper cells and decreasing subsequent cytokine secretion were exclusively found in nearly all commonly used CHM; however, only HF had bronchodilation effect (Table
Possible pharmacological mechanisms of CHM frequently used for asthmatic children.
CHM | Active ingredients | Possible mechanisms | References |
---|---|---|---|
|
|||
Zhe Bei Mu | Ethanol extract | Decreases serum IL-6, IL-8, and TNF-alpha and inhibits MAPK pathway | [ |
Xing Ren | Aqueous extract | Inhibits type 2 T-helper cell | [ |
Jie Geng | Aqueous extract | Decreases serum IgE, ROS scavenger | [ |
Saponin | Decreases Syk-dependent cascades and inhibits downstream MAPK, Akt pathway | [ | |
Ethanol extract | Decreases serum IL-6, prostaglandin D2, leukotriene C4, beta-hexosaminidase, and COX-2 protein | [ | |
Huang Qin | Skullcapflavone II | Bradykinin antagonist, decreases type 2 T-helper cells and increases transforming growth factor-beta 1 | [ |
Ethanol extract | Restores serum IL-8 and TNF- |
[ | |
Yu Xing Cao | Ethanol extract | Suppresses Th2 responses decreases subsequent expression of IL-4 and thymus and activation-regulated chemokine | [ |
Gan Cao | Flavonoid | Inhibits secretion of eotaxin-1 and decreases eosinophil recruitment in asthmatic airway | [ |
Prevents eosinophil cationic protein interaction | [ | ||
Suppresses memory Th2 cells, decreases serum IL-4, IL-13, and increases IFN-gamma | [ | ||
Sang Bai Pi | Water extract | Suppresses Th2 cell and enhances regulatory T cells | [ |
Qian Hu | (±)-Praeruptorin A | Relieves airway inflammation, airway hyperresponsiveness, and remodeling via suppressing Th2 cell | [ |
Kuan Dong Hua | Water extract | Antitussive effect | [ |
Chlorogenic acid, 3,5-dicaffeoylquinic acid, and rutin | |||
| |||
|
|||
Ma-Xing-Gan-Shi-Tang (MXGST) | Antitussive effect | [ | |
Beta-2 androgenic effect | [ | ||
Xiao-Qing-Long-Tang (XQLT) | Relieves airway inflammation, remodeling, and immunomodulation, via downregulating IL-10, IL-13, RANTES, Eotaxin, and MCP-1 | [ | |
Anti-inflammatory effect, decreasing IgE titer in bronchial lavage fluid. | [ | ||
Suppresses histamine signaling | [ | ||
Suppresses Th2 cell, decreasing IL-4 and IL-5, restoring IFN-gamma | [ | ||
Bronchial dilation, partial beta-2 androgenic effect | [ | ||
Mai-Men-Dong-Tang | Antitussive effect | [ | |
Bronchial dilation via beta-2 androgenic effect | [ |
The key finding of this study is the recognition of MXGST as the core formula for treating asthmatic children. To the best of our knowledge, this is the first study to demonstrate the prevalence and the prescription patterns of CHM prescribed for asthmatic children. Also, the combination patterns were discovered by applying ARM to the nationwide prescription database. Further careful examination for efficacy and safety of MXGST should be carried out since the prevalence is relatively high among all available CHM, and this result from a nationwide study could be regarded as a form of consensus from the majorities of TCM physicians. MXGST is one of the most famous formulas, and composition and indication were shown in the ancient TCM classic, “Shanghanlun”. MXGST has been widely used for infectious and inflammatory airway diseases for thousands of years. The indication for MXGST includes wheezing, cough, coarse breathing sound, thirst, and fever; the hot-type asthma in TCM terminology [
The high prevalence of MXGST raises demand on evidence of efficacy which is still lacking. The effects on asthmatic children may be related to beta-2 adrenoreceptor agonist with subsequent bronchodilation and antitussive effect [
In addition to use alone, MXGST was usually used with other CHM, especially SH, and phlegm was the most common indication of SH (Table
In contrast to synergistic effects, the SH commonly used to combine with other SH or HF raises the importance of safety issues, for example, food allergy from Xing Ren (
Furthermore, different triggers for asthma attack among children may influence the prescription of CHM. Xiao-Qing-Long-Tang (XQLT) is the second most commonly used HF for asthmatic children, and it has extensive effects on decreasing Th2-related immune response, anti-inflammatory, reducing secretion, and has potent bronchodilator effect [
The major advantage of this study is to explore core formula and common CHM combinations for asthmatic children among a nationwide clinical database. ARM is a powerful tool to find out the important CHM combinations among large number of prescriptions, and the combination patterns are an important reference for TCM doctors because TCM doctors usually use multiple CHM in one prescription when treating diseases. Also, the commonly used CHM is not necessarily the important CHM used in combination, and therefore, the study of combination patterns is important when analyzing CHM prescriptions; for example, XQLT was not found in commonly used two CHM in combination although it was the second most HF used for asthma (Tables
However, there are still some limitations for this study. First, the number of CHM prescriptions for final analysis may be substantially reduced since the single diagnosis of asthma is used. However, this procedure may promptly reduce the influence on prescription from comorbidities, such as allergic rhinitis. Since allergic rhinitis highly coexisted with asthma, the use of single diagnosis was helpful in focusing on prescriptions for asthma. Second, the folk medicine in Taiwan was not included in the analysis in this study since folk medicine was not reimbursed by the NHI. Only CHM that is produced by the GMP factory and that passed the examination of the CCMP is approved to be prescribed, and thus, the result of this study is readily used for further studies. Third, CHM with higher prevalence could not directly correspond to better effectiveness. The precise efficacy of CHM should be evaluated carefully by conducting well-designed randomized, placebo-controlled, and double-blinded clinical trials. The CHM explored in this work would be valuable references to the choice of CHM for clinical trials.
MXGST was the core formula for asthmatic children, and it was commonly used with phlegm-dispelling SH. By analyzing the nationwide clinically relevant prescription database, this result may be regarded as a form of consensus from TCM physicians in Taiwan. Still, further stringently designed clinically relevant randomized clinical trials for efficacy and safety study about the commonly used herbal preparations in the treatment of asthma discovered in this study are warranted.
The authors declare that they have no conflict of interests in this study.
The data of this study is retrieved from the NHIRD, provided by the Bureau of National Health Insurance (BNHI). The interpretation and conclusion do not represent the opinion of the BNHI.