Electronic cigarettes (e-cigarettes), the electronic devices that simulate tobacco smoking, consist of an atomizer, a power source of battery generally, and a container such as a cartridge or a tank. An e-cigarette is different from a combustible cigarette in that the user inhales vapor generated by the atomizer, which heats the e-liquid. Therefore, using e-cigarettes is often called “vaping.” In addition, the various appearances and flavors of e-liquid are more appealing than conventional cigarettes, especially to teenagers and young adults (1). Invented by a Chinese pharmacist in 2003 (2) and further improved in the following years, e-cigarettes were exported all over the world in 2006 (3). Thereafter, the sale of e-cigarettes has been increasing yearly (4) and is currently a multibillion dollars industry (5, 6).

The e-cigarette vapor contains propylene glycol, nicotine, glycerin, flavors, traces of nitrosamines (7), other toxicants, carcinogens (8), heavy metals, and nanoparticles (9). The exact composition of these components varies and depends on several factors, including the manufacturer and user behavior (10). E-cigarettes are considered less harmful than conventional cigarette smoking (8, 11), as they contain fewer toxic chemicals and lower nicotine concentration. However, public concerns have increasingly emerged over the new health problems that e-cigarettes may cause. Most researchers believe that e-cigarettes contain harmful chemicals not found in tobacco smoke, which may lead to some healthy effects on the cardiovascular system, respiratory tract system, and pregnancy (12–14). In the past decade, a number of e-cigarettes have hit the market, rapidly gaining consumers, especially among the younger population, for example, cases of vaping in America, Poland, and Hungary (15) rose by 900% between 2011 and 2015 (16). Since e-cigarettes have increasingly drawn more researchers' attention worldwide, thousands of studies related to e-cigarettes have been published so far. However, bibliometric analysis for this field is limited. It enables us to unpack the evolutionary research topic of e-cigarettes while shedding light on the emerging areas in this field. Bibliometrics is widely used in many fields, such as exploring the focus shift in Covid-19 literature (17), analyzing the evolution of service networks for sustainable business (18), revealing the adoption of new technologies in different areas (19), or providing a science mapping of emerging contaminants in the wastewater (20).

Bibliometric analysis is a popular method for exploring and analyzing large volumes of scientific data. With computer assistance, this method can investigate core research, author, as well as a hot research domain in a specific field. A study by Solla Price reveals variation in sciences which shows that the differences between sciences arise from differences in the process by which scientists cite each other's results. Only 2 percent of published papers are cited more than 5 times among a small group of authors. This small part of the papers supports a smaller part of the new literature which shapes the active research front (21). To understand the research trends of publications, the context of knowledge and relationship among researches can be displayed methodically through visualized quantitative analysis for scientific citations. Therefore, this study aimed to (1) analyze the distribution of publications by years, authors, countries, institutions, and keywords on e-cigarettes research; (2) identify the cooperation of countries and institutions; (3) and explore the existing hot topics and prospects of e-cigarette research.

This study aimed to provide a comprehensive bibliometric analysis of e-cigarette research from 2000 to 2021. Based on the purposes, a flowchart of the research framework was developed and shown in Figure 1. Considering the data analysis, two software tools were applied, VOSviewer (version 1.6.9) and CiteSpace (version 5.8 R2). VOSviewer tool can for example be used to construct maps of countries or institutions based on co-authorship data or construct maps of keywords based on co-occurrence data (22). It used the mapping technique of visualization with superior functionality that processes at least a moderately large number of items. CiteSpace was characterized by detecting and understanding emerging trends and abrupt changes (23). Two complementary visualization views were designed, including cluster views that reflected the co-citation clusters of prominent trends for emergent research-front terms, and time-zone views that reflected the abrupt changes of keywords (24).

This study explored the landscape of 7,979 publications on e-cigarette research from 2000 to June 2021. The results showed that most publications (57.81%) were articles and almost all of these documents were published in English (98.22%). Research related to e-cigarettes has been increasing rapidly during the last 6 years (from January 2016 to June 2021), with a total of 6,773 publications (84.89%) being done within this period. Of the 19,837 authors who contributed to all researches, only 1.70% published more than ten documents, while 71.7% of them published only one document.

The consumption of e-cigarettes has increased since it was introduced into the market in 2004 (27). In 2011, there were seven million e-cigarette users worldwide. However, the number rose to 41 million by 2018 (28). The trend of publications on e-cigarettes showed that there were sporadic researches before 2010. However, the number of publications soared afterward. This phenomenon may be due to increased awareness of the health concerns of using e-cigarettes as they contain harmful chemicals and toxins. Although the popularity of e-cigarettes was rapidly increasing after their marketing, a comprehensive chemistry of the e-liquids was seldom disclosed by the manufacturers. Various debates and questions such as whether e-cigarettes should be regarded as harm-reduction device and whether e-cigarettes promote the use of conventional tobacco products among non-smokers added to the confusion of the safety of e-cigarettes. Therefore, large different regulations on e-cigarettes were implemented among countries and states, ranging from no regulation to banning them entirely (29, 30).

Our analysis showed that 94 countries involved the e-cigarette research with the United States playing a predominant role in this field. With the debated issue and popularized use of e-cigarettes, US Food and Drug Administration tried to ban e-cigarettes in 2012, but it lost the case at the court (31). In recent years, e-cigarette use has increased rapidly, especially among youth and young adults, who use e-cigarettes more than any other age group (32, 33). For example, the current use rate (used an e-cigarette in the past 30 days) among American high schoolers increased substantially from 11.7% in 2017 to 27.5% in 2019 (34). Owing to medical drug policies and the companies that manufacture e-cigarettes pushing for laws that support their interests, e-cigarette legislation is being debated in many countries. In Japan, e-cigarettes containing nicotine were banned for use as cigarette alternatives. Some other countries have licensed e-cigarettes as medical devices including South Korea and the United Kingdom. While national policies result in a decrease in the use of e-cigarettes currently (35), the prevalence is still alarmingly high with nearly one in five youths reporting current e-cigarette use.

The author analysis revealed that each of the top ten authors published more than 50 articles, which showed a small variation among them (range: 58 to 86). However, the citations of the top ten co-cited authors were categorized obviously into two sets, with the top three authors being cited more than 1,500 times by other scholars, and the citations of remainders ranging from 765 to 996 times. Co-cited author analysis had been a generic method in indicating intellectual structures of scholarly fields and inferring some of the characteristics of the corresponding scientific community. WoS database only took into account first authors in the definition of co-citation, which meant the first authors were considered as being co-cited when their papers occurred in the other publications' reference list. In our study, the author and co-cited author analysis exhibited the fundamental aspect-the most active scholars in the e-cigarette research for the past decades.

Similarly, the distribution of e-cigarette research in countries and institutions showed that 59.02% of the publications were done in the USA, and most of the institutions were also from the USA. Furthermore, the visualization network analysis of cooperation among countries showed that the USA, UK, and Australia had strong co-authorship in this field. However, there was no apparent cooperation to be found in the visualization network map of institutions. The prevalence of e-cigarette use has increased in recent years (27, 36) despite the uncertainty of long-term health effects and harm reduction treatment. This uncertainty created unique challenges for governments as they attempted to optimally regulate e-cigarettes in a way that maximized the public's health (29). Hence, countries and institutions have conducted numerous studies to evaluate the safety of e-cigarettes, including animal experiments, cell research (in vivo and in vitro), and population-based studies.

In our study, the most productive journals and co-cited journals were also investigated. Such Tobacco, Nicotine, and Respiratory journals played a predominant role in both top ten journals and co-cited journals. A co-cited journal is defined as a journal that is cited together by other scholars (37). In particular, Nicotine & Tobacco Research was the journal that has the most published articles on e-cigarette research and the highest number of co-citations. Top journals in the medical field such as BMJ and New England Journal of Medicine were also in the list of top ten journals or co-cited journals. In addition, the dual-map showed an intuitive, and easy to interpret the representation of citations made by a wide variety of journals and cited journals (38), providing an explicit gateway to integrate the citing and cited clusters which included four and five major subjects, respectively. The accelerated adoption of e-cigarettes and the increase in popularity among consumers of all ages were caused by the common belief that they were indeed safer alternatives to traditional tobacco products. Nowadays, there have been relatively accurate conclusions on many aspects of e-cigarettes (39), such as the chemical profile of e-liquid, toxicology, public health, regulatory issues etc. However, the potential pathogenic impact associated with e-cigarette and the effect on smoking cessation should be further investigated in future, thus providing a sound scientific basis for regulated policy.

In this study, the common keywords of publications were categorized into four clusters that represented four different domains of e-cigarette research. The keyword of “Electronic cigarettes” was at the core and closely related to keywords of “Smoking,” “Nicotine,” “Tobacco,” “adolescents” “smoking cessation,” and “perception.” The four integrated clusters resulting from co-occurrence analysis of the keywords involved all the topics aforementioned, suggesting that the research on e-cigarettes was extensive. The common belief that e-cigarettes were safer than combustible tobacco products was inconclusive. Although a large, randomized, controlled trial found that smokers who used e-cigarettes to promote their smoking cessation were less likely to start again for at least a year, compared with those who used other nicotine-replacement therapy (40), the benefit may be slight-75% of study participants had already failed to quit using the other cessation aids before enrollment, so it was very likely that they failed again in a long period. In addition to nicotine, e-cigarettes contained potentially toxic substances such as acetaldehyde, acrolein, formaldehyde, and traces of heavy metals derived from flavoring additives, which were possibly associated with carcinogenicity and teratogenicity.

Burst keywords can represent the change in the research front over time (23, 41). It provides a timeline-analysis method to visualize the development process of hot topics in a specific field (24). As shown in Supplementary Figure S3, there were apparent time-series phases in the detection of burst keywords, with the first burst period starting in 2010 and the second, third, and fourth period starting in 2012, 2014, and 2016, respectively. However, some keywords lasted longer than others, such as “Nicotine delivery system,” “Withdrawal,” “Trial,” “Safety,” “Cytotoxicity,” and “Benefit,” indicating that scholars were more concerned with different themes during different periods. Finally, the analysis of the top ten co-cited articles provided a perspective of searching literature that impressed other scholars in e-cigarette research. With the increase in popularity of e-cigarettes some study results published in high-impact sources were indeed helpful to the practice of decision-making and regulation policy of e-cigarettes. However, if some debated questions are not settled, then e-cigarettes should be treated on par with tobacco products and regulated likewise.

There has been a rapid increase in research related to e-cigarettes since 2016. Goniewicz ML and Polosa R are shown to be the top ten authors and co-cited authors, indicating that these two scholars are active and influential researchers in the e-cigarette research field. Moreover, the network maps for countries and institutions showed strong cooperation between the USA, UK, and Australia. The Nicotine & Tobacco Research journal was the source with the highest number of publications and co-citations. This study also revealed four main aspects of e-cigarette research and changes in research topics over time through co-occurrence and burst keywords analysis.

XL and TZ conducted the literature search and drafted the manuscript. TZ and LZ developed the search strategy and performed database retrieval. XL conducted data analysis, visualization, and interpretation. RZ, QL, and CS conceptualized the idea and study design revised the manuscript and verified the entry of key data. All authors agreed to be accountable for all aspects of this study and contributed to the article and approved the submitted version.

This study was supported by Special R&D Program Project of Chinese Academy of Se-enriched Industry, No. 2020FXZX05-01, Key Research and Development Program in Shaanxi Province, No. 2020SF-076, and the Subject Innovation Team of the Shaanxi University of Chinese Medicine, No. 132041933.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.