The 100 Top-Cited Studies on Loneliness: A Bibliometric Analysis

The present study explores characteristics of the top 100 most-cited articles on loneliness. A systematic search was carried out using the Thomson Reuters Web of Science Core Collection to collect studies on loneliness from inception to June 1, 2022. The initial search resulted in 6,250 search results, which included articles, book chapters, conference proceedings, editorials, and letters. Two authors independently screened the literature and extracted the data. The study supervisor removed any discrepancies. Top 100 papers (articles and reviews) on loneliness published in English were extracted. Data analysis and visualization were performed on Excel, Web of Science (WoS) Data Analyzer, and VOSviewer 1.6.16. The total number of citations of the 100 top-cited articles was 42,044, ranging from 203 to 2,201 per article. All of the studies were published from 1989 to 2021, and the years with the highest number of top-cited articles published were 2003 and 2008. Most publications were from the following journals: Computers in Human Behavior, Developmental Psychology, Psychological Science, Psychology, and Aging (n=4 each). The most cited article was titled “UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure” by Russell, DW, in the Journal of Personality Assessment. The most productive institute was the University of Chicago. The two most productive authors were Cacioppo, JT, and Hawkley, LC. Of the 100 top-cited publications, 87 were original articles and 13 were reviews. The top three WoS categories were psychology multidisciplinary, gerontology, and psychiatry. In total, 37 author keywords were elicited and further clubbed into eight distinct clusters. The study provides new insight into loneliness research, which may help doctors, researchers, and stakeholders achieve a more comprehensive understanding of trends and influential contributions to the field and highlight under-researched areas, which could be the basis for future investigation.


Introduction And Background
Although many definitions exist [1], loneliness is generally defined as the "unpleasant experience that occurs when a person's network of social relations is deficient in some important way, either quantitatively or qualitatively" [2]. Loneliness is linked to early mortality risk [3], depressive symptoms [4,5], and sleep disturbance [6], which leads to increased use of health services [7], and is found in both men and women [8] across ages [9][10][11][12][13][14]. The issues of loneliness have increased during COVID-19 [15][16][17]. Even prior to COVID-19, loneliness was considered one of the emerging public health concerns, and increasing attention was being paid to loneliness due to an improved understanding of the impacts it has on individuals and communities, as highlighted by the appointment of the Ministry of Loneliness in the UK and Japan [18]. With the increased interest in loneliness research, many are published yearly. Thus, a bibliometric analysis helps summarize the existing research and elicit key insights into loneliness research.
Bibliometric analysis is an emerging field that evaluates a series of academic publications such as journal articles, chapters, and other scientific publications to provide evidence and insights into a field, subject, or discipline [19]. It enables external quality evaluation of studies published within a defined timeline. It also enables progress mapping, identifying trends, and gap finding for future studies [20]. Bibliometric analysis assesses authors, institutions, and countries of research articles by mapping the structure and dynamics of discipline using a specific source [21].
The bibliometric analysis focuses on data from academic databases such as the Web of Science Core Collection (WoSCC), Scopus, and Dimension. One of the frequent types of bibliometric analysis, citation analysis, informs that the frequency of an article mentioned reflects its relative importance within that area. Recognizing the highly referenced papers and emerging papers in a field may help researchers point them in new directions. Since predatory journals are difficult to track, citation analysis helps find premier authentic journals with authors with the most impact in the field. Several bibliometric publications have recently focused on identifying the top 100 studies in a field. The focus has been on fields that publish too many papers. Therefore, keeping track of the latest development and pioneer works within a field is essential. Bibliometric analysis is a helpful tool to get an overview of major public health concerns such as cancer [22], diabetes [23], rheumatoid arthritis [24], global malnutrition [25], and ageing [26], along with new emerging 1 UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure [31] Russell Journal of Personality  Year of publication   Table 2. Psychology and Aging, Developmental Psychology, Computers in Human Behavior, and Psychological Science had the highest publication (n=4) each. The top 10 journals published 34 articles cumulatively, and 44 journals had a single publication in the list of the top 100. The impact factor of journals ranged from 2.16 (Journal of Aging and Health) to 8.19 (Perspectives on Psychological Science). We did not find a relationship between the number of published studies and the journal's impact factor (p > 0.05). Also, we did not find a relationship between the average number of citations for each journal and the journal's impact factor (p > 0.05).  Latest journal impact factors were updated from the journal homepage.

Country
A total of 16 countries (excluding Belgium and Scotland) contributed to the top 100 cited publications ( Table  3). The most productive countries were the USA (n=51), England (n=17), Australia, Canada, Germany, Israel, and the Netherlands (n=4). The other countries contributed to less than three publications. The largest set of connected countries consisted of nine countries that could be divided into three distinct clusters (Figure 4). The first cluster consisted of England, New Zealand, China, and Scotland; the second cluster consisted of the USA, Germany, and Israel; and the third cluster consisted of the Netherlands and Belgium. Network overlay visualizations of the co-authorship of countries are shown in Figure 3. The USA dominated the earliest publications, with recent publications from Belgium and Scotland.  If a study had a different first and corresponding author's country, the corresponding author's country was considered in the list.

FIGURE 4: Network overlay of co-authorship map of country of top 100 publications Authors
Nine authors published at least two top-cited papers as first or corresponding authors listed in Table 4.
Hawkley and Cacioppo published seven publications, each as either the first author or corresponding author. They also published most papers together. The co-authorship map of authors in Figure 5 shows that Cacioppo is the most influential author in loneliness publications.   The institute of the first author was used for data analysis

Study type and Web of Science categories
Of the top 100 cited publications, 87 were original articles and 13 were reviews ( Table 6). The average citation for articles was 420.1 and that for reviews was 424.7. Eighteen Web of Science (WoS) categories had at least two top-cited papers. The top three categories were psychology multidisciplinary (n=24), gerontology (n=21), and psychiatry (n=17).

TABLE 6: Publication by WOS category of top-cited papers with at least two papers
If an article had more than one WoS category, both categories were used for data analysis.

Author keywords
The author keyword with a minimum of two occurrences and without "loneliness," which occurred in 69 titles, was used for analysis ( Figure 6). A total of 37 keywords were elicited, which could be clubbed into eight distinct clusters: 1) attachment, longitudinal study, mortality, social network, and social support; 2) emotional loneliness, health, loneliness scale, partner status, and social loneliness; 3) age, older people, social isolation, social neuroscience, and systematic review; 4) ageing, blood pressure, cortisol, and longitudinal; 5) anxiety, cognition, depression, and prevalence; 6) COVID-19, loneliness risk factors, mental health, and older adults; 7) intervention, physiology, sleep, stress; 8) Facebook, Internet, personality, and shyness.

Discussion
Research in the field of loneliness has increased over time. A review of the most influential papers and authors can provide a detailed overview of research trends for researchers and practitioners. In this review, we tried identifying and assessing the bibliometric characteristics of the top 100 loneliness-related publications. To our knowledge, no citation analyses have examined publications on loneliness. Therefore, this study represents the first comprehensive analysis of published literature in the field of loneliness. WoSCC database, which provides in-depth insights into multiple study characteristics [25], initially resulted in 6,250 publications. The top 100 publications (original articles and reviews) on loneliness based on the number of citations were extracted.
The study found that the total number of citations of the 100 top-cited articles was 42,044, ranging from 203 [121] to 2,201 [31]. The average citation per document was 420.44. The most cited article was published in the Journal of Personality Assessment. The University of California, Los Angeles (UCLA) loneliness (version 3) is the most frequently used uni-dimensional measure of loneliness. The measure is a revised version of the original UCLA Loneliness Scale [122] and the Revised UCLA Loneliness Scale [123]. The scale has been used extensively for a diverse population with different norms available, showing researchers' preference to use this scale as a measurement of loneliness. The current review's finding aligns with other bibliometric studies, which found that self-report measures usually receive high citations [20]. However, the scale has been critiqued as it measures loneliness as a personality trait or a state-related entity [124]. An overview of the publication year suggests that research on loneliness is relatively new compared to other psychological issues such as depression [125,126]. (Perspectives on Psychological Science). We did not find a relationship between the number of published studies and the journal's impact factor. Also, we did not find a relationship between average number of citations for each journal and journal's impact factor. Similar results were found in past research [20] in the study of top-cited papers in neuropsychology showing that journal impact factor is not the most important factor in journal selection.
The most productive country was the USA, followed by England. The largest set of connected countries comprised nine countries that could be divided into three distinct clusters. The first cluster consisted of England, New Zealand, China, and Scotland; the second cluster consisted of The USA, Germany, and Israel; and the third cluster consisted of the Netherlands and Belgium. Network overlay visualizations of the coauthorship of countries show the USA dominated the earliest publication, with recent publications emerging from Belgium and Scotland. The impact of developed countries such as the USA and England can be attributed to more available research resources. Moreover, most countries in top-cited papers are individualistic [127]. Past studies have shown that culture plays an important role in the experience of loneliness [128][129][130]. Therefore, more research and newer findings from collectivist countries can enrich our knowledge on loneliness. Hawkley and Cacioppo published seven publications, each as the first or corresponding author. Cacioppo et al. have published extensively on the relationship between loneliness and its impact on the brain and physical health, mainly focusing on an evolutionary psychology approach [131]. The most productive institute was the University of Chicago, with 17 publications, followed by the University of London, with seven publications. Out of 100 top-cited publications, 87 were original articles and 13 were reviews. The top three WoS categories were psychology multidisciplinary with 24 publications, gerontology with 21, and psychiatry with 17. Out of the top five categories, two WoS categories were of geriatrics, highlighting that loneliness has been extensively studied as a problem of the elderly. Focusing on other age groups can open new research avenues.
The co-occurrence of author keywords showed that loneliness research had been conducted in different viewpoints and fields. A total of 37 author keywords were elicited, which could be clubbed into eight distinct clusters. The first cluster comprises loneliness attachment, longitudinal study, mortality, social network, and social support. A multi-dimensional perspective of loneliness [132] dominates the second cluster with keywords such as emotional loneliness, social loneliness, health, loneliness scale, and partner status. The third cluster is dominated by the elderly and geriatric population, which consists of keywords such as age, older people, social isolation, social neuroscience, and systematic review. The fourth cluster consists of keywords from the evolutionary theory of loneliness [131], such as ageing, blood pressure, cortisol, and longitudinal. The fifth cluster comprises psychological problems such as anxiety, cognition, depression, and prevalence. The sixth cluster shows emerging topics of loneliness during COVID-19, such as loneliness risk factors, mental health, and older adults. The seventh cluster consists of intervention, physiology, sleep, and stress. The eighth cluster consists of keywords focusing on enduring traits of individuals, such as personality and shyness. This cluster also focuses on lonely individuals' online behavior and consists of keywords such as Facebook and Internet.

Limitations
The current research has some limitations. We did not collect data from other databases, such as Scopus, Medline, or Google Scholar, because we used the WoS database exclusively for our analysis. Second, we only considered research that had been published in English. Third, since the number of citations to each publication changes over time, the top 100 cited papers will also vary. Fourth, we may have missed some studies that did not contain the term "Loneliness" or "Lonely" or "Loner" in the title, as we did not consider studies where the core word was not used in the title. Fifth, citation rates are affected by many factors, many of which are beyond the scope of this study. In the future study, we will include more databases and dynamically track changes in these studies.

Conclusions
In the present bibliometric study, we identified and analyzed the 100 top-cited publications on loneliness. With a recent surge in top-cited publications, loneliness has become a growing public issue. However, Individualized countries still dominate influential publications with leading researchers from the USA. The review provides insight into the most impactful loneliness studies and highlights critical and insufficiently investigated topics. More research from Asian countries will help practitioners and other stakeholders to enhance their understanding of the trends and influential contributions to loneliness research.

Conflicts of interest:
In compliance with the ICMJE uniform disclosure form, all authors declare the following: Payment/services info: All authors have declared that no financial support was received from any organization for the submitted work. Financial relationships: All authors have declared that they have no financial relationships at present or within the previous three years with any organizations that might have an interest in the submitted work. Other relationships: All authors have declared that there are no other relationships or activities that could appear to have influenced the submitted work.