key: cord-0958997-7721wcy7
authors: Ang, Lin; Song, Eunhye; Lee, Myeong Soo
title: Randomized controlled trials of traditional, complementary, and integrative medicine-based interventions for coronavirus disease 2019 (COVID-19): a bibliometric analysis and review of study designs
date: 2021-09-23
journal: Integr Med Res
DOI: 10.1016/j.imr.2021.100777
sha: d21addde8b312b78b4124e85b25075c5eaf804a9
doc_id: 958997
cord_uid: 7721wcy7

BACKGROUND: To date, the coronavirus disease 2019 (COVID-19) pandemic remains ongoing and continues to affect millions of people worldwide. In the effort of fighting this pandemic, there has been an increasing interest in the potential of traditional, complementary, and integrative medicines (TCIMs) in engaging COVID-19. This study presents a bibliometric analysis of the research trends of TCIMs for COVID-19. METHODS: Six databases were searched on July 15, 2021, to retrieve all the citations on TCIM-focused RCTs available on COVID-19. Only RCTs that mentioned at least one TCIMs for the treatment and/or management or COVID-19 were eligible. Data such as number and countries of trials conducted, publication journal, research focus, studies designs, and sample size were extracted for analysis. RESULTS: The resulting 56 articles included 28 English articles, 19 Chinese articles with English abstracts, and 9 Chinese articles with 553 unique authors. Analyses had shown that China was the dominant country with TCIM related RCT publications, followed by India and United States. The included articles were published across 24 English journals and 22 Chinese journals with a wide range of impact factors from 0.220 to 56.272. The most commonly studied TCIM modalities included Chinese herbal decoction (n=12) and Chinese patent medicine (n=16). In terms of study designs, TCIM interventions were integrated with standard medicine across the trials with most trials having a small to medium sample size and open-labeled. CONCLUSION: This bibliometric analysis of RCTs demonstrated the research trends and characteristics of TCIM mentioned in COVID-19 research. Although there are still many research gaps and limitations for pandemic research, the publication of TCIM-focused RCTs is anticipated to show a continuously increasing trend.

As of August 3, 2021, coronavirus disease 2019 (COVID- 19) has affected more than 198 million people worldwide. 1 While appropriate precautions such as social distancing, staying home, wearing a mask, and frequent handwashing are continuously taken, the world remains fully engaged and continues its effort to tackle the ongoing pandemic. Being part of the effort, there has been a growth in the interest to include traditional, complementary, and integrative medicines (TCIMs) in treating or preventing COVID-19. TCIM includes traditional medicine indigenous to different amorphous rationalities, traditions, and cultures, as well as a broad range of health care practices and supplements that are yet to be integrated into the dominant health care system.

Existing systematic reviews and meta-analyses with a focus on TCIMs for COVID-19 have shown positive effects on improving clinical symptoms and complementing standard medications. [2] [3] [4] Additional evidence has also shown that TCIMs may strengthen physical and psychological resilience which is worth to be studied further. 5, 6 Despite the considerable number of TCIM-related studies published on COVID- 19 , there have been fairly few efforts taken to systematically analyze and describe the trends of these research. The assessment of publication trends is necessary to understand the quantity and quality of these TCIM-focused COVID-19 studies while identifying their research gaps and avenue for future research. A bibliometric approach is useful in quantitatively analyze literature characteristics and publications trends, where its findings can be used to identify strengths and limitations in the field of research.

This bibliometric study aimed to analyze the publication trends in TCIM randomized controlled trials (RCTs) since the outbreak of this ongoing pandemic, including (1) quantity of RCT publications, (2) most productive countries based on where the trials were conducted, (3) most common journals with their subject categories and recent impact factors, (4) most cited articles, (5) most commonly used TCIM modalities, and (6) the characteristic of study designs.

Six bibliographic databases including PubMed, Embase, Allied and Complementary Medicine Database PubMed is the key database among health science researchers as it comprises extensive citations for biomedical literature with free access and user-friendliness. As PubMed alone does not cover all relevant literature, Embase is used as a supplement to identify studies that are not indexed in PubMed. AMED, on the other hand, is a specialized database designed on alternative and allied therapies-related subjects. Being the country that first considered TCIM approach for COVID-19 and issued relevant national guidelines, TCIM has been widely studied in China for its effectiveness in the treatment and prevention of COVID- 19. 7 Therefore, three major Chinese databases were also included in the literature search. CNKI is the world's largest Chinese journal database with constant updates and free access to wide-ranging coverage of citations in various fields. As a supplement to CNKI, Wanfang which is currently indexed in EBSCO Information Services (EBSCO) and VIP which has the highest indexing of Chinese journals are both preferred leading databases for the literature search of clinical researches.

The literature search was performed on July 15, 2020. The authors used Medical Subject Heading (MeSH) terms, title/abstract words, and author keywords in PubMed, and equivalent indexed subject headings and terms were used in the other databases. The list of TCIM therapies provided by the World Health Organization 8 and National Center for Complementary and Integrative Health 9 was also referred to extend the search coverage. The complement search strategies are provided in the supplement. In addition to the comprehensive database search, relevant systematic reviews with or without meta-analysis were manually screened for additional eligible RCTs.

Only randomized controlled trials (RCTs) relating to TCIM for the management and/or treatment of COVID-19 were included in this bibliometric analysis. RCTs mentioning COVID-19 but not TCIM, or vice versa, were excluded. Other clinical studies such as case-controlled trials, observational studies, and case series/reports were also excluded. Research protocols, methodological studies, reviews, dissertations, conference proceedings, and abstracts were not eligible. As the COVID-19 is an emergency in public health, preprints indexed in the aforementioned electronic databases were considered for inclusion.

Titles and abstracts screening we performed by one author using reference manager software, Endnote 20.1. The full text of the potentially eligible RCTs were retrieved and assessed for inclusion by one author. All the screening and selection were verified by a second author and any disagreement was resolved by consensus with an arbiter.

Bibliometric data were retrieved using export function of the electronic databases and was manually extracted for the following information: title, authors' names, affiliations and affiliated countries, DOI, language of publication, publication type, publication month and year (issue online), publication journal, research focus, trial designs, country of the trial conducted, sample sizes, TCIM modalities, intervention and comparison regimes, outcome measured, and trial registry number. All results were verified by a second author. The raw data were exported on the same day of the literature search to avoid discrepancies between the database's daily updates.

Bibliometrics is the use of data from citation indexes to analyze, monitor, and visualize study publication trends, impacts, and research gaps. Specifically, the publications and citations, journal performance, and other information of interest such as types of TCIM modalities and outcome measures were analyzed in this study. This bibliometric analysis did not involve human research or protocol that require ethical approval from the relevant review board or committee.

Data were summarized using frequencies with percentages. For the analysis of study characteristics, the number of citations was retrieved from Thomson Reuters' Web of Science (WoS) database for all studies regardless they were indexed in English or Chinese databases. To describe the journal's performance and whether the journal is TCIM-focused journal, the authors used Journal Citation Reports (JCR) released impact factor and journal category for English articles, and CNKI reports for Chinese articles. Both JCR and CNKI defined the journal impact factor as the ratio between the yearly number of citations and the number of citable articles published during the preceding 2 years. The research and publication trends associated with the eligible RCTs were also identified and presented. Visualization of the research trend using mapping and word clouds were created with mapchart.net and wordart.com.

Results revealed 90 eligible publications on TCIM focused RCTs. Among them, 56 (63%) were published as original full-length journal articles and 34 (37%) were published as protocols that reported the study design and rationale. As the consistency of protocols with the final study report might not be definite, protocol publications were excluded in the final analysis. Therefore, only a total of 56 articles were included and reviewed in the bibliometric analysis. All the 56 RCTs mentioned at least one TCIM for the management and treatment of COVID-19. A flowchart on the study selection process is provided in Fig. 1 .

Across 574 authors of the included publications, 553 authors were unique and the number of publication per author ranged from 1 to 4 (1 article (n=537); 2 article (n=12); 3 article (n=3); and 4 articles (n=1)). In terms of language, 28 articles (50%) were published in English and 28 articles (50%) were published in Chinese of which English titles and abstracts were provided for 19 articles. The number of TCIM focused publications (Fig. 2 ) showed a fluctuation seesaw over the pandemic.

Throughout the pandemic, China was the leading country with TCIM related RCT publications, followed by India, United States, and Iran. Publications were affiliated from the following 13 countries: China (n=41),

Germany (n=1), Pakistan (n=1), Saudi Arabia (n=1), Spain (n=1), and Turkey (n=1). The number of countries affiliated per publication only involved one (n=51) to two (n=7) countries.

Most of the journals that published TCIM RCTs were from the field of general medicine, health care sciences, pharmacology, and integrative medicine. In total, 56 articles were published in a total of 46 journals, of which 24 were English journals and 22 were Chinese journals. Only 5 journals (21%) were identified as TCIM-focused English journals and 12 journals (55%) as TCIM focused Chinese journals. The impact factor of these journals ranged widely from 0.220 to 56.272. The number of articles published per journal ranged from 1 to 3. The articles that had been cited at least 5 times as of the search date in Web of Science databases are provided in Table 1 and the details of the journals are provided in Table 2 . To analyse the terminology used for study design in the title, the broad range of terms that were used to describe the title of the RCTs is illustrated in Fig. 3 .

The most commonly studies TCIM modalities included Chinese herbal decoction (n=12), Chinese patent medicine (n=16), Chinese herbal injection (n=3), combined Chinese medicine interventions (n=5), moxibustion (n=1), vitamin D (n=5), vitamin C (n=3), herbal extractions (n=3), zinc (n=1), propolis (n=1), omega 3 fatty acid (n=1), zinc and ascorbic acid (n=1), Ayurvedic medicine (n=1), guided imagery (n=1), progressive muscle relaxation (n=1), and ozone therapy (n=1), as shown in Table 3 . The details of the modalities used is provided in Table 4 .

Among the retrieved data, 54 trials focused on the treatment and 2 trials focused on the rehabilitation of COVID-19 using TCIM approach. Due to the nature and complexity of the disease, all participants in the intervention group or the control group of the 54 trials focused on COVID-19 treatment were given standard medications for COVID-19 such as antibiotics, antiviral and anti-inflammation drugs. None of the studies used inactive control (no intervention) except for one study that focused on the rehabilitation of discharged patients. In the other words, TCIM interventions were not used alone but integrated with standard medicine across the trials. Twenty-five trials were registered in the trial registry while 31 trials did not report on the relevant information. The visualization of the countries that conducted TCIM-focused RCTs on COVID-19 is presented in Fig. 4 . For sample size, a total of 5932 study participants were included across the RCTs. Small to medium sample-sized studies were dominant with only a few studies including larger sample size of more than 200 study participants (Fig 5) .

Over the studied population, 5169 (87%) participants were positive cases, 343 (6%) were suspected or positive cases, and 148 (2%) were suspected cases of COVID-19. The severity of the diseases ranged from mild to critical to recovery. After standardizing and merging similar outcomes, there were 33 outcomes reported in the included RCTs, including admission to intensive care unit, adverse events, Apache II score, arterial blood gas parameters, changes in oxygenation index, chest X-ray or CT findings, clinical symptoms score, duration of hospital admission, frequency and duration of mechanical ventilator assistance, Glasgow coma scale, health-related laboratory markers, hospitalization rate, incidents of acute respiratory distress syndrome (ARDS) development, inflammatory markers, medication status, mortality, need for vasoactive drugs, organ functions, overall improvement rate, overall rate of clinical cure, peak expiratory flow rate, prevalence of antibiotic, quality of life, rate and time to negative conversion of viral tests, rate of advancing to critical, rate of clinical deterioration, remission rate and time of clinical symptoms, retest of viral tests, severe adverse events, survival rate, traditional Chinese medicine (TCM) syndrome score, TCM tongue and pulse diagnosis score, and time to disease recovery. Details of the outcomes measured according to the severity of disease are provided in Table 5 .

The full information of all RCTs containing all of the aforementioned characteristics is available in supplement for the convenience of clinical researchers who are interested in pursuing the original publication or foster further researches.

This study quantitatively explored the current publication trends of TCIM-focused RCTs, as of July 2021. This bibliometric analysis also found that small to medium-sized studies were dominant, which is often the common limitation that leads to a certain degree of bias in the study findings. In terms of outcome measures, it is seen that the outcomes measured varied across the trials with many less critical outcomes in each intended population. Implications of such heterogeneity should not be ignored in future studies and researchers should consider selecting their outcomes by referring to the core outcome set developed by Core Outcome Measures in Effectiveness Trials (COMET) Initiative. 11 There is also a lack of consistency in the terminology for the study design in the title, suggesting lack of proper reporting as CONsolidated Standards Of Reporting Trials (CONSORT) guidelines included identification of a randomized trial in the title.

It is not doubting that clinical research on COVID-19 faces many ethical challenges due to the complexity of the disease. Yet, well-designed trials are inexcusably necessary as ill-designed trials are likely to produce potential misleading or exaggerated information; in the meantime exposing trial participants to a higher risk of harm. The perception that core methodological components of high-quality RCTs such as adequate sample size, proper randomization or blinding, and validated outcome measures are expendable given the urgency of COVID-19 situation is highly inappropriate. Several rapidly published research trials with suboptimal designs have faced retraction, but not without having clinical practice to face the consequences. As exemplified by hydroxychloroquine, its usage had increased after the rapid publication of an open-label, non-randomized, small-sampled trial claiming its effectiveness and was later shown to be no different from placebo and revoked in a well-designed RCT with large sample size. 12, 13 Therefore, RCTs with suboptimal designs will not help either patients or clinicians and may result in significant ethical lapse.

Instead of allowing exceptionalism, researchers should coordinate their studies to uphold the gold standard of RCT to contribute to the advancement of clinical evidence base for COVID-19. 14 To the authors' knowledge, this study is the first bibliometric analysis of TCIM-focused RCTs. In contrast to a number of COVID-19 bibliometric analyses published over the past year, the authors centered their study to RCTs only and manually screened and extracted all the data to provide a comprehensive review on the state of TCIM clinical research. Particularly, a bibliometric analysis on the global trends of overall TCIM research providing a larger view of the publication landscape 15 and a bibliometric analysis focusing on traditional Chinese medicine 16 Nevertheless, the authors acknowledge the limitation of this study. Firstly, the authors did not include explicit search terms related to each TCIM modalities such as specific herb/product names or technique used (e.g. curcumin, calcifediol, Baduanjin); however, these interventions were included if they were described as TCIM modalities. Therefore, there might be RCTs on these TCIM interventions that the authors' missed. Second, a number of non-English or non-Chinese journals may not be indexed in the aforementioned databases; hence, it is possible that several TCIM-focused RCTs are published in their native language or local journals. Regrettably, such articles are often hard to capture as they are less accessible to authors and there could be other forms of less known TCIM modalities that were overlooked in this analysis.

This bibliometric analysis of RCTs demonstrated the research trends of TCIM mentioned in COVID-19

research. There are still many research gaps in the pandemic research of TCIM which requires evaluation of study designs for more rigorous research from various countries. The burgeoning literature on TCIMfocused clinical trials for COVID-19 in the future is highly anticipated. 

LA, ES, and MSL are editors of IMR but their status had no bearing on editorial consideration. This article was externally peer-reviewed. There is no other conflicts of interest.

This research was supported by Korea Institute of Oriental Medicine (KSN2013210).

Not applicable.

The data that support the findings of this study are available as supplementary material. 

WHO Coronavirus (COVID-19) Dashboard

Herbal Medicine for the Treatment of Coronavirus Disease 2019 (COVID-19): A Systematic Review and Meta-Analysis of Randomized Controlled Trials

The Effects of Traditional Chinese Medicine as an Auxiliary Treatment for COVID-19: A Systematic Review and Meta-Analysis

Therapeutic effects and safety of oral Chinese patent medicine for COVID-19: A rapid systematic review and meta-analysis of randomized controlled trials

The Relevance of Complementary and Integrative Medicine in the COVID-19 Pandemic: A Qualitative Review of the Literature

Public Health Approach of Ayurveda and Yoga for COVID

Herbal medicine and pattern identification for treating COVID-19: a rapid review of guidelines

Health topic: Traditional, complementary and integrative medicine

Open access (OA) Online-first publishing of research papers on COVID-19

Core Outcome Measures in Effectiveness Trials (COMET) Initiative. Core outcome set developers' response to COVID-19

Effect of Hydroxychloroquine in Hospitalized Patients with Covid

Misguided Use of Hydroxychloroquine for COVID-19: The Infusion of Politics Into Science

Against pandemic research exceptionalism

Global research trends at the intersection of coronavirus disease 2019 (COVID-19) and traditional, integrative, and complementary and alternative medicine: a bibliometric analysis

Bibliometric analysis of researches on traditional Chinese medicine for coronavirus disease 2019 (COVID-19)

Reflections on treatment of COVID-19 with traditional Chinese medicine

D-Dimer, erythrocyte sedimentation rate (ESR), ferritin, fibrinogen, high-sensitivity-CRP, interferon gamma (IFN-γ), interleukin-2 receptor

NEU%, neutrophil-to-lymphocyte (N/L) ratio, and white blood cell (WBC) counts. Organ functions, including myocardial, liver, kidney, and Sequential Organ Failure Assessment (SOFA) score. Medication status, including Chinese medicine, antibiotics, anti-viral drug, anti-inflammatory drug etc. Emotional wellbeing and quality of life outcomes including 12-Item Short Form Survey (SF-12)