Mapping the Knowledge Landscape of Atopic Dermatitis and Allergic Rhinitis Comorbidity: A Bibliometrics Study

Summary

Atopic dermatitis and allergic rhinitis are common comorbid conditions, yet research on their co-occurrence remains to be fully elucidated. This study aims to employ a bibliometric approach to map the knowledge structure, identify research hotspots, and trace the evolving trends in this field.

Abstract

Atopic dermatitis (AD) and allergic rhinitis (AR) are common allergic diseases that frequently co-occur, sharing common underlying pathophysiological mechanisms. This study retrieved comorbidity-related publications on AD and AR from the Web of Science (WOS) database over the past decade. Bibliometric analysis was performed using software tools including Microsoft Excel, VOSviewer, CiteSpace, and Scimago Graphica to examine countries, research institutions, journals, authors, and keywords. The results indicate that a total of 1571 relevant publications were published in the past 10 years, showing a fluctuating upward trend. The United States contributed the most publications (244 articles), cluster analysis identified 3 major research hubs, represented by the United States, China, and Germany. In terms of research institutions, China Medical University was the most productive (42 publications). Cluster analysis revealed 4 primary institutional research centers, represented by the University of Zurich, China Medical University, Seoul National University, and the University of Colorado. Among journals, Allergy published the most articles (56), while the Journal of Allergy and Clinical Immunology demonstrated the strongest academic influence, with an average of 78.77 citations per article. Regarding authors, Akdis, Cezmi A., had the highest output (15 publications), while Irvine, Alan D., exhibited the strongest impact, with a total citation count of 2711. Keyword analysis identified 11 clusters, which were primarily associated with immune regulation. This protocol provides researchers with an effective strategy to identify potential research gaps and guide future studies in these interrelated allergic diseases.

Introduction

Atopic Dermatitis (AD) and Allergic Rhinitis (AR) are two common chronic inflammatory allergic diseases, which together form the core components of the atopic march^1,2,3. Atopic Dermatitis is a chronic inflammatory skin disease associated with a genetic predisposition to allergy, characterized by impaired skin barrier function, immune dysregulation, and intense pruritus^4,5. Allergic rhinitis is an IgE-mediated chronic inflammation of the nasal mucosa in response to environmental allergens, typically presenting with nasal congestion, rhinorrhea, nasal itching, and sneezing⁶. In recent years, extensive epidemiological and clinical studies have confirmed a significant comorbidity between Atopic Dermatitis and Allergic rhinitis^7,8. Patients often develop these two conditions simultaneously or sequentially, which not only increases the disease burden but also leads to a significant decline in quality of life and higher healthcare costs. Concurrently, the association between the two diseases provides important clues for research into their pathogenesis. It is widely accepted in the academic community that these two diseases share complex genetic backgrounds, immunological mechanisms, and environmental factors⁹. However, although substantial progress has been made in basic and clinical research on either Atopic Dermatitis or Allergic Rhinitis, systematic and macro-level studies on their comorbidity remain relatively limited.

Bibliometrics, as a scientific discipline that employs mathematical and statistical methods to quantitatively analyze academic literature, demonstrates unique advantages¹⁰. It transcends the subjective limitations inherent in traditional reviews by analyzing data such as publication counts, authors, institutions, national/regional collaborations, keyword co-occurrence, and document co-citations¹¹. This allows for an objective and systematic revelation of the research output, knowledge structure, evolutionary dynamics, and future directions within a specific field¹². Applying bibliometrics to the research domain of Atopic Dermatitis (AD) and Allergic Rhinitis (AR) comorbidity can help researchers quickly grasp the overall landscape of this field. This study integrates bibliometric methods to systematically review the literature related to AD and AR over the past decade. This study contributes a reproducible bibliometric protocol that integrates multiple software tools to systematically process and analyze literature data. Through this data-driven approach, we comprehensively present the knowledge structure of the AD and AR comorbidity field, including research hotspots, collaboration networks, and thematic evolution over the past decade. This methodological framework moves beyond traditional narrative reviews by providing an objective and quantitative synthesis of the literature, offering researchers a valuable tool for understanding the current state and future directions of this field.

Protocol

1. Literature search and acquisition

1. Access the Medical Subject Headings (MeSH) database (https://meshb.nlm.nih.gov/).
2. Enter the terms "Atopic Dermatitis" and "Allergic Rhinitis" separately to retrieve their respective controlled vocabulary terms and subheadings.
3. Based on the identified MeSH terms and subheadings, formulate a comprehensive search strategy.
4. Navigate to the Web of Science (WoS) database (https://www.webofscience.com/wos/). Select the "Advanced Search" option.
5. Input the following search strategy into the query box: TS=("Atopic Dermatitis" OR "Atopic Eczema" OR "Infantile Eczema" OR "Atopic Neurodermatitis" OR "Disseminated Neurodermatitis") AND TS=("Allergic Rhinitis"). Limit the publication date range from January 1, 2015, to December 31, 2024.
6. Exclude non-research document types such as meeting abstracts, conference papers, editorial materials, book chapters, and retracted publications; retain only English-language literature.
7. For publications with inconsistent author names or institutional affiliations, use the affiliation of the first author as the standard to ensure consistency in network analyses.
8. Export the obtained dataset in plain text format and store it on the computer for subsequent analysis.
9. Record the annual number of published articles and click on 'Citation Report' to export the citation counts.
10. Click "See all" in the "Countries/Regions" section and export the national publication output to Excel.
11. Under the "Publication Years" column, click "See all", then sequentially select "2015" and "Refine". Next, click "See all" in the "Countries/Regions" column.
12. Record the specific publication counts for the top 10 countries in the year 2015.
13. Following Step 10, record the annual publication counts for the top 10 countries from 2015 to 2024.

2. Trend analysis of annual publication output and citation frequency

1. Open Microsoft Excel and sequentially enter the year, the corresponding number of published articles, and the citation count.
2. Select the input data, then navigate to Insert > Chart > Combo Chart. Set the annual number of published articles as a Column Chart, and the citation count as a Line Chart.
3. Select the chart and label the X-axis title as "Year" and the Y-axis title as "Publications".
4. Select the data labels and set them to display the specific values.
5. Select the figure and save it in TIFF format.

3. National publication output and cross-border research networks

NOTE: VOSviewer was used to construct the national collaboration network due to its strength in bibliometric network visualization. Scimago Graphica was subsequently employed to generate world maps for clearer geographic presentation of the network data.

1. Launch the software, then navigate through and click: Create → Create a map based on bibliographic data → Read data from bibliographic database files.
2. Import the plain text dataset, then click 'Next' to read the data.
3. In the "Type of analysis" field, select "Co-authorship"; then choose "Countries" under "Unit of analysis" and click "Next". Set the value in "Number of organizations to be selected" to 22, proceed by clicking "Next", and initiate the data analysis.
4. Right-click on the data in the "Source" column and export metrics, including national publication output and citation counts for subsequent data analysis.
5. Click "Finish" to generate the national collaboration network, then select the "Items" tab in the left toolbar to view the number of clusters.
6. Navigate to "Analysis" in the left toolbar, select "Minimum cluster size", and adjust this parameter to maintain the number of clusters within an appropriate range.
7. Access the "Analysis" and then "Layout" sections within the left toolbar. Adjust the numerical values for the "Attraction" and "Repulsion" parameters to optimize the scale of the visualization.
8. Select "Visualization" on the right to modify the appearance of the visual map. Drag the "Scale" slider to adjust country/region labels to an appropriate size, use the "Labels" control to set node dimensions, manipulate the "Lines" parameter to achieve suitable connection line thickness, and click "Curved lines" to select straight or curved lines for the connections.
9. Navigate to File > Screenshot > Save to export the visualization; then select File > Save > GML to preserve the GML format file for subsequent use.
10. Launch the software and import the GML format file.
11. Set the "lable" column as "Country", change the data type of the "cluster" column to "String", and click the visualization icon in the upper-left corner.
12. Drag "weight<documents>" from the left panel to the "Size" box, move "cluster" to the "Color" box, and place "lable" into the "Label", "Tooltip", and "Unit" boxes respectively.
13. Select "Map" under the "Marks" section in the right toolbar to generate the world map; then use the "Edges" option to adjust connection line curvature and modify "Edge color" to configure node colors.
14. Select "Color" in the left toolbar to adjust cluster colors; then configure "Legend position" to modify the placement of the legend.
15. Click "Export" in the top toolbar to save the image in PNG format.

4. Analysis of institutional publication output and collaboration networks

NOTE: VOSviewer was chosen for institutional collaboration analysis because it effectively handles large-scale bibliometric data and provides a clear visualization of institutional clusters.

1. Launch the software, then navigate through and click: Create → Create a map based on bibliographic data → Read data from bibliographic database files.
2. Import the plain text dataset, then click 'Next' to read the data.
3. In the "Type of analysis" field, select "Co-authorship"; then choose "Organizations" under "Unit of analysis" and click "Next". Set the value in "Number of organizations to be selected" to 50, proceed by clicking "Next", and initiate the data analysis.
4. Right-click on the data in the "Source" column and export metrics, including institutional publication output and citation counts for subsequent data analysis.
5. Click "Finish" to generate the institutional collaboration network, then select the "Items" tab in the left toolbar to view the number of clusters.
6. Navigate to "Analysis" in the left toolbar, select "Minimum cluster size", and adjust this parameter to maintain the number of clusters within an appropriate range.
7. Access the "Analysis" and then "Layout" sections within the left toolbar. Adjust the numerical values for the "Attraction" and "Repulsion" parameters to optimize the scale of the visualization.
8. Select "Visualization" on the right to modify the appearance of the visual map. Drag the "Scale" slider to adjust institution names to an appropriate size, use the "Labels" control to set node dimensions, manipulate the "Lines" parameter to achieve suitable connection line thickness, and click "Curved lines" to select straight or curved lines for the connections.
9. Click "File", then "Screenshot", and finally "Save" to export the visualization.
10. Launch Microsoft Excel, sort the research institutions by publication output and citation counts, select the top 10 institutions, identify their respective countries, and prepare this dataset for visual presentation.
11. Select the three data columns containing the top 10 research institutions by publication output, their respective publication counts, and host countries. Then navigate through Insert > Charts > Column Chart to create a visualization of the top 10 research institutions ranked by publication volume.
12. Select the data labels to display the actual values, and export the chart in TIFF format.
13. Select the three data columns containing the top 10 research institutions by citation count, their corresponding citation numbers, and host countries. Then navigate through Insert > Charts > Column Chart to create a visualization of the top 10 research institutions ranked by citation count.

5. Journal publication output and citation analysis

NOTE: Perform analysis of journal publication output and citation using VOSviewer software.

1. Launch the software, then navigate through and click: Create → Create a map based on bibliographic data → Read data from bibliographic database files.
2. Import the plain text dataset, then click 'Next' to read the data.
3. In the "Type of analysis" section, select "Citation"; then choose "Sources" under "Unit of analysis" and click "Next". Set the value in "Minimum number of documents of a source" field to 5, set "Minimum number of citations of a source" to 0, click "Next", and proceed with data analysis.
4. Right-click on the data in the "Source" column and select the "Save As" option to export both the journal publication output and citation counts for subsequent data analysis.
5. Click "Finish" to generate the journal collaboration network map, then select the "Items" tab in the left toolbar to view the cluster count.
6. Click "Analysis" in the left toolbar and set the "Minimum cluster size" parameter to maintain the number of clusters within an appropriate range.
7. Navigate to "Analysis" and then "Layout" in the left toolbar. Adjust the values for the "Attraction" and "Repulsion" parameters to achieve an optimally scaled visualization.
8. Select "Visualization" on the right to modify the appearance of the visual map. Drag the "Scale" slider to adjust journal names to an appropriate size, use the "Labels" control to set node dimensions, manipulate the "Lines" parameter to achieve suitable connection line thickness, and click "Curved lines" to select straight or curved lines for the connections.
9. Click "File", then "Screenshot", and finally "Save" to export the visualization.
10. Launch Microsoft Excel and calculate the average citations per journal based on publication output and citation counts.
11. Select the three columns of data (journal titles, publication output, and average citations), then navigate to Insert > Charts > Column Chart to create a visualization comparing journal publication output and average citation rates.
12. Select the data labels to display the actual values, then export the chart in TIFF format.

6. Collaboration network analysis of high-productivity authors

NOTE: Perform collaboration network analysis of high-productivity and highly-cited authors using VOSviewer software.

1. Launch the software, then navigate through and click: Create → Create a map based on bibliographic data → Read data from bibliographic database files.
2. In the "Type of analysis" field, select "Co-authorship"; then choose "Authors" under "Unit of analysis" and click "Next". Set the value in "Number of organizations to be selected" to 95, proceed by clicking "Next", and initiate the data analysis.
3. Right-click the data in the "Source" column and export metrics, including publication counts and citation numbers of high-productivity authors, for subsequent data.
4. After clicking "Finish" to generate the institutional collaboration network, select the "Items" tab in the left toolbar to view the cluster count.
5. After clicking "Finish" to generate the author collaboration network, select the "Items" tab in the left toolbar to view the cluster count.
6. Navigate to "Analysis" and then "Layout" in the left toolbar. Adjust the values for the "Attraction" and "Repulsion" parameters to achieve an optimally scaled visualization.
7. Select "Visualization" on the right to modify the appearance of the visual map. Drag the "Scale" slider to adjust journal names to an appropriate size, use the "Labels" control to set node dimensions, manipulate the "Lines" parameter to achieve suitable connection line thickness, and click "Curved lines" to select straight or curved lines for the connections.
8. Navigate to File > Screenshot > Save to export the visualization.
9. Launch Microsoft Excel, sort the high-productivity authors by publication output, select the top 10 authors, and prepare this selection for visual presentation.
10. Select the top 10 high-productivity authors and their corresponding publication counts. Then navigate through Insert > Charts > Column Chart to create a visualization with author names on the horizontal axis and publication counts on the vertical axis. Enable the display of data labels. Finally, right-click the chart and save it in TIFF format.

7. Keyword co-occurrence and thematic evolution analysis

NOTE: CiteSpace was selected for keyword analysis due to its advanced capabilities in detecting research hotspots, cluster analysis, and temporal evolution patterns.

1. Launch the software, duplicate records were identified and removed using CiteSpace's deduplication function. Navigate to Data > Import/Export, set the Input Directory to the "input" folder, and configure the Output Directory to the "output" folder.
2. Navigate to Deduplication & Organization, select both "Article" and "Review" document types, and click "Start". Then import the plain text dataset and proceed by clicking "Start" again.
3. Click "Close" to exit the settings page and return to the main interface.
4. Click "New" on the main interface, set the Project Home to the "project" folder, configure the Data Directory to the "data" folder, select "WoS" for Data Source and "English" for Preferred Language, and finally save the settings.
5. In the timeslice panel on the right, set the time range from January 2015 to December 2024. Select "Keyword" as the node type. Choose the g-index (k=25) for network scaling. Enable the "Pathfinder", "Pruning sliced networks", and "Pruning the merged network" options in the pruning settings to simplify and clarify the network visualization.
6. Click "Start" to initiate data processing. After the operation is complete, select the "One-Click Clustering Label Optimization" function to generate the keyword clustering diagram.
7. Adjust the node labels and cluster labels in the right control panel to modify elements such as node size in the visualization.
8. Navigate to Toolbar > Label > Label Color > Article Labels to customize the label colors; then select the Node option in the toolbar to modify node attributes such as colors and shapes.
9. After completing the adjustments, navigate to File > Save As and export the image in PNG format.
10. Navigate to Layout > Timeline in the control panel to generate the keyword timeline view.
11. In the control panel, adjust the node labels and cluster labels to modify the node size and cluster label size.
12. Navigate to File and save the image in PNG format.
13. Navigate to Burstness in the control panel, sequentially click Refresh and View, enter "20" in the data field, confirm the selection, and generate a research hotspot analysis diagram of the top 20 keywords.
14. Capture and save the research hotspot analysis diagram of the top 20 keywords using a computer screenshot tool.

Results

Over the past decade, 1,571 articles on atopic dermatitis and allergic rhinitis have been published. The annual publication volume showed an overall upward trend, while citation frequency demonstrated steady growth (Figure 1). In terms of countries, the United States published the most articles (244). Further visualization of research clusters and collaborative networks among countries revealed three major national research clusters, represented by the United States, China, and Germany. The ...

Discussion

The bibliometric analysis conducted in this study provides a comprehensive mapping of the research landscape on the comorbidity between AD and AR. This approach not only highlights global research trends but also identifies key contributors, emerging themes, and potential future directions for research in this important field.

Over the past decade, publications on the comorbidity of AD and AR have shown a fluctuating upward trend, reflecting growing global interest in this field. However, afte...

Materials

List of materials used in this article

Name	Company	Catalog Number	Comments
CiteSpace	Chaomei Chen, Drexel University	CiteSpace 6.2.R4 (64-bit) beta Basic	Used for keyword co-occurrence analysis, cluster analysis, timeline visualization, and burst detection to identify research hotspots and thematic evolution.
Scimago Graphica	Spanish National Research Council (CSIC)	Scimago Graphica Setup 1.0.49	Used to visualize national collaboration networks and generate world maps based on VOSviewer output data.
VOSviewer	Centre for Science and Technology Studies, Leiden University, The Netherlands	VOSviewer version 1.6.19	Used to construct and visualize collaboration networks for countries, institutions, and authors, as well as journal citation analysis.