Bibliometric study of the global scientific research on pain and disability


 Background A growing number of relevant studies on the comorbidity of pain and disability have been published in high-quality journals; however, few studies have analysed quantitatively the characteristics of the topic via bibliometrics. Purpose The study aims to systematically analyze various aspects of global scientific research on pain and disability comorbidity from 1980 to 2019 and determine the global research trend and knowledge structure. Methods A collection of publications on pain and disability comorbidity between 1980 and 2019 was retrieved from the Web of Science Core Collection. CiteSpace, Microsoft Excel 2016, and IBM SPSS Statistics 25 software were used to analyze publication outputs, authors, journals, institutions, countries, subject category, types of pain, references, and keywords. Results We collected a total of 3,570 articles in March 9, 2019. Results indicated that the publication outputs had a significant growth in the recent 40 years via the linear regression model (P<0.001). The journal Pain had the maximum output (178), the most cited frequencies in WoS (15,418), the highest impact factor index in 2017 (5.559), and the highest H-index (68). Linear regression analysis revealed that the tendency of multiple authors’ papers increased significantly (t=8.065, P<0.001). In accordance with the top 20 subject categories, Clinical Neurology was the most productive academic area (1095). With regard to types of pain, low back pain and headache were the most widely concerned with a total of 1,114 publications.Conclusion The study outlines the evolution and development of comorbid pain and disability research, reflects the overall situation and lays the groundwork for future research.


Introduction
By almost any measure, pain is a global public health problem (1). It has multiple sequelae, including disability, low quality of life, depression and anxiety, sleep disturbances, and even suicide (1)(2)(3).
Approximately 20% of adults are afflicted with pain every year (1). The weighted mean prevalence of chronic pain in adults is 20% (2), which is higher than that of diabetes (4) or asthma (5). In 2010, the totals cost in the USA ranged from $560 to $635 billion, and additional healthcare expenditure due to 3 pain ranged from $261 to $300 billion (6). Disability is widely prevalent among adults globally, which increases with age (7). Approximately 22.2% of American adults suffer from any disability in 2013 (7), and the global morbidity of disability was 14% (8). Among eight common diseases requiring rehabilitation, stroke is the most pervasive cause of disability, followed by arthritis and back pain (9).
In 2006, the annual healthcare costs associated with disability reached nearly $400 billion in the USA (10).
In the top 10 diseases contributing to the most years lived with disability, five are defined or characterized by pain. Thereinto, low back pain (LBP) is ranked first, followed by neck pain, other musculoskeletal disorders, migraine, and falls (11). Accumulating evidence suggests that chronic pain is a serious problem for physically disabled adults (12)(13)(14). The fear avoidance model of chronic pain indicates that patients develop fear of movement, leading to activity avoidance, further increasing disability (15). One study found that the prevalence of disability was 65% among LBP patients, and among them, 80.7% had moderate to severe disability (16). Individuals who deal with pain more actively are more likely to prevent and intervene in disability (17).
A growing number of relevant studies on the comorbidity of pain and disability have been published in high-quality journals; however, few studies have explored the characteristics of the topic via bibliometrics. Bibliometrics is a cross-disciplinary science that uses statistical and mathematical methods to quantitatively analyze all knowledge carriers (18). It is commonly regarded as a useful and effective tool to evaluate developing trend and scientific productions in a specific domain (19,20). This analysis can assess the productivity of authors, countries, and institutions; display collaborations in countries, institutions, and geographical distribution; and reveal research frontiers, hotspots, and trends (21,22). Recently, bibliometric analysis has been widely used in different areas, such as headache (23), neuroscience (24), drug delivery (25), long noncoding RNA (26), and so forth.
The objective of the present study is to systematically analyze various aspects of global scientific research on pain and disability comorbidity from 1980 to 2019. We used CiteSpace V to analyze pertinent papers retrieved from the Web of Science Core Collection (WoSCC), determine emerging trends, extract the knowledge pattern, and capture hidden information. 4 2. Methods

Sources of data
The WoSCC was selected as the major database to accomplish our search. We obtained literature published between 1980 and 2019 from the Science Citation Index-Expanded and Social Sciences Citation Index. The following retrieval strategy was used: TI=(pain* or headache* or "head ache*" or head-ache* or migraine* or cephalalgi* or "stomach ache*" or "tummy ache*" or "abdominal ache*" or "belly ache*" or earache* or ear-ache* or toothache* or tooth-ache* or odontalgi* or dysmenorrh* or neuralgi* or cervicodyn* or analg* or nocicept* or hyperalg* or hypoalg* or fibromyalg* or radiculalg* or colic or sciatic* or arthralg* or causalg* or eudyn* or maldyn* or brachialg* or ophthalmodyn* or cephalalg* or otalg*) and TI=(disability or disabilities or Disabled or dysfunction or dysfunction* or disfunction* or Impairment* or handicap*).

Inclusion criteria
The document types were limited to articles and reviews. Other types, such as meeting abstracts, editorial materials, letters, proceedings papers, corrections, book reviews, notes, news items, discussions, correction additions, early accesses, reprints, and retracted publications, were excluded.
With respect to language, we only selected literature in English. Moreover, we limited our analyses to publications related to the comorbidity of pain and disability research. No more restrictions were made.

Data extraction
We collected a total of 3,570 articles in March 9, 2019. The publications were extracted and downloaded as text-and table-based documents, with some basic information presented intuitively through various charts. The bibliometric indicators were classified into two categories: (i) Publictivity indicator: annual publication outputs, distribution of journals, countries, institutions and authors, subject category distribution, and pain type distribution. (ii) Impact indicator: H-index of each journal or researcher, citation counts that papers received, IF of the journals, and relative citations per paper.
The relative citations of top articles were calculated as follows: (the number of citations that a paper received in WoS)/(2019-the year that the paper was published).

Publication outputs and growth trends
We collected a total of 5,569 papers at first and excluded 1,885 non-article or non-review papers and 114 non-English papers. Eventually, a total of 3,570 papers met the inclusion criteria ( Figure 1A).  We divided the period studied into eight time partitions, 1980-1984, 1985-1989, 1990-1994, 1995-1999, 2000-2004, 2005-2009 Table 1 The dual-map overlay of journals presented the distribution of journals across subject and paths of citation ( Figure 2). The citing pattern was on the left, and the cited pattern was on the right with labels regarded as the disciplines involved in corresponding periodicals. The citation line was connected from the citing periodicals to the cited periodicals. Notably, these articles were mainly published by journals in the field of neurology, sports and ophthalmology, and a majority of them cited psychology, education and social journals.

Distribution by countries and institutions
The 3,570 papers on pain and disability comorbidity were contributed by 82 countries (Additional file 2, Figure 3A). In terms of publications, USA was the most prolific country (1,267), followed by England

Distribution by authors
Included papers on pain and disability comorbidity were distributed over 12,200 authors. Figure 4A shows that the node representing Lipton RB was the largest in terms of maximum publications (37), followed by Bussone G (28), Jensen MP (26), D'Amico D (23), and Grazzi L (20). Only the proportion of Lipton RB was more than 1% (1.036%). It also outlined the cooperation between different authors.
From the data generated by CiteSpace V, the value of every author centrality was 0, indicating that collaboration among these authors was not sufficient and frequent. The author's co-citation map demonstrated that Waddell G had the largest node with 386 citation counts, followed by Turk DC with 302 citation counts, Jensen MP with 300 citation counts, Deyo RA with 299 citation counts, and Ware 8 JE with 277 citation counts ( Figure 4B, Table 3). Figure 4C shows the trends in percentage of multiple- Arthritis was the topic that received the highest number of citations per paper. All top 10 types of pain increased significantly in publication over time through linear regression analysis.

Analysis of references
The co-citation reference map presented clear scientific relevance of pertinent literature ( Figure 5C).
These references were divided into separate categories and arranged by timeline, which were labeled with title terms extracted from themselves. The modularity Q score was 0.9157 (close to 1), indicating that the map had a good and reasonable clustering effect and distinctive knowledge feature. The "pain lead" was the largest cluster #0 with most papers on the topic, "affecting pain" was the second 9 largest cluster #1, and "predicting disability" was the third largest cluster #2. Apparently, the node marked as "Fear-avoidance and its consequences in chronic musculoskeletal pain: a state of the art," belonging to cluster #21 (pain-related fear), was the largest. The next most cited were "Pain-related fear is more disabling than pain itself: evidence on the role of pain-related fear in chronic back pain disability" and "The fear-avoidance model of musculoskeletal pain: Current state of scientific evidence" (Table 2). References published in the past several years were commonly concentrated in cluster #0 (pain lead), cluster #4 (subtle objective), and cluster #17 (patient-reported outcome instrument).

Analysis of keywords
A total of 68 keywords with the strongest citation bursts were obtained ( Figure 6). Some keywords, including behavior, arthriti, health status, rheumatoid arthriti, fibrositi, illness, compensation, and sickness impact profile, were the first cited frequently in 1991. Basically, the citation burst continued until around the year 2000. Rheumatoid arthriti had the longest citation burst time and maximum strength. Disease, association, women, people, and children were all cited until 2019 and will likely to last over the coming years. Table 3  Among the top 10 countries, two were in North America, six were in Europe, one was in Oceania, and one was in Asia. Only Turkey was the developing country, which revealed that developed countries were on a leading position and accomplished quite a few research achievements in the field. The USA had a prominent scientific research strength with the maximum number of publications. Figure 3C shows that England had comparably strong and extensive international cooperation compared with others. The closest partnership was identified in England, Scotland, Canada, Turkey, Finland, Sweden, and Germany. Taking into account publications and centrality, England and Australia were the main core power in the comorbidity of pain and disability research. The USA was the most productive country, but its value of centrality was 0. From the perspective of academic exchanges and research quality, the USA was in the marginal position.
Among the top 10 institutions, six were from the USA, which is why the country had significant advantage in the quantity of publication. Harvard University made great research progress with the maximum number of publications. Figure 3E shows that the University of South Australia had relatively stronger and broader cooperation compared with others. The strongest collaboration identified was among the University of South Australia, University of Sydney, University of Alabama Birmingham, Vrije Universiteit Amsterdam, and Neuroscience Research Australia. Compared with the map of national cooperation, the institutions were less sufficient.

Citation data and research focuses
Among the top 10 authors according to publications and co-cited counts, Jensen MP, who published several high-quality papers and promoted the progress of the comorbidity of pain and disability research, could be regarded as the most influential researcher. He and his team mainly studied the influence of psychological intervention, cognitive restructuring intervention, psychosocial factors for chronic pain in individuals with disabilities, the associations between chronic pain and disability, and the assessment of function and psychology of chronic pain patients with disabilities (12,13,(27)(28)(29).
However, most prolific authors did not have corresponding high co-citation frequency, indicating that their research results did not make a great difference on the development of this field and that they still need to further improve article quality. Waddell G, who had the maximum number of co-cited rates, analyzed the influence of fear-avoidance beliefs on chronic LBP (CLBP) and disability, explored the association between LBP and disability, and effectively evaluated physical impairment due to CLBP (30)(31)(32). In Figure 4A, we could plainly identify that the network was very loose, hinting that the authors seldom communicated and cooperated with each other. Through analysis of multiple-and signal-authored articles, the proportion of collaboration among multiple authors gradually rose, and the network also became denser. In the timeline map of co-citation reference, cluster #0 (pain lead) was the largest, revealing that several articles cited references related to "pain lead." The period from 1998 to 2016 contained several influential references with highly cited frequencies, tallying with the golden stage between 1999 and 2018 in terms of publications (as mentioned earlier). The reference entitled "Fearavoidance and its consequences in chronic musculoskeletal pain: a state of the art" was cited the most; it mentioned that fear and avoidance caused by pain was a reason for some patients with musculoskeletal pain to develop chronic pain disability (34). In the last 40 years, the focus on comorbid pain and disability research gradually transformed from characteristics to application.
The top 10 papers were selected according to citation counts in WoS. "A study of the natural-history of back pain .1. development of a reliable and sensitive measure of disability in low-back pain" had the maximum number of citations, which provided a questionnaire for assessing disability due to back pain (35). "Burden of migraine in the United States -Disability and economic costs," with the highest value of impact factor, quantified the economic costs of migraine in the USA, which caused a loss of approximately $13 billion to the employers and cost around $1 billion of treatment per year (36). "The global burden of headache: a documentation of headache prevalence and disability worldwide" 13 received the largest number of relative citations. In the article, Stovner, LJ reported that headache disorders would be one of the 10 most disabling reasons in the whole crowd, and one of the five most disabling reasons among females (37). A number of articles published in Pain attracted widespread attention. These high-impact articles were the knowledge base in the research. Analysis of the top 10 papers revealed popular topics covered in the past 40 years, including i) LBP/headache/migraine/neck pain, ii) Disability, iii) economic burden, iv) questionnaire, v) treatment, and vi) pain-related fear.
Burst keywords, which were suddenly cited frequently for some time, could predict future development direction on comorbid pain and disability research. The words related to research frontiers were as follows: children, people, women, association, and disease. The risk of pain and disability increases with age (7,38). Compared with males, females showed higher prevalence (2,7).
Patients with comorbid pain and disability are accompanied by many other diseases. In terms of cooccurrence counts and centrality, depression and validity could be considered as the core keywords reflecting research focuses and hotpots, which implies that psychosocial factor plays a significant role in the development of comorbid pain and disability.

Strengths and limitations
This is the first bibliometric study to assess the state and developing trends of pain and disability comorbidity research during the last 40 years. WoS was chosen as the database to extract abundant information, covering not only publications, citations, authors, countries, institutions, journals, and IF but also contained abstracts, references, subject categories, and so forth. On this basis, we could make an in-depth analysis and excavate valuable information hidden in the original data. Furthermore, our research was not limited to the last 20 years or a single country or journal, which provides more reliable and complete analysis.
Notwithstanding these strengths, some limitations of the study should be considered. First, we only retrieved relevant papers in WoS; other databases (e.g., PubMed, Scopus, Embase, and Ovid) were not analyzed. Second, we excluded some non-English papers, which may have high quality and impact. Third, these data are constantly changing because the WoS database updates regularly.
Lastly, some potential influential papers were not outstanding with few cited frequencies due to time 14 lag. These biases generated differences between theoretical analysis and actual scientific condition.
Despite these limitations, we believe our conclusion could reflect the overall and comprehensive situation, laying the groundwork for future research.

Conclusions
Overall, on the basis of bibliometric analysis, this study outlines the evolution of comorbid pain and disability research from 1980 to 2019. The peroid of 1999-2018 was considered as the golden age for the development of the study, with lots of influential papers. Pain and Spine were the core journals, making a great contribution on disseminating research findings and attracting researchers' attention.
The proportion of multi-author articles has increased significantly, but the collaboration between authors is not very sufficient. In addition, researchers expressed much interest on low back pain and headache according to the type of pain. This study helps relevant researchers to pioneer directions for future new research in the coming years.

Consent for publication
Not applicable.

Availability of data and materials
All data generated or analysed during this study are included in this published article [and its supplementary information files].

Competing interests
The authors declare that they have no competing interests.

Authors' contributions
Lin-Man Weng and Xue-Qiang Wang conceived and designed the study. Lin-Man Weng, Yi-Li Zheng, Meng-Si Peng, Juan Wang, and Rui Wang collected and analysed data. Lin-Man Weng was the major contributor in writing the manuscript. All authors read and approved the final manuscript.    Dual-map overlay of journals related to comorbid pain and disability research.  Top 68 keywords with the strongest citation bursts.

Supplementary Files
This is a list of supplementary files associated with this preprint. Click to download.