Complementary and Alternative Medicine for Long COVID: Scoping Review and Bibliometric Analysis

Prolonged symptoms after the clearance of acute coronavirus disease 2019 (COVID-19) infection, termed long COVID, are an emerging threat to the post-COVID-19 era. Complementary and alternative medicine (CAM) interventions may play a significant role in the management of long COVID. The present study aimed to identify published studies on the use of CAM interventions for long COVID and provide an overview of the research status using bibliometric analysis. The present scoping review searched MEDLINE, Embase, and Cochrane Library from inception until November 2021 and identified published studies on CAM interventions for long COVID. A narrative analysis of the study types and effectiveness and safety of the CAM interventions are presented and a bibliometric analysis of citation information and references of the included publications were analyzed using the Bibliometrix package for R. An electronic database search identified 16 publications (2 clinical studies and 14 study protocols of systematic reviews or clinical studies) that were included in the present study. Dyspnea or pulmonary dysfunction, quality of life, olfactory dysfunction, and psychological symptoms after COVID-19 infection were assessed in the included publications. The two clinical studies suggested that Chinese herbal medications were effective in relieving symptoms of pulmonary dysfunction. Bibliometric analysis revealed the current trend of research publication in this area was driven by study protocols written by Chinese, Korean, and Indian authors. Thus, the present scoping review and bibliometric analysis revealed that there are few studies published about the use of CAM for long COVID and long-term management for COVID-19 survivors. Original studies on CAM interventions, including randomized controlled trials and systematic reviews, are required to actively support evidence for their use in the management of long COVID. PROSPERO registration: this trial is registered with CRD42021281526.


Introduction
Since the first report of the coronavirus disease 2019 case  in 2019, the impact of the pandemic has continued for more than 2 years worldwide. Although the number of people vaccinated has greatly increased, the number of new COVID-19 cases is still increasing due to factors, such as the delta and omicron variants [1]. However, a new health problem that is different from the acute increase of new confirmed cases is emerging, namely long COVID, which is defined as prolonged symptoms after the clearance of acute COVID-19 infection [2]. e number of patients experiencing long COVID is significant and is expected to increase as the COVID-19 pandemic continues. It is important to prepare for the post-COVID-19 era, and the healthcare strategy for long COVID patients is critical.
Several organs and systems are speculated to be involved in long COVID, and the spectrum of symptoms ranges widely, from dyspnea, sequelae of lung inflammation due to severe acute respiratory syndrome coronavirus 2 infection to "brain fog" and cognitive impairment due to chronic damage of the central nervous system [2]. e symptoms of long COVID are diverse, and the mechanisms involved remain unclear, making it difficult to establish a treatment strategy for these patients [2]. Many countries have adopted complementary and alternative medicine (CAM) in their healthcare services during the COVID-19 pandemic [3][4][5], and surveys from different regions have reported a significant number of individuals using CAM interventions as a means to boost immunity and prevent acute COVID-19 infection [6,7]. CAM interventions are expected to play a significant role in the management of long COVID. erefore, understanding the efficacy and safety of CAM interventions is urgent and necessary for their implementation. e present study comprised a scoping review and bibliometric analysis to explore the status of research evidence for the use of CAM interventions for long COVID.
e study objectives were to identify publications and evidence for various CAM interventions for the treatment of long COVID and long-term management of survivors of COVID-19 infection. Furthermore, the current research status of long COVID was overviewed using bibliometric analysis.

Search and Selection of Published
Articles. Core medical databases, including MEDLINE, Embase, and Cochrane Library, were searched from their inception until November 2021 to identify publications related to long COVID and long-term management using CAM interventions for COVID-19 survivors. e following PICO components were used for the literature searches: Population: patients with long COVID symptoms, such as fatigue, mild cognitive dysfunction, anosmia, dysgeusia, and other pulmonary dysfunction or neurological or psychological problems after recovery from acute COVID-19 infection.
Intervention: any type of CAM intervention, including traditional medicine, mind-body therapy, homeopathy, and dietary supplements. Limitations were not imposed on the type of comparators and outcomes of the included publications.
Publication type: any type of publication, including systematic reviews (SRs), randomized controlled trials (RCTs), and observational studies, as well as study protocols were included for human studies. e search strategy for each database was developed according to the specific features of each database using the keywords "CAM interventions," "long COVID," and "related symptoms" (Supplementary File 1). e titles and abstracts of the publications were independently reviewed by two authors (S-RK and T-HK), and hard copies of potential articles were assessed for selection. Decisions about selection of the publications were made in the following discussion.

Data Extraction.
In the present scoping review, data of publications, including condition, type of publication, type of the study, number of patients, country of the authors, intervention types, and description of interventions (frequency, duration, potential effectiveness, and safety) were extracted from the included publications. To conduct a more productive analysis, we searched the Web of Science (WoS) database for each study which was located through MEDLINE, Embase, and Cochrane Library database searching. Bibliometric analyses were conducted by retrieving citation information and references of the included articles in the WoS database format. Citation information included keywords plus, which is a specific index term for each publication defined by the WoS [8]. Two authors (S-RK and T-HK) independently extracted data, and the third author (SOK) made a final decision about instances of disagreement.

Bibliometric Analysis.
Bibliometric analyses were conducted by narratively analyzing information about the included publications. Relevant sources (journals), top 20 authors, relevant institutions, country of the corresponding author, country contribution, keywords, and title words were narratively analyzed. e current research trend was identified using a three-field plot, which links the title, authors, and country of the corresponding author in all included publications. Word cloud, which shows the most relevant keywords of the publication, was also used. Trends in publications by year of publication were not analyzed because the COVID-19 pandemic occurred after 2019. e degree of collaboration, which can be calculated as the ratio of multiauthored articles to the sum of single and multiauthored articles, was suggested for identifying how active research collaboration was during the last two years. To access the pattern of co-authorship, we calculated the coauthorship index (CAI). In addition to this, the collaborative coefficient (CC) was calculated for assessing dominant patterns of single-authored or multiauthored publications [9]. We analyzed publication data using the Bibliometrix package for R (version 3.1).
Between 2020 and 2021, the degree of collaboration was calculated to be 1, which means that all the published articles had multiauthors and there was no single-authored article.
e CAI was 100 which implied that the pattern of coauthorship of the included literature between 2020 and 2021 was similar to that of the world average. e CC was calculated to be 0.8607, which suggested that multiauthored publications were dominant in this research area.

Discussion
e present scoping review and bibliometric analysis revealed a lack of published studies about use of CAM on long COVID and long-term management of COVID-19 survivors. Only 16 publications were available, including two observational studies that examined the effects of CHM treatment for recovery from lung inflammation after acute COVID-19 infection and 14 protocols for RCTs and SRs. In addition, the clinical effectiveness of CHM decoctions for lung inflammation was found to be suggested as limited evidence. Most prevalent and severe long COVID symptoms including brain fog, fatigue, and olfactory dysfunctions were not tested or were not evaluated, which suggested that urgent evaluation would be necessary for this area. From the bibliometric analysis, we found that China was the most active country where most research literature has been published already. e most frequently published journal was "Medicine," which may be because it publishes original research protocols. Several indicators for co-authorship suggested that multiauthored publication was dominant between 2020 and 2021. Frequently tested interventions were CHM, acupuncture, and Tai Chi. ese results indicate that research on CAM interventions for long COVID by Chinese researchers has focused on TCM interventions. Furthermore, pulmonary rehabilitation and improvement of quality of life were reported to be the main target of the CAM interventions assessed in these publications. In addition to this, the current research trend in this area was mainly driven by Chinese, Korean, and Indian authors, and the main publication types were protocols of RCTs or SRs for long COVID symptoms. is suggests that more time will be required until the evidence for CAM treatments to handle these long COVID symptoms to be prepared. e present study has some limitations. First, only a small number of publications are currently available; therefore, any evidence for the effectiveness and safety of CAM interventions for long COVID cannot be suggested in the present study. Long COVID includes a range of symptoms; however, the publications included in the present study only covered a limited number of symptoms. RCTs and SRs for various symptoms of long COVID are required in the future. Second, most of the publications used protocols that may reflect the current research status. However, we did not assess the registry for clinical trials and SRs, which contained research protocols. e primary objectives of the present study were to assess the publication status of this research area and conduct a bibliometric analysis; therefore, we did not include a search of the registries. Future studies should include clinical trials and SR registries to include more diverse studies.
ird, there could be publication bias due to the limited searching strategy of this study. We only assessed core databases, and there may be more studies that have been published in local databases.
ese points need to be considered when interpreting the results of our study.
Long COVID is an emerging health problem, and the socioeconomic burden of this condition is expected to be severe [2]. erefore, an appropriate therapeutic plan needs to be prepared for the post-COVID-19 era. e use of CAM interventions for various diseases and conditions is increasing worldwide; however, the evidence supporting their appropriate use remains insufficient [27]. As COVID-19 and long COVID are new health problems, there is a lack of evidence in the field of CAM. Considering that CAM interventions are expected to be used frequently for long COVID, the substantial gap between knowledge and practice should be resolved through rigorous clinical studies and RCTs.
In conclusion, there is a lack of published studies about the effectiveness and safety of CAM interventions for long COVID. Original studies on CAM interventions, including RCTs and SRs, are required in the future to provide evidence for their use in this condition.

Data Availability
All data of this study are included in the appendix file.

Conflicts of Interest
e authors declare that there are no conflicts of interest regarding the publication of this article.

Authors' Contributions
T-HK, S-RJ, SK, and JWK conceived the study design. T-HK, S-RJ, SK, and JWK conducted the literature searching and bibliometric analysis. T-HK, S-RJ, SK, and JWK wrote the final draft of this manuscript.