Global Scientific Trends on Anxiety Disorders in the Early Twenty-First Century: A Data-Driven Bibliometric Analysis
Abstract
Objective: Interest in the study of anxiety disorders is growing rapidly. However, there is still a lack of comprehensive bibliometrics research in measuring and analyzing global scientific publications. The aim of this article is to review the current research status and trends of anxiety disorders worldwide through bibliometrics and to provide reliable support for future in-depth research.
Methods: In this bibliometric study, we utilized VOSviewer to evaluate the Web of Science Core Collection (WoSCC) publications and to build visualizing maps to analyze the research progress on this topic between January 1, 2000, and August 7, 2024.
Results: Up to August 7, 2024, the study included a total of 67,386 English original articles and review articles published in 4054 journals by 206,496 authors from 164 countries/regions. The United States was a leader in articles, researchers, and institutions in the field of anxiety disorders. Using cluster analysis, the keywords were divided into five major clusters: (1) etiology and pathogenesis, (2) clinical classification and interventions, (3) comorbidity, (4) anxiety caused by the COVID-19 pandemic and specific groups, and (5) anxiety disorders in children and adolescents.
Conclusion: Molecular and neural circuit mechanisms, adolescent psychological problems, objective indicators of diagnosis and classification, and technology-assisted therapy are still the focus of future research. The findings provide a comprehensive overview of anxiety disorder research, which may help physicians, researchers, and local authorities and community health staff to more fully understand trends and influence in the field and highlight under-researched areas, which could be the basis for future research.
Summary
- •
This is a comprehensive bibliometric analysis in the field of anxiety disorders.
- •
The current research status and hotspots of anxiety disorders are discussed.
- •
The United States leads the way in anxiety disorders.
- •
This study predicted the research interests of anxiety disorders in the future.
1. Introduction
Anxiety disorders are one of the most prevalent and debilitating psychiatric disorders worldwide, characterized by persistent, excessive anxiety and worry. There are many forms of anxiety disorders, according to the symptoms and form of attack, can be divided into the following categories: generalized anxiety disorder, social anxiety disorder, separation anxiety disorder, panic disorder, and agoraphobia [1]. Although the global research on anxiety disorders has never stopped, there are still many unresolved problems, such as (1) the etiology and pathogenesis are still unclear, (2) diagnosis still relies on subjective methods and lacks biomarkers for accurate diagnosis, and (3) treatment effects are poor. Furthermore, anxiety disorders are often recurrent and associated with considerable morbidity and comorbidity, excessive disability, increased risk of suicide, and significant costs worldwide, which have become one of the difficulties in clinical treatment.
The global prevalence of anxiety disorders ranges between 3.8% and 25% [2] and can affect individuals of all ages [3]. No reduction in the global prevalence or burden of anxiety disorders has been detected since 1990 [4]. The age-standardized disability-adjusted life year (DALY) rate for anxiety disorders increased the most between 2010 and 2021 (16.7%) [4]. Unlike chronic physical illnesses, anxiety disorders have their strongest foothold in youth [5] but unfortunately remain largely unrecognized and untreated [6]. The 3-year global epidemic of the COVID-19 poses an unprecedented and potentially long-term threat to mental health, triggering a range of global issues that need to be addressed accordingly [7]. How to reduce the high burden and prevalence of anxiety disorders is a topic that researchers have been exploring. Over the past 20 years, a huge amount of research articles have appeared on anxiety disorders, published in thousands of journals and categorized into hundreds of disciplines such as psychiatry, neurology, behavioral sciences, education, health policy services, cell biology, immunology, genetics, pediatrics, and gerontology. An objective and comprehensive understanding of the historical overview, research hotspots, and development trends of anxiety disorders is very meaningful and necessary for every researcher and healthcare policy maker working on anxiety disorders.
Over years of development, a large number of research articles have been published on anxiety disorders, and it is difficult to manually extract their information to conduct a comprehensive analysis of the research in this field. Bibliometric analysis is an interdisciplinary field that combines mathematics, statistics, and linguistics to quantify and visualize research trends. It can quickly organize a large number of published articles, with the advantages of simple, rapid, comprehensive, and accurate, which is an effective tool to objectively study the current situation of the subject and reflect the development of the subject [8, 9]. Although there have been several bibliometric studies to explore diagnosis, treatment, and comorbidities of anxiety disorders [10–12], there is still no comprehensive study to summarize and analyze the research trends in the whole field of anxiety disorders.
As a result, we conducted a comprehensive bibliometric analysis to assess research distribution, trends, and active topics of anxiety disorders, as well as to analyze and predict future research hotspots on anxiety disorders, which is critical for researchers, funding agencies, and policy makers.
2. Method
The Web of Science Core Collection (WoSCC) of Thomson Reuters contains more than 12,000 influential academic journals and is recognized globally as a critical academic resource [13]. We conducted an advanced search in the Social Sciences Citation Index (SSCI) and Science Citation Index Expanded databases (SCIE) within WoSCC on August 7, 2024. To ensure a comprehensive search, we employed Medical Subject Headings (MeSH) to identify all relevant terms associated with anxiety disorders. The publication was included when the search term appears in any of the titles, abstracts, author keywords, and keywords plus of the publication. The full search terms used can be found in the electronic Supporting Information file (see online Supporting Information Methods). The flowchart of the study was shown in Figure 1. We refined the time span of publication from January 1, 2000, to August 7, 2024, and excluded nonarticle document types and non-English language publications. To ensure consistency and avoid potential bias from database updates, we downloaded the full bibliometric data, categorized by year, on August 7, 2024. The downloaded bibliometric data included bibliometric index data such as title, abstract, journal, author, institution, country, citation times, and publication year.

Visualization tools for bibliometric analysis include HistCite, RefViz, DIVA, VOSviewer, CiteSpace, and so on [14]. VOSviewer was developed by Van Eck and Waltman of Center for Science and Technology Studies (CWTS). VOSviewer is extensively utilized in bibliometric and citation studies due to its robust visualization and data processing capabilities [15]. We used VOSviewer version 1.6.20 to perform bibliometric analysis in this study. We performed citation analysis to evaluate the influence of journals, creating visual maps of high-yield journals. Additionally, we performed a coauthorship analysis for both authors and institutions to uncover collaboration networks, revealing prominent researchers, major research hubs, and the strength of institutional partnerships, with results visualized in detailed network maps. We performed co-occurrence analysis of keywords, identifying frequently co-occurring keywords, and mapping their relationships to reveal dominant research themes and emerging topics. Furthermore, we utilized VOSviewer to extract bibliometric data on publication years, journals, countries/regions, institutions, authors, and highly cited articles, performing a descriptive analysis.
3. Results
3.1. General Trends in the Publications and Citations
We included a total of 67,386 publications related to anxiety disorders, three times the number of articles included in previous study (22,267) [11]. The reason is that Liu et al. [11] only obtained publications from the Science Citation Index subdatabase which focused on the natural sciences and did not include publications in the SSCI subdatabase. The chart showed the trends in annual publications and citations from 2000 to 2023 (2024 was not a full year and was not shown in Figure 2).

From 2000 to 2023, the annual publication volume and citations of publications related to anxiety disorders showed a significant growth. From 2000 to 2007, the number of publications per year was between 1000 and 2000, between 2000 and 3000 from 2008 to 2000, and 3000–4000 from 2013 to 2020, and the number exceeded 4000 from 2021 to 2023. 2022 was the peak year for the number of annual publications (reaching 4992), 4.99 times the number in 2000.
3.2. Distribution Characteristics of Countries/Regions
In total, 164 countries/regions globally participated in the anxiety disorder research. Table 1 presents the top 20 countries ranked by the number of publications. The United States (26,919 counts, 1,296,489 citations) ranked first in terms of the number of publications and all cited. It was followed by the United Kingdom (6824 counts, 301,752 citations), and the third was Germany (5612 counts, 254,276 citations). Among the top 20 countries, Belgium had the highest average citation (50.74 citations), followed by France (49.14 citations), and the third was United States (48.16 citations).
Rank | Country | Counts | Citations | Avg. citations | Avg. year pub |
---|---|---|---|---|---|
1 | United States | 26,919 | 1,296,489 | 48.16 | 2014.19 |
2 | United Kingdom | 6824 | 301,752 | 44.22 | 2015.26 |
3 | Germany | 5612 | 254,276 | 45.31 | 2015.43 |
4 | Canada | 5370 | 205,511 | 38.27 | 2015.55 |
5 | Australia | 5127 | 200,470 | 39.10 | 2015.57 |
6 | China | 5007 | 110,656 | 22.10 | 2019.40 |
7 | Netherlands | 4218 | 196,463 | 46.58 | 2014.43 |
8 | Italy | 2896 | 109,985 | 37.98 | 2014.90 |
9 | Spain | 2452 | 89,544 | 36.52 | 2016.23 |
10 | Sweden | 2072 | 82,142 | 39.64 | 2016.14 |
11 | Brazil | 2025 | 59,974 | 29.62 | 2015.29 |
12 | Turkey | 1842 | 24,637 | 13.38 | 2016.83 |
13 | Japan | 1615 | 53,596 | 33.19 | 2014.72 |
14 | France | 1575 | 77,394 | 49.14 | 2015.15 |
15 | Switzerland | 1472 | 65,968 | 44.82 | 2015.62 |
16 | Israel | 1412 | 54,111 | 38.32 | 2014.75 |
17 | South Korea | 1301 | 27,692 | 21.29 | 2017.01 |
18 | Belgium | 1203 | 61,046 | 50.74 | 2015.22 |
19 | Norway | 1091 | 37,419 | 34.30 | 2015.70 |
20 | Denmark | 991 | 37,062 | 37.40 | 2016.91 |
The average publishing year of 20 countries was generally between 2014 and 2016, among which the average publishing year of United States was the earliest, and the average publishing year of China was the latest, which ranked sixth in the number of publications, indicating that China’s research on anxiety disorder had increased significantly in recent years (Table 1).
Figure 3 illustrates the global distribution of the 164 countries/regions that participated in the research on anxiety disorders. Countries/regions with more than 5000 publications were in red, countries/regions with 1000–4999 were in orange, countries/regions with 100–999 were in dark green, countries/regions with 10–99 were in yellow, and countries/regions with 1–9 were in light green. It can be seen that almost all of the world had contributed to anxiety disorders research, with North America, East Asia, and Oceania leading the way.

3.3. Analysis of Highly Productive Journals
Four thousand fifty-four journals had published publications on anxiety disorder research. Table 2 showed the top 20 journals in publication volume. The Journal of Affective Disorders (n = 1861 counts) was the most prolific magazine in this field, followed by the Journal of Anxiety Disorders (n = 1389 counts) and the Behaviour Research and Therapy (n = 1211 counts). The Behaviour Research and Therapy (n = 80,094 citations) was the journal with the highest total citations, followed by the Journal of Affective Disorders (n = 58,077 citations) and the Journal of Anxiety Disorders (n = 58,017 citations).
Rank | Journal | Counts | Citations | Avg. citations | Avg. year pub |
---|---|---|---|---|---|
1 | Journal of Affective Disorders | 1861 | 58,077 | 31.21 | 2016.20 |
2 | Journal of Anxiety Disorders | 1389 | 58,017 | 41.77 | 2012.16 |
3 | Behaviour Research and Therapy | 1211 | 80,094 | 66.14 | 2010.88 |
4 | Psychiatry Research | 1,143 | 41,884 | 36.64 | 2014.25 |
5 | PLOS One | 1040 | 34,028 | 32.72 | 2017.32 |
6 | Depression and Anxiety | 914 | 47,427 | 51.89 | 2011.28 |
7 | Frontiers in Psychiatry | 912 | 9490 | 10.41 | 2021.21 |
8 | BMC Psychiatry | 762 | 17,099 | 22.44 | 2018.25 |
9 | Journal of Psychiatric Research | 735 | 27,866 | 37.91 | 2015.57 |
10 | Psychological Medicine | 688 | 44,078 | 64.07 | 2013.84 |
11 | Frontiers in Psychology | 596 | 9119 | 15.30 | 2019.91 |
12 | Journal of Obsessive-Compulsive and Related Disorders | 556 | 7158 | 12.87 | 2017.60 |
13 | Journal of Behavior Therapy and Experimental Psychiatry | 546 | 16,007 | 29.32 | 2014.12 |
14 | Comprehensive Psychiatry | 540 | 19,453 | 36.02 | 2012.11 |
15 | Cognitive Therapy and Research | 525 | 15,217 | 28.98 | 2014.07 |
16 | Journal of Clinical Psychiatry | 518 | 33,568 | 64.80 | 2009.86 |
17 | BMJ Open | 490 | 5423 | 11.07 | 2020.57 |
18 | Biological Psychiatry | 480 | 53,871 | 112.23 | 2009.61 |
19 | Behavior Therapy | 475 | 19,666 | 41.40 | 2014.19 |
20 | International Journal of Environmental Research and Public Health | 408 | 8180 | 20.05 | 2020.82 |
Of the top 20 journals, the one with the highest average citations was Biological Psychiatry (n = 112.23 citations). The earliest journal in average publication year was Biological Psychiatry (2009.61), and the latest journal in average publication year was Frontiers in Psychiatry (2021.21).
We set the threshold for journal publication to 100, and 132 high-yield journals were obtained. There were 38,478 publications in 132 high-yield journals, accounting for 57.10% of all publications. We used VOSviewer to perform citation analysis on 132 high-yield journals and constructed overlay visualization map (Figure 4). The size of circle represented the amount of publications, and the color from blue to red represented the average number of citations from low to high. There were 19 journals with on average more than 60 citations (shown in red).

3.4. Analysis of High-Yield Institutions
There were 31,577 institutions involved in the publication of anxiety disorders articles, and the top 20 institutions ranked by the number of publications were shown in Table 3. Harvard University was far ahead of other institutions in both the number of publications (2389 counts) and the number of citations (176,341 citations), Columbia University was second in the number of publications (1544 counts), and the third was King’s College London (1328 counts). The second in citations was Columbia University (123,595 citations), and third was Yale University (71,904 citations).
Rank | Institution | Country | Counts | Citations | Avg. citations | Avg. year pub |
---|---|---|---|---|---|---|
1 | Harvard University | United States | 2389 | 176,341 | 73.81 | 2014.34 |
2 | Columbia University | United Kingdom | 1544 | 123,595 | 80.05 | 2013.31 |
3 | King’s College London | United Kingdom | 1328 | 65,552 | 49.36 | 2015.74 |
4 | Yale University | United States | 1226 | 71,904 | 58.65 | 2014.34 |
5 | University of California, Los Angeles | United States | 1201 | 71,102 | 59.2 | 2013.6 |
6 | University of Toronto | Canada | 1078 | 43,703 | 40.54 | 2015.59 |
7 | Karolinska Institute | Sweden | 1074 | 43,972 | 40.94 | 2017.27 |
8 | Washington University | United States | 1046 | 57,854 | 55.31 | 2014.06 |
9 | Massachusetts General Hospital | United States | 1026 | 68,007 | 66.28 | 2013.25 |
10 | Boston University | United States | 973 | 63,186 | 64.94 | 2013.5 |
11 | Vrije University Amsterdam | Netherlands | 926 | 47,154 | 50.92 | 2014.57 |
12 | University of Pennsylvania | United States | 900 | 51,844 | 57.6 | 2013.75 |
13 | University of California, San Diego | United States | 879 | 56,355 | 64.11 | 2012.44 |
14 | University of Pittsburgh | United States | 854 | 52,620 | 61.62 | 2013.16 |
15 | Stanford University | United States | 837 | 48,729 | 58.22 | 2014.78 |
16 | The University of Melbourne | Australia | 818 | 35,991 | 44 | 2016.13 |
17 | University of Michigan | United States | 798 | 45,154 | 56.58 | 2014.77 |
18 | University of Amsterdam | Netherlands | 790 | 35,791 | 45.31 | 2015.12 |
19 | Johns Hopkins University | United States | 762 | 36,658 | 48.12 | 2014.33 |
20 | University of Groningen | Netherlands | 748 | 40,282 | 53.85 | 2014.46 |
Among the top 20 institutions, Columbia University had the highest average citation (80.05 citations), followed by Harvard University (73.81 citations) and Massachusetts General Hospital (66.28 citations). Among the top 20 institutions, University of California, San Diego, had the earliest average publication year (2012.44), and Karolinska Institute had the latest average publication year (2017.27).
Of the top 20 institutions, 12 institutions were from the United States, 3 from the Netherlands, 2 from the United Kingdom, and 1 each from Canada, Sweden, and Australia.
We set the threshold 150 publications for institutions and identified 206 high-yield institutions from 31,577 institutions. We used VOSviewer to conduct coauthorship analysis on 206 high-yield institutions, all of which were in a coauthorship network, as shown in Figure 5, consisting of 8 clusters.

The red cluster was the largest, consisting of 73 institutions. Harvard University was the largest node, in close cooperation with the Massachusetts General Hospital and the Boston University, all of which were from American. The second was the green cluster, consisting of 51 institutions, and King’s College London was the largest node, with close cooperation with University College London and University of Oxford, all of which were British. The dark blue cluster was the third largest, and consists of 24 institutions, and the Dresden University of Technology was the largest node, with close cooperation with the University of Wurzburg and University of Basel, which were all German.
Harvard University cooperated with 199 high-yield institutions, followed by Columbia University with 183 high-yield institutions, and University of Toronto and Yale University with 182 high-yield institutions.
3.5. Visualization of Coauthorship Analysis of Authors
There were 206,496 authors which participated in anxiety disorder publications, and the top 20 authors by number of publications were shown in Table 4. Storch, Eric A. (n = 411 counts) were the most prolific authors, followed by Dan J. Stein (n = 377 counts) and Brenda WJH Penninx (n = 311 counts). Kessler, Ronald C. (n = 67,145 citations) were the most cited authors, followed by Dan J. Stein (n = 27,348 citations) and Stein, Murray B (n = 26,248 citations). Of the top 20 authors, Kessler, Ronald C. (n = 271.84 citations) had the most average citation times, followed by Wittchen, Hans-Ulrich (n = 104.94 citations) and Stein, Murray B. (n = 101.74 citations). Stein, Murray B. had the earliest average publication year (2009.89), and Storch, Eric A. (2015.87) had the latest average publication year.
Rank | Author | Counts | Citations | Avg. citations | Avg. pub. year |
---|---|---|---|---|---|
1 | Storch, Eric A. | 411 | 13,303 | 32.37 | 2015.87 |
2 | Dan J. Stein | 377 | 27,348 | 72.54 | 2013.00 |
3 | Brenda WJH Penninx | 311 | 18,044 | 58.02 | 2015.61 |
4 | Daniel Pine | 288 | 19,520 | 67.78 | 2013.16 |
5 | Michelle G. Craske | 283 | 16,489 | 58.27 | 2013.02 |
6 | Zvolensky, Michael J. | 268 | 11,529 | 43.02 | 2012.15 |
7 | Stein, Murray B. | 258 | 26,248 | 101.74 | 2009.89 |
8 | David Mataix-Cols | 250 | 12,268 | 49.27 | 2015.16 |
9 | Schmidt, Norman B. | 248 | 9674 | 39.01 | 2012.42 |
10 | Kessler, Ronald C. | 247 | 67,145 | 271.84 | 2013.03 |
11 | Wittchen, Hans-Ulrich | 236 | 24,765 | 104.94 | 2012.44 |
12 | Rapee, Ronald M. | 232 | 12,317 | 53.09 | 2014.29 |
13 | Kendall, Philip C. | 227 | 11,929 | 52.55 | 2014.08 |
14 | Andersson, Gerhard | 205 | 13,213 | 64.45 | 2015.12 |
15 | Piacentini, John | 188 | 11,004 | 58.53 | 2014.30 |
16 | Hofmann, Stefan G. | 178 | 14,504 | 81.48 | 2012.83 |
17 | Abramowitz, Jonathan S. | 177 | 12,254 | 69.23 | 2012.36 |
17 | Heimberg, Richard G. | 177 | 12,122 | 68.49 | 2011.78 |
19 | Tolin, David F. | 173 | 10,088 | 58.31 | 2013.66 |
20 | Fontenelle, Leonardo F. | 166 | 3766 | 22.69 | 2015.32 |
We set the threshold of author publications to 60 and identified 164 high-yield authors. We used VOSviewer to perform coauthorship analysis on 164 high-yield authors (Figure 6), and 160 of the 164 high-yield authors formed the largest coauthor network, consisting of 12 clusters. Dan J. Stein collaborated with 77 high-yield authors, followed by Christine Lochner with 62 high-yield authors, and Eric A. Storch with 57 high-yield authors.

We counted the research areas of eight authors with more than 250 publications, as shown in Figure 7. We made specific statistics on the top five research areas with the most published articles by the eight authors, and the rest were classified into others. Psychiatry was the largest research area for seven of the eight authors; other significant research areas included clinical psychology, psychology, clinical neurology, neurosciences, etc.

3.6. Analysis of Highly Cited Articles
Table 5 shows the top 20 highly cited articles. The highly cited one of the 67,386 articles was Robert L. Spitzer’s article, named “A brief measure for assessing generalized anxiety disorder: the GAD-7” [16]. The study developed a short self-report scale based on questionnaires and telephone interviews with mental health professionals from 15 primary care clinics in the United States. Comparing the GAD self-report scale diagnosis with an independent diagnosis by a mental health professional confirmed that GAD-7 was an effective tool for screening for generalized anxiety disorders and assessing disease severity in clinical practice and research. Ronald C. Kessler’s article named “Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication” was close behind in citations. The study was a nationally face-to-face household survey of 9282 English-speaking respondents aged 18 years and older, using a comprehensive international diagnostic interview from the fully structured version of the World Health Organization’s World Mental Health Survey [5]. The result was that about half of Americans would meet the criteria for DSM-IV disorder at some point in their lives, usually with first onset in childhood or adolescence, so interventions need to focus on youth. The citations of these two papers were far ahead of the others.
Rank | Title | Publication year | Journal | First author | Citations |
---|---|---|---|---|---|
1 | A brief measure for assessing generalized anxiety disorder: the GAD-7 | 2006 | Archives of Internal Medicine | Spitzer | 17,375 |
2 | Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication | 2005 | Archives of General Psychiatry | Kessler | 16,908 |
3 | Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990−2010: a systematic analysis for the Global Burden of Disease Study 2010 | 2012 | The Lancet | Vos, Theo | 4469 |
4 | Lifetime prevalence of mental disorders in U.S. adolescents: results from the National Comorbidity Survey Replication–Adolescent Supplement (NCS-A) | 2010 | Journal of the American Academy of Child and Adolescent Psychiatry | Merikangas | 3955 |
5 | Global burden of disease attributable to mental and substance use disorders: findings from the Global Burden of Disease Study 2010 | 2013 | The Lancet | Whiteford | 3931 |
6 | The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication | 2007 | Biological Psychiatry | Hudson | 3203 |
7 | The psychological impact of the COVID-19 epidemic on college students in China | 2020 | Psychiatry Research | Cao | 2982 |
8 | Anxiety disorders in primary care: prevalence, impairment, comorbidity, and detection | 2007 | Annals of Internal Medicine | Kroenke | 2868 |
9 | Prevalence and development of psychiatric disorders in childhood and adolescence | 2003 | Archives of General Psychiatry | Costello | 2718 |
10 | Validation and standardization of the Generalized Anxiety Disorder Screener (GAD-7) in the general population | 2008 | Medical Care | Lowe | 2709 |
11 | The Patient Health Questionnaire Somatic, Anxiety, and Depressive Symptom Scales: a systematic review | 2010 | General Hospital Psychiatry | Kroenke | 2646 |
12 | The prevalence and correlates of adult ADHD in the United States: results from the National Comorbidity Survey Replication | 2006 | American Journal of Psychiatry | Kessler | 2631 |
13 | The size and burden of mental disorders and other disorders of the brain in Europe 2010 | 2011 | European Neuropsychopharmacology | Wittchen | 2510 |
14 | The effect of mindfulness-based therapy on anxiety and depression: a meta-analytic review | 2010 | Journal of Consulting and Clinical Psychology | Hofmann | 2269 |
15 | Psychiatric disorders in children with autism spectrum disorders: prevalence, comorbidity, and associated factors in a population-derived sample | 2008 | Journal of the American Academy of Child and Adolescent Psychiatry | Simonoff | 2225 |
16 | Generalized anxiety disorder, depressive symptoms and sleep quality during COVID-19 outbreak in China: a web-based cross-sectional survey | 2020 | Psychiatry Research | Huang | 2196 |
17 | Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic | 2021 | The Lancet | Santomauro | 2176 |
18 | Prevalence and co-occurrence of substance use disorders and independent mood and anxiety disorders: results from the National Epidemiologic Survey on Alcohol and Related Conditions | 2004 | Archives of General Psychiatry | Grant | 2168 |
19 | The obsessive–compulsive inventory: Development and validation of a short version | 2002 | Psychological Assessment | Foa | 2053 |
20 | The role of childhood trauma in the neurobiology of mood and anxiety disorders: preclinical and clinical studies | 2001 | Biological Psychiatry | Heim | 2001 |
The psychological impact of the COVID-19 epidemic on college students in China ([7], IF 11.3); generalized anxiety disorder, depressive symptoms, and sleep quality during COVID-19 outbreak in China [7]; and a web-based cross-sectional survey ([17], IF 11.3) [17], global prevalence, and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic [18] were the representative papers published in the past 4 years. However, the number of citations ranked in top 20. On the one hand, these papers were of high quality, and on the other hand, anxiety and depression caused by novel coronavirus were also hot topics in recent years, with great potential.
3.7. Keyword Analysis
There were 53,765 author keywords in 67,386 articles. We set the threshold for high-frequency keywords to 80 and obtained 394 high-frequency author keywords. The co-occurrence analysis of 394 high-frequency author keywords was performed, and the co-occurrence network map of high-frequency author keywords was constructed. The co-occurrence analysis of 394 high-frequency keywords formed 5 clusters, represented by different colors. The red cluster was composed of 116 author keywords and was the largest cluster. Cluster 1 (red) represented etiology and pathogenesis, Cluster 2 (green) represented clinical classification and interventions, Cluster 3 (blue) represented comorbidity, Cluster 4 (yellow) represented anxiety caused by the COVID-19 pandemic and specific groups, and Cluster 5 (purple) represented anxiety disorders in children and adolescents, focusing on five research topics (Figure 8).

We also built overlay map for 394 high-frequency author keywords (Figure 9). The color of each node in the figure represents the average publication year of that node, becoming increasingly new from blue to red.

4. Discussion
4.1. General Information
In this study, we analyzed 67,386 articles on anxiety disorders obtained from the WoSCC, providing an in-depth analysis of the literature by country, institution, journal, author, and keyword. Judging from the number of annual publications, the extensive and close scientific collaboration, the dense citation network, and the numerous subject areas involved, there is no doubt that anxiety is still at a hot phase. The life, career, and financial changes caused by the COVID-19 pandemic have been associated with worsening mental health, which may be responsible for the sharp increase in the number of publications on this topic in 2020−2022 [7, 17–20].
There were 164 countries/regions; almost all regions of the world participated in anxiety disorder research, especially North America, Europe, East Asia, and Oceania. The leading position of the United States in the field of anxiety disorders was solid. Of the 67,386 articles we included, the United States topped the list with 26,919 in absolute terms, significantly outpacing other countries. The United States had six of the top 10 research institutions by output and four of the top 10 journals published. All these indicated that the robust national research capabilities of the United States and its substantial contributions to this field. Although Belgium ranked 11th in the volume of research outputs, it was the highest in terms of total citations, followed by France and the United States in third, which indicated that these countries had a certain influence in the study of anxiety disorders. Also of note was China’s rapid growth, with the latest average publication year (2019) indicating great potential for future research.
As for contributing institutions, Harvard University in the United States was the core institution of anxiety disorder research. Harvard University ranked first in the world in both the number of research outputs and the cited publications, which illustrated that Harvard University was the most influential institution and the quality of the published literature was widely recognized by peers. However, none of the Asian institutions made it to the list of top 20 institutions in terms of scientific output or most cited publications on anxiety disorder research. In terms of international research cooperation in this field, the cooperation between countries/regions was very extensive; the United States, the UK, and Germany were more influential. Notably, Harvard University also was the center of global research collaboration in this field, with 199 high-output research institutions, the closest being Boston University and Massachusetts General Hospital in the United States. In addition, renowned research institutions such as Columbia University, University of Toronto, and Yale University had established substantial collaborative networks, engaging with over more than 180 institutions each. But our analysis revealed that there was a high degree of cooperation among the developed European and American countries, with much less collaborative activity with institutions in developing countries. Moreover, the pattern of cooperation between these institutions tends to be relatively fixed. It is necessary to strengthen cross-regional collaboration among global institutions to ensure a broader and more diverse contributions, which is essential in future research in key areas of anxiety disorders.
The author’s analysis can objectively evaluate the contribution of the researchers, as well as their research level and academic standing on the topic. Professor Eric A. Storch from Baylor College of Medicine in the United States produced the most research documents in this field with 411 publications, plus with the latest average publication year, indicating his continuing influence in the field. Following Eric A. Storch is Dan J. Stein from the University of Cape Town in South Africa and Brenda WJH Penninx from Vrije Universiteit, Amsterdam, in Netherlands. We found that most of the authors came from universities and hospital institutions, and the most active scholars with the most research achievements contributed greatly to clinical psychology, psychology, clinical neurology, and neuroscience, especially as they formed a close scientific cooperation relationship. The three groups with the highest link strength were centered on Dan J. Stein from the University of Cape Town in South Africa, Christine Lochner from Stellenbosch University in South Africa, and Eric A. Storch from Baylor College of Medicine in the United States.
Remarkably Professor Ronald C. Kessler from Harvard Medical School in the United States was the most cited author in both total and average citations times, indicating that he had a very high level of influence in the field and his articles were of very high quality and worth studying.
4.2. The Hotspots and Frontiers
Highly cited references form the source of a knowledge base for a particular subject area. The top 10 cocited references were mainly related to the following aspects attracting researchers: (1) development and validation of the screening assessment tools for anxiety disorders [16, 21, 22]; (2) epidemiological evidence [5, 23]; (3) psychiatric disorders in childhood and adolescence [3, 24]; (4) comorbidity of anxiety disorders [25, 26]; and (5) the psychological impact of the COVID-19 epidemic [7].
Keyword analysis can help us understand the hotspots and frontiers in the field. In the high-frequency keyword analysis, the main focus was on five clusters, which shaped the current research trends in anxiety disorders.
4.2.1. Cluster 1 (Red Cluster): Etiology and Pathogenesis
Due to its multifactorial nature and heterogeneous symptomatology, the precise etiology of anxiety disorders is largely unclear. The possible pathogenesis of anxiety disorders includes genetics [27, 28], neurotransmitter dysfunction[29–31], neuroendocrine dysfunction [32], and immune-inflammatory activation [33], molecular [34], neural circuit [34], social factors [35], and so on. However, the status of each pathogenic factor in the pathogenesis needs to be further elucidated.
Previous studies mainly focused on animal models or psychosocial factors, but in recent years, the rapid development of neuroimaging and the use of MRI to analyze brain structure, brain network mechanism, and brain function have provided support for neuropathological studies of anxiety disorders. In this domain, keywords such as “amygdala,” “prefrontal cortex,” “hippocampus,” “nucleus accumbens,” “insula,” and “cingulate gyrus” were often mentioned, which together described the main brain regions involved in the development of anxiety disorders [36]. “Genetics” has attracted the attention of researchers, mainly focusing on candidate genes related to monoaminergic neurotransmission or stress axis function [37]. Basic neuroscience research has been working to the diagnosis, treatment, and prediction of anxiety disorders.
4.2.2. Cluster 2 (Yellow Cluster): Clinical Classification and Interventions
The DSM-5 classifies anxiety disorders into generalized anxiety disorder, social anxiety disorder, panic disorder, specific phobias, separation anxiety disorder, selective mutism, agoraphobia, etc. based on the molecular genetics and neuroimaging findings of fear circuit disorders, as well as the presentation of clinical features, to further improve diagnostic validity and facilitate clinical application [38]. However, this classification has been controversial eversince its promulgation.
Anxiety disorders are treated with varying degrees of consistency. Mentally, anxiety can be addressed through psychological treatments such as cognitive behavioral therapy (CBT), implosive therapy, mindfulness, and virtual reality. Physically, anxiety can be addressed with medications, including selective serotonin reuptake inhibitors (SSRIs)/selective serotonin–norepinephrine reuptake inhibitors (SNRIs), benzodiazepines, and 5-HT1A agonist [38].
Despite this, over the past 5–10 years, there has been far less research on novel medication treatments or new therapies for anxiety disorders. It is in the context of the lack of head-to-head comparison that “network analysis” has become a co-occurrence of emerging high-frequency keywords in 2021. Network analysis can draw on data comparing one treatment to a placebo or another, or compare and rank all treatments, even if they are not directly compared [39, 40].
4.2.3. Cluster 3 (Green Cluster): Comorbidity
Individuals diagnosed with anxiety disorders have an increased risk of developing almost all other types of mental disorders, especially depression, bipolar disorder, autism spectrum disorder, and feeding and eating disorder, and the relationship is generally bidirectional as well as they are also more likely to have alcohol and substance-related disorders [41].
The close relationship between anxiety and depression has been clinically observed for a long time [42]. Comorbid anxiety and depressive disorder, also known as anxious depression. Thirty percent to 63% of people with anxiety disorders meet the criteria for major depressive disorder (MDD) [43], similarly about 85% of patients with depression have significant anxiety [44]. Comorbidity of depressive and anxiety disorders was associated with more childhood trauma [43]. Comorbidity of depressive and anxiety disorders has earlier age of onset [43, 45, 46], more severe and long-lasting conditions [43, 47], more frequent episodes of MDD, and a higher risk of suicidal ideation and suicide attempts [48, 49]. Molecular genetic study explained high comorbidity among most psychiatric disorders, such as anxiety disorders, depression, and neuroticism [50].
Anxiety disorders can occur with the development or worsening of medical conditions, including neurological disorders [51], cardiovascular [52] and respiratory illnesses [53], gastrointestinal diseases [54], cancer [55], chronic pain [56], endocrine system diseases [57, 58], and chronic kidney disease [59]. Up to 71% to 97.8% of people with anxiety disorders are not properly diagnosed [60], and about 41% are not treated [21]. Kinesiophobia is observed in a wide range of chronic diseases, such as chronic pain, neurological disorders, and cardiovascular and respiratory illnesses [61]. The usage of the term “kinesiophobia” has increased dramatically after 2019 and is likely to continue to be a relevant topic.
At present, the identification of anxiety disorders is mainly based on clinical symptoms and anxiety course. The clinical features of anxiety disorders include psychiatric symptoms and physical symptoms. Psychiatric disorders are intertwined. Psychiatric symptoms of anxiety are common in all psychiatric disorders, and the clinical allocation of “typical” and “atypical” somatic symptoms for medical or psychiatric diagnosis is notoriously difficult. In recent years, network analysis has also been used to to explore complex relationships between symptoms [62].
4.2.4. Cluster 4 (Blue Cluster): Anxiety Caused by the COVID-19 Pandemic and Specific Groups
The COVID-19 pandemic has further exacerbated the global burden of mental illness. In 2020, the prevalence of anxiety disorders increased by about 76 million, an increase of about 25.6% [18]. Within 6 months of follow-up, an estimated 17.39% of COVID-19 patients had anxiety disorders, and 19.15% of critically ill patients admitted to the ICU [63]. One year after follow-up, 17.9% of COVID-19 survivors admitted to ICU reported symptoms of anxiety [64]. The long-term consequences of coronavirus infection and mechanisms by which pandemic stressors cause negative mental health are still unknown. All these are hot spots of the future research. The COVID-19 pandemic has accelerated the use of technology-based interventions in mental health treatment. We found that the hot keywords that emerged sharply after 2019 are “internet-based intervention” and “machine learning,” underscoring the recent attention paid to these topics.
Referring to the field of special groups, it is worth noting that “pregnancy,” “postpartum period,” “older adults,” “young adult,” “loneliness,” “HIV,” “caregivers,” “stroke,” “sleep disorders,” “asthma,” “cognitive dysfunction,” “multiple sclerosis,” and “Alzheimer disease” are the most frequently mentioned terms, indicating these are the groups that need to focus on, and society needs to give more support.
4.2.5. Cluster 5 (Purple Cluster): Anxiety Disorders in Children and Adolescents
There is no doubt that researchers have been concerned about the psychological problems of adolescents. Three of the top 10 cited articles focused on psychiatric disorders in childhood and adolescence [3, 5, 24]. For example, in 2005, Kessler published a study on the psychiatric epidemiological survey, which has reached 16908 citations, ranking second in total citations, revealing that anxiety disorders are the most common mental disorders among American adults, with a lifetime prevalence of 28.8%, and first onset usually in childhood or adolescence, as 50% of them first experienced symptoms by age 14 and 75% by age 18 [5]. Anxiety disorders are most common in adolescents (31.9%) [3], and usually begin during childhood, appearing as early as age 6 [3], and are associated with a lifetime of chronic and recurrent course, as well as the onset of other psychiatric disorders, with a report showing a significant proportion (25.5%) of affected adolescents meeting the criteria for two or more psychiatric disorders [3], ultimately resulting in widespread and lasting impairment in interpersonal, social, educational, and occupational functioning.
The prevalence of anxiety disorders in children and adolescents has risen sharply due to various factors such as family pressure, social pressure (e.g., bullying, discordant relationships, or stressful life events), excessive academic burden, and insufficient sleep [43, 65], which has become an ongoing global concern. In this area, keywords such as “assessment,” “reliability,” and “validity” are also the most common co-occurrence keywords. There is still a lot of hard work to be done to carry out psychological assessment for children and adolescents, but it is a very valuable work. After 2019, the sharp increase in usage of keywords such as “students,” “young adult,” “college students,” and “university students” has increased dramatically, which indicates that the study of anxiety disorders in children and adolescents has become an important research hotspot.
The above five aspects have been the hot spots in the field of anxiety disorders in the past 20 years, but there are still many problems to be further studied and explored, and there is still huge research potential in this field.
5. Limitations
Our bibliometric study has some limitations. First, only English literature was included in this study; language bias may inevitably occur. Second, high-quality articles published recently cannot attract enough attention due to their short publication time and low citation frequency. Finally, there may be a small amount of bias in the results for institutions and authors due to differences in the expression of author and institution names.
6. Conclusion
In conclusion, this study will fill the gaps in the existing literature. Through a comprehensive analysis of the current research results and trends in the field of anxiety disorders, we predict that “molecular and neural circuit mechanisms,” “adolescent psychological problems,” “objective indicators of diagnosis and classification,” and “technology-assisted therapy” will continue to be hot topics of anxiety disorders in the future.
Ethics Statement
No ethical approval and patient consent were required for all analyses.
Consent
The authors have nothing to report.
Conflicts of Interest
The authors declare no conflicts of interest.
Author Contributions
Research design: Fang Wang, Zhongqing Wang, Bailing Tian, and Nannan Li. Data collection: Zhongqing Wang, Zhiying Cui, Fanyu Meng, and Haoxin Guo. Data analysis: Zhongqing Wang, Fang Wang, Bailing Tian, and Nannan Li. Original writing: Nannan Li, Bailing Tian, and Zhongqing Wang. Manuscript editing: Nannan Li, Bailing Tian, and Zhongqing Wang. Zhongqing Wang, Fang Wang, and Zhiying Cui contributed equally to this work and should be regarded as co-first authors. Bailing Tian and Nannan Li are corresponding authors.
Funding
This study was funded by the foundation from the Chinese National Key Technology R&D Program (2021YFC2501303 to Pingting Yang), the Doctoral Fund of Liaoning Province (2022-BS-141 to Bailing Tian), and the Natural Science Foundation of Liaoning Province (2023-MS-07 to Liu Hu).
Supporting Information
Additional supporting information can be found online in the Supporting Information section.
Open Research
Data Availability Statement
The data that support the findings of this study are available from the corresponding authors upon reasonable request.