Based on the research scope and objectives, the following four research questions formulated:

A bibliometric study enables researchers to explore patterns, trends, associations, and scientific developments related to searched domains, along with interrelated fields over publication data. This analysis requires a structured bibliometric database to obtain appropriate data for answering research questions [37-39].

Bibliometrics is also defined as a statistical method to analyze bibliometric publications data over a wide spectrum such as peer-reviewed journal articles, books, conference proceedings, periodicals, reviews, reports, and related reports. There are various analysis methods for a literature review along with bibliometric tools [37,40-43]. This approach allows for further obtaining more in-depth understanding of a given topic and its publication trends.

In this study, we employed a bibliometric approach for obtaining descriptive publication results, author/coauthor productivity metrics, source impact analysis, along with keyword and most common co-word (text) analysis [38,44-47].

A word cloud, also known as a “tag cloud,” is a visual representation of text data [39] from various keywords or any given text material [42]. According to the Web of Science dataset structure, a word cloud has four main categories to analyze: abstract, title of the paper, author keywords, and keyword plus (see Multimedia Appendix 1). Depending on the frequency of text data regarding the main categories, the significant terms and tags are highlighted, which are usually single words represented by a single font size and color based on their relative importance. Bold and larger-sized words indicate that the word has more importance and has attracted researchers’ increasing attention in the subject domain field [48]. Keywords or any other text datasets among these four categories were collected from the articles to conduct the co-word analysis and to construct a word cloud to illustrate the power of words based on their frequencies in the literature [49].

A bibliometric study requires a structured database to analyze publication data. The main two bibliometric databases available for this purpose are ISI Web of Science and Scopus. ISI Web of Science provides data on the highest ranked and impactful (prestigious) sources, whereas Scopus also ranks the same sources in addition to other sources with wider coverage, including conferences, symposia, and congress proceedings. In addition, PubMed is a commonly used database in the medical field. Journals focused on mHealth mostly rank in the highest quartile (Q1) in Web of Science as well. Therefore, we chose to focus on these highly ranked (Q1) and high-impact sources in Web of Science to maintain consistency of citations in a single database with SCI-Expanded and SSCI indexing. We only used PubMed for comparison purposes and as a secondary source of indexing. All databases have their citation count categories. The citation results are significant to keep all publications within the same quartile and for consistency of comparisons at the same level (Q1 ranking and indexing). Finally, publication data were retrieved from the Web of Science database using the search strategy shown in Textbox 1 [9,37,38,40,42].

^a64 were papers found in the Web of Science list, including 40 matching papers in PubMed.

The data were retrieved as plain .txt, .xls, .csv, and .bib file formats for further analysis. The Microsoft Excel and R Language (version R x64 3.6.1) R Studio software with the “bibliometrix” package [39] were used for descriptive and bibliometric data analysis, respectively [48].

A total of 64 publications for the research domain were retrieved from the ISI Web of Science database (Multimedia Appendix 2). In comparison to the PubMed results, 40 papers were found in the same dataset of the Web of Science–retrieved publications. The vast majority of the papers were written in the English language (63/64, 98%) with the remaining article written in French (1/64, 2%). Overall, there were 323 authors/coauthors from 11 different countries, with the United States being the most common, followed by England, Canada, Italy, the Netherlands, Australia, France, Germany, Mexico, Sweden, and Vietnam. The descriptive summary of these publications showed that the majority were peer-reviewed journal articles (48/64, 75%), followed by reviews (8/64, 13%), meeting abstracts (5/64, 8%), news items (2/64, 3), and proceedings (1/64, 2%). There was a mean number of 6.4 papers published on the topic per year, with a 100% increase in the number of publications found for the period of 2016-2019, along with a corresponding increase in the mean number of papers published per year during this period to 13.33.

Web of Science-Core Collection subject category data were used to categorize the related research domains under the top 12 major subjects, which are summarized in Table 1, comprising topics with at least 3 publications. Other minor subject categories with 2 publications included Computer Science Information Systems and Neurosciences, and those with 1 paper each included Computer Science Interdisciplinary Applications, Education Educational Research, Environmental Sciences, Health Policy Services, Information Science Library Science, Medicine Research Experimental, Nursing, Oncology, Pediatrics, Psychology Experimental, Rehabilitation, and Respiratory System.

Figure 1 presents the publication records by year starting from 2010 through to the end of the decade in 2019. There was only one publication recorded for 2010 and 2011, followed by a slight increase from 2012 to 2013. From 2013 to 2014, the number of publications increased sharply from 2 to 5, with a steady increase until 2016 (n=5 to n=7). A sharp increase was observed from 2016 (n=7) to 2019 (n=14), representing a 100% increase in the publication count. The last 3 years reflect increasing research interest in the topic, as the period with the highest count in the dataset. The mean number of publications per year for the entire period was 6.4, increasing to 13.33 from 2016 to 2019. The publication distribution also supports this upward movement as shown in Figure 1.

Table 2 provides the descriptive results for authors and coauthors of the retrieved publications. Only 6 papers were published as single-authored documents, whereas 317 papers were multiple-authored documents, and there were 323 different authors with 366 appearances.

Among the top 20 most productive authors, Kuhn (n=7 records) was identified as a leading author; followed by Beckham (n=5 records); Calhoun (n=4 records); and Dennis, Marx, and Moore (n=3 records each) in the top category. Others at the same level with 2 papers each include Acierno, Carpenter MJ, Carpenter VL, Dahne, Dedert, Dennis, Diaz, Felton, Hertzberg, Hoffman, Jovanovic, Kirby, and Kizakevich.

Table 3 shows the top 10 most productive countries that contributed to the research domain fields. The top country was the United States, followed by Germany, Italy, the Netherlands, United Kingdom, Australia, Canada, France, Mexico, and Sweden.

Table 4 shows the contributed institutions from various countries around the world. Harvard University and Stanford University emerged as the leading institutions with 8 papers each, and 13 universities ranked at the top of the dataset among the total 156 contributing institutions.

The citation report of the 64 publications derived from the Web of Science Core Collection statistics between 2010 and 2019 showed that the H-index was 12 and the average number of citations per item was 7.9. The sum of times an article was cited in total was 509, which was reduced to 471 after excluding self-citations. The number of cited articles was 421 in total and was 402 after excluding self-citations. Figure 2 shows the total citation distribution throughout the research timeframe. Although the citation statistics are available as of 2012, an increase was only observed as of 2015 (n=28), followed by a steady increase until 2019. From 2016 to 2017 and from 2017 to 2018, the citation counts nearly doubled, reaching 217 by 2019.

Table 5 presents the top 10 journals by citation counts. Olff [2] was the leading paper with the highest number of citations (n=58) during the search period. The top 10 citations comprised 299 of the total 509 citations (58.7%) starting from 2012 to 2019. There was a mean of 29.9 per year. Therefore, 4 papers were above the mean: Olff [2], Hertzberg et al [19], Van Ameringen et al [50], and Repetto et al [29], published in 2015, 2013, 2017, and 2013, respectively. Although the papers only started to be cited as of 2012, the citation counts showed a substantial (3-times) increase from 2014 to 2015 (from 6 to 20 counts). The most citations were recorded in 2019 (n=108).

The leading journal publishing these studies was Journal of Medical Internet Research (JMIR) with 3 articles, and the remaining journals only included 1 article each published in the top 10 list. The majority of these papers were related to mobile technology and apps, PTSD-related topics, virtual reality related to mobile devices, treatment, and other psychological disorders.

Table 6 shows the publication frequency for the most relevant sources. Although there were no notable differences in publication history, the top contributing journals of 8 sources (n≥2 records) are listed in Table 6.

The top publication sources distributed and sorted from the highest on the list with 4 (19.36%) records were European Journal of Psychotraumatology, JMIR mHealth and uHealth, and Military Medicine. The next most common sources with 3 (14.52%) records were JMIR, Professional Psychology Research and Practice, and Psychological Services, whereas Applied Psychophysiology and Biofeedback and Depression and Anxiety included 2 records each (6.45%).

There were 39 single-publication sources (n=1, 59.68%) among the total 64 articles in the dataset retrieved: Addiction, Addictive Behaviors, American Journal of Preventive Medicine, Annales Medico Psychologiques, Asian Journal of Psychiatry, Behavior Therapy, Biological Psychiatry, Computers In Human Behavior, Current Psychiatry Reports, ETR D Educational Technology Research and Development, Health Informatics Journal, Implementation Science, International Journal of Clinical Practice, International Journal of Environmental Research and Public Health, International Journal of Methods In Psychiatric Research, International Journal of Psychiatry In Medicine, Jama Journal of The American Medical Association, JMIR Medical Informatics, JMIR Mental Health, Journal of Affective Disorders, Journal of Child And Adolescent Psychopharmacology, Journal of Clinical Psychiatry, Journal of Dual Diagnosis, Journal of Head Trauma Rehabilitation, Journal of Investigative Medicine, Journal of Medical Systems, Journal of Pain, Journal of Psychiatric Research, Journal of Psychosocial Nursing And Mental Health Services, Journal of The American Medical Informatics Association, Journal of Traumatic Stress, Nicotine Tobacco Research, Personal and Ubiquitous Computing, Psychiatry Interpersonal and Biological Processes, Psycho-Oncology, Social Psychiatry and Psychiatric Epidemiology, Global Mental Health, Systematic Reviews, and Thorax.

The JMIR group of journals had the highest amount of publication records (n=9) for the research domains in the search period. In addition, the source impact metric (H-index, n≥2) of the journals ranked as follows: JMIR mHealth and uHealth (H-index=4); Military Medicine, JMIR, and Professional Psychology Research and Practice (H-index=3); and European Journal of Psychotraumatology, Psychological Services, and Depression and Anxiety (H-index=2).

Keyword analysis is used to reveal the frequency of keywords in publications. There are two types of keywords: author keywords and Keywords Plus. The author keywords are the words (terms) that authors prefer to use for their papers and Keyword Plus includes terms from the preset list of related science domains, which is created by the editorial experts of Web of Science.

Table 7 shows the most relevant keywords used in the publications associated with mobile technologies and apps for PTSD treatment. The most common author keywords included “m-health,” “PTSD,” “mobile health,” and “depression.” The most common Keyword Plus terms included “posttraumatic stress disorder,” “PTSD,” “randomized controlled trial,” and “depression.”

The word dynamic-growth graph (Figure 3), prepared with the top keywords, was used to evaluate the keyword dynamics over the research period. The repetition trend of each word (ie, the frequency of appearances in the dataset over the search period) represents occurrences. The graph shows the trend direction to analyze either upward or downward movement over the linear line according to the annual distribution of keywords. The most popular terms and keywords can be tracked during the period to understand trends in subject domain interest and importance in the research field. Identifying topics of growing interest helps researchers to concentrate on new subject areas and can also provide valuable results to contribute to these fields.

Figure 3 illustrates the word dynamics-growth for the research domains. Although the dataset shows that publications started in 2012, the interest and popularity became significantly more visible after 2015, followed by a sharp increase in the subject areas until the middle of 2017. In those years, “PTSD,” “posttraumatic stress disorder,” “smartphone,” and “randomized controlled trial” had peak levels, but the usage rates of these terms declined from the middle of 2017 to 2019. At this point, the terms “depression” and “prevalence” became highly popular with a sharp increase on the graph, followed by “symptoms” and “meta-analysis.”

Multimedia Appendix 1 illustrates four different world cloud–based data distributions. The top left word cloud is for abstract text data, whereas the upper right word cloud is for author keywords text data. Similarly, the bottom left word cloud is from Keyword Plus text data and the bottom right word cloud shows a paper title text data representation. As can be seen in each word cloud, the highlighted words differ in terms of importance for each group. In particular, “health,” “mobile,” and “apps” were of the highest importance, followed by “mental,” “treatment,” and “PTSD” in the abstract word cloud. By contrast, in the author keyword cloud, “m-health” and “PTSD” were both dominant keywords, followed by “depression,” “mobile health,” and “posttraumatic stress disorder.” The word order changed for Keyword Plus to “depression” and “PTSD” at the top, followed by “care,” “smartphone app,” and “symptoms.” The title word cloud was comparable to the others with “mobile” being far more important than other key terms as a highly dominant term, followed by “health,” and the terms “stress,” “app,” “disorder,” “mental”, “posttraumatic,” and “application” appeared in another layer.

Mobile apps are recognized as efficient tools in the assistance of both health care patients and staff [1,3]. A mobile platform allows for predicting the symptoms of depression and PTSD [51] to collect data and conduct research on health interventions using the technology [53]. Mobile technology in the treatment for military staff and veterans with PTSD and other mental disorders has been reported to be feasible [8,15]. Engagement with mHealth apps in the transition period improved communications [18] and mental well-being [52]. Several mobile apps have been considered to be useful supportive tools in the treatment of managing pain [26,27], stress reduction [54], anger [25], and mental disorders, including PTSD, depression, anxiety, and addictions [50,55], as well as in identifying suicidal behaviors [7].

The efficiency of mobile apps in the psychological treatment of mental disorders, including CBT [9], cognitive rehabilitation [28], exposure therapy [10,29,30], and visualization [56] has been verified. The use of mobile apps allows for evaluating CBT precision [34], to conduct a cognitive assessment [35], to deliver a PTSD checklist [36], and to monitor mental disorders [33].

By analyzing the results of the dataset regarding the publication pattern on research related to mobile technology and apps associated with PTSD treatment, we found a rapid increase and growth of subject interest in the last decade. The trend sharply increased in the most recent years, from 2016 to 2019.

This growth of productivity could reflect the improvements, functionality, and developments of mobile technology and apps in parallel with comparison to other areas of usage. Therefore, these technologies have become the center of human life to provide new opportunities, convenience, and address potential benefits. Despite the negative insights and perceptions of such technologies, this would be promising for patients to adapt their health behaviors as a supportive tool with their clinical treatments. Cooperation among clinical experts, app designers, and technology providers is necessary to reach the ultimate goal and objectives, which should concentrate on patients’ needs and treat them positively and objectively.

This analysis was conducted to highlight the most frequent subject categories, along with popular keywords and terms. These aspects were reflected in keywords, and the same terms were similarly represented in the text of the abstract and title of the paper. The text analysis showed the critical terms used in this field of research and also represents the popularity of subdomain searches. The aim of this study was to discover the publication trend and to identify the critical areas in the dataset to ultimately provide insights and research directions for academics, practitioners, and readers who wish to collaborate in these domains in the future.

According to the papers retrieved and analyzed (N=64), the majority were peer-reviewed journal articles (75%), with a mean of 6.4 publications per year from 2010 to 2016, which then sharply increased to 13.33 (doubled) between 2016 and 2019. The most productive countries were the United States, with far greater representation than any other country, followed by Germany, Italy, the Netherlands, and the United Kingdom. One of the main reasons for this difference is attributed to differences in military engagement in various regions. For example, in the Middle East and Asia, soldiers return to their countries with different intensities of PTSD symptoms and other comorbid disorders.

The results also revealed the distribution of the publications, demonstrating that the top category sources support previous arguments; namely European Journal of Psychotraumatology, JMIR mHealth and uHealth, and Military Medicine were among the highly popular sources for these publications. The JMIR group of journals emerged as the leading sources (n=9, 14% coverage) compared to other single-publication sources. However, the vast majority (60%) of sources were equally distributed among the 39 single-publication sources. Similar results were found in the authorship analysis, in which the top contributors in the field are Kuhn (n=7 records) as the leading author; followed by Beckham (n=5 records); Calhoun (n=4 records); and Dennis, Marx, and Moore (n=3 records each) in the top categories.

Co-word (text) analysis showed that the most common vital terms overall (for the four different word clouds created) were “mobile,” “PTSD,” “posttraumatic stress disorder,” “m-health,” “depression,” “health,” “treatment,” “smartphone apps,” and “mobile health.”

Despite growing interest for the research domains, no publication was identified that analyzed the state of the field with a bibliometric approach. Therefore, the main strength of this study could be considered as the uniqueness of the research design itself. This study is the first bibliometric-related research in the domain. The contribution of the study is revealing the scientific patterns and future research gaps to academics and practitioners. The text analysis also highlighted and supported popular subject areas to clarify the research scope and future directions.

One of the limitations of this study is that we used only the Web of Science Core Collection database in comparison to PubMed. A single database was selected to ensure a simple and accurate analysis, and to effectively eliminate duplications and avoid errors. In this regard, Web of Science covers the highest impact journals and has unique indexing and ranking with its citation categories. The research domain of health and mobile internet–related publications is ranked in the SCI-Exp index. Another limitation is that only documents published in the English language were selected. Although various bibliometric analysis methods are available, given the scope and size of this topic, we decided to concentrate on more specific analyses such as descriptive statistics regarding the dataset from 2010 to 2019.

According to the findings, the research domains are prevalent, and growing interest can be seen as an upward trend in the publication records since 2016. In particular, the majority of the subject category records were found in Psychiatry and Psychology, especially in the clinical and multidisciplinary domains, followed by Health Care Science Services, Medical Informatics, and Medicine General Internal. Further research is needed adopting various aspects of bibliometric analysis. More empirical and case studies should also be conducted in parallel with the improvement of technology and apps perspectives that would be tested and clinically validated.

However, the research analyzed indicates the importance of further explorations to develop appropriate and feasible mobile technology for PTSD treatment. The necessity to manage the challenges related to the development of mHealth tools were underlined [2]. Establishing the standards for creating PTSD-related mHealth apps and following them seem to be essential in transferring mobile apps to the clinical field [4]. Moreover, practitioners need to explore the factors facilitating and limiting the effective use of mHealth for PTSD treatment [57,58]. Cooperation between mobile app creators, researchers, and practitioners is essential in creating new technology that will match the needs and expectations of both health care staff and patients [50]. Finally, the majority of the available mobile apps require more clinically validated evidence of their efficacy before they can be adopted in the psychological treatment of PTSD [33].

This study explored and analyzed the scientific patterns and relations of scholarly publications related to the use of mobile technologies in PTSD treatment. We therefore provide a general overview of the field based on co-word (text and keyword) analysis of research domains, and various forms of bibliometric methods were employed along with a data visualization approach to establish a clear picture. The analysis included 64 papers published between 2010 and 2019.

The results identify the most frequent subject categories, popular keywords, critical terms, and the popularity of subdomain searches. With this study, we attempted to investigate the patterns of publications to provide insights and research directions for academics, practitioners, and readers who wish to collaborate in these domains in the future. The data highlight the significance of further explorations in this field to improve mobile technology for PTSD treatment. Conducting studies and analyzing the practical use of these tools will improve the technology and apps that would be tested and clinically validated.