Psycho-Oncology: A Bibliometric Review of the 100 Most-Cited Articles

(1) Background: A bibliometric review of psycho-oncology research is overdue. (2) Methods: The 100 most-cited journal articles were compiled and ranked according to Scopus. (3) Results: The total citation count for the results ranged from 488–8509 (Mean = 940.27; SD = 1015.69). A significant correlation was found between years since publication and number of citations (p = 0.039). The majority of research originated from the United States (66%). The vast majority of research publications were original articles (80%). Observational research study designs represented the majority of studies (37%). Mixed cancer population research studies represented the largest cancer research population (36%). Positive psychology topics represented the most prolific proportion of studies (30%). Findings were reported in line with PRISMA-ScR guidelines. (4) Conclusions: This analysis offers a comprehensive account of seminal journal articles in psycho-oncology, identifying landmark contributions and areas for future research developments within the field, namely highlighting a need for more RCT studies. This analysis serves as an educational tool for interdisciplinary researchers and clinicians to support compassionate cancer care.


Introduction
Psycho-oncology is a collaborative, cross-disciplinary subspecialty of oncology with domains in the psychological, social, behavioural, and ethical aspects of cancer in clinical care [1,2]. The discipline provides clinical and research material about issues clinically relevant to health professionals who provide psychosocial services to cancer patients, their families, and their caregivers [3]. The foundations of the field first came into existence in the 1970s [1], the evolution of which has previously been detailed by the founder of the field, Dr Jimmie Holland [1,3].
Overtime a large body of literature has been published comprising a wide range of relevant research and clinical themes. A previous review of this wealth of literature by Greer outlines the important need to "close the yawning gap between current knowledge and therapeutic skills on the one hand and actual clinical care of cancer patients on the other" [4]. As the discipline approaches fifty years since formal foundation, a bibliometric review of the literature is warranted to aid the synthesis and implementation of the evidence base.
Citation count is an important metric in understanding the significance of a research contribution to a research field [5][6][7]. Situational analyses which identify research that has exerted significant citation influence offers researchers and clinicians an introduction to seminal research publications. It can be argued that the most-cited publications of a research field theoretically contribute the most to the respective field [8][9][10][11]. Notably, the approach has proved useful in practice-driven research funding decision-making by offering objective and reliable bibliometric quantitative analysis of grant productivity [8,12]. Bibliometric analyses with the aid of bibliographic electronic databases offer a systematic overview of peer-reviewed research in a range of disciplines and research 1.
Journal articles with primarily medical foci despite the inclusion of brief quality of life measures; 2.
Journal articles which described mixed patient populations beyond oncology; or 3.
Did not have psycho-oncology research or practice as key foci.

Data Extraction
Data were extracted independently by the two reviewers. Information was extracted on the following variables: (1) title; (2) authorship and publication year; (3) country of publication and first author's affiliation at the time of publication; (4) journal; (5) article type (e.g., intervention, systematic review); (6) article global subject (e.g., cancer prevention, psychoneuroimmunology or post-traumatic growth); (7) number of citations; (8) and citation rank. High percentage agreement between raters was found (percentage agreement = 94%).

Self-Citations
Using the "exclude self-citations" tool in Scopus, the percentage of self-citations within the list of 100 most highly cited articles derived from Scopus was calculated.

Statistical Analysis
The Pearson correlation coefficient (r) was calculated to determine whether the number of years since publication was correlated with total number of citations among the included articles.

Publication Trends
Additional searches using the terms "psycho-oncology" and "psychosocial oncology" were conducted within Scopus. These searches and the resulting data provide a broad overview of the publication trends of articles using these terms.

Study Selection
A PRISMA flow diagram for the Scopus results is provided in Figure 1. The initial search returned 197,569 results, of which the titles and abstracts of the 2000 highest-cited articles were screened using the eligibility criteria. Full-text screening was completed for 351 articles. Results were ranked according to citation counts to represent the 100 mostcited articles. A table of the included 100 publications and a citation details are presented in the Appendix A.

Self-Citations
Self-citations were found to represent 4.4 per cent of total citations retrieved from Scopus.

Study Characteristics
The characteristics of the articles retrieved are provided in Table 1. Of the 100 included papers, the highest-cited articles were published between 1992 and 2005. A significant correlation was found between years since publication and the number of citations (p = 0.039). The citation range was 488-8509 (mean = 940.27, SD = 1015.69). Similarly to recent bibliometric reviews [22,39] a word cloud of the words contained in the titles of the 100 included studies was generated using wordle.net in order to depict influential prevailing words and Healthcare 2021, 9, 1008 4 of 24 themes within the field of psycho-oncology. Popular words and phrases are highlighted based on frequency and relevance to the titles of the 100 included papers (see Figure 2).

Self-Citations
Self-citations were found to represent 4.4 per cent of total citations retrieved from Scopus.

Study Characteristics
The characteristics of the articles retrieved are provided in Table 1. Of the 100 included papers, the highest-cited articles were published between 1992 and 2005. A significant correlation was found between years since publication and the number of citations (p = 0.039). The citation range was 488-8509 (mean = 940.27, SD = 1015.69). Similarly to recent bibliometric reviews [22,39] a word cloud of the words contained in the titles of the 100 included studies was generated using wordle.net in order to depict influential prevailing words and themes within the field of psycho-oncology. Popular words and phrases are highlighted based on frequency and relevance to the titles of the 100 included papers (see Figure 2).

The 100 Most-Cited Articles
The distribution of results for the 100 most-cited articles is presented in Table 1. A comprehensive list of results is presented in the Appendix A.

Country of Publication
The country of origin of the first author for each article represented study origin data. Overall, 10 nations contributed to included study origin. The United States of America represented the largest contribution of studies (66%), followed respectively by the United Kingdom (12%) and Canada (10%). See all contributory countries in Table 1

The 100 Most-Cited Articles
The distribution of results for the 100 most-cited articles is presented in Table 1. A comprehensive list of results is presented in the Appendix.

Country of Publication
The country of origin of the first author for each article represented study origin data. Overall, 10 nations contributed to included study origin. The United States of America represented the largest contribution of studies (66%), followed respectively by the United Kingdom (12%) and Canada (10%). See all contributory countries in Table 1, panel 1.

Publication Type
The distribution of document type is presented in panel 2. Original articles represented 80% of the studies. Review papers and conference papers represented 18% and 2% of studies respectively.

Type of Study
Observational research study designs represented the majority of studies (37%). Cross-sectional observational designs represented the largest cohort of studies (20%) followed closely by prospective designs (17%). Tool development/evaluation, intervention

Publication Type
The distribution of document type is presented in panel 2. Original articles represented 80% of the studies. Review papers and conference papers represented 18% and 2% of studies respectively.

Type of Study
Observational research study designs represented the majority of studies (37%). Crosssectional observational designs represented the largest cohort of studies (20%) followed closely by prospective designs (17%). Tool development/evaluation, intervention and review studies each represented 21% of studies. Comprehensive distribution of study methodology is presented in panel 3.

Global Subject Topic
Positive psychology represented the largest proportion of included studies (30%), where the overarching global subject topic of studies examined psychological well-being and post-traumatic growth (14%), quality of life (14%), and mindfulness (2%). Clinical psychology global topics represented the second-largest cohort of studies (12%), where topics included psychological distress and mental health outcome including depression and suicidality were explored. Parallel psychological and physical health outcomes were the global subject topic for 2% of studies. Symptom prevalence represented 15% of study global subject topics, where 5% of studies examined pain prevalence and 10% explored additional cancer-related sequelae including the prevalence of cancer-related fatigue. Health promotion studies represented 6% of studies. These studies explored cancer prevention including self-monitoring behaviour, genomic testing, and survivorship intervention studies. Patient-physician communication and patient communication needs represented 10% of studies. Patient treatment choices including complementary and alternative medicine (CAM) represented 4% of studies. Palliative or supportive care studies represented 6%. Psychoneuroimmunology research represented 6% of the studies. Survivorship analyses represented 6% of the studies. Family system outcomes represented 1% of studies. See panel 4 for comprehensive results.

Cancer Population
The largest proportion of studies explored mixed cancer populations (36%) followed by breast cancer populations (25%), and advanced/terminal cancer populations (15%). A significant proportion of studies did not define the cancer population (13%). A further 5% of studies included prostate cancer patients. Lung and malignant melanoma patients each represented 2% of studies. Brain, cervical, gastric, laryngeal, and colorectal cancer populations each represented 1% of studies. See panel 5 for comprehensive results. The vast majority of the included studies examined adult populations (97%). The remaining 3% of the studies examined child populations.

Major Contributing Journals and Periods
The 100 most-cited articles were published in 46 journals; 17 journals represented more than one study. The major contributing journals are presented in Table 2  The 100 most-cited studies were published from 1975 to 2016. Figure Figure 4 provides an overview of publication trends within Scopus under the key terms "psycho oncology" and "psychosocial oncology". Publications under the term "psychosocial oncology" precede "psycho oncology" publications commencing in 1973. The term "psycho oncology" presents initially in 1979, demonstrating the evolution of the discipline. A peak in publications was observed in 2018 for both search terms.   1970 1972 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998   Publications by year for the terms "psychosocial oncology" and "psycho-oncology" in Scopus.

The 10 Most-Cited Articles
The ten most-cited studies are presented in Table 3. The articles included the following: three studies detailing the development and assessment of psychometric measures, two quality of life measures [40,41] and one pain measure [42]; two reviews, the first explores psychological adjustment to breast cancer diagnosis [43], and the latter explores the role of mindfulness in psychological well-being and includes a prospective mindfulness-based intervention for early-stage cancer patients [44]; one cross-sectional observational study which compares psychological distress prevalence by cancer site [45] two controlled trial studies, one randomised controlled trial exploring an early palliative care intervention for metastatic lung cancer patients [46]; and one prospective controlled trial exploring a psychosocial group-based intervention for metastatic breast cancer patients [47]; and finally two prospective cohort studies, one identifying the determinants of quality of life and satisfaction among prostate cancer survivors [48]; and one determining the impact of end-of-life patient-physician communication on patient mental health, medical care near death, and caregiver bereavement adjustment in advanced cancer patients and their family systems [49].

Major Contributing Authors
Overall, a total of 158 authors contributed to the results. There was wide, disparate authorship for first authors where 91 first authors represented the 100 included studies. Of these included studies, only one first author had published three studies as first author [47,50,51]. Three other first authors each published two studies as first author [40,[52][53][54][55][56]. Each of these authors contributed as co-authors to other studies indicating a psycho-oncological focus in their published work. Cella, D. contributed the largest number of studies to the research (n = 7) [40,52,[57][58][59][60][61]. Table 4 presents results for authors who contributed three or more of the 100 most-cited psycho-oncology articles.

Discussion
The aim of this review was to perform a bibliometric analysis of the 100 most-cited journal articles in psycho-oncology. It is, to the best of our knowledge, the first study to identify and describe the characteristics of highly cited journal articles and publication trends that have contributed to the development of the field.
The results of the bibliometric review provide a systematic overview of seminal research in psycho-oncology overtime. Our review presents a body of evidence which may have multiple applications for researchers and clinicians alike working in the field of psycho-oncology, including potential for the development of educational materials, journal editorial strategy, and future research.
In accordance with Scopus, our analysis revealed that the 100 most-cited articles were published between 1975 and 2016. This finding is in keeping with the timeline of previous reviews which describe the evolution of the discipline, from a time when a diagnosis of cancer was stigmatised and not openly disclosed to patients, and towards a time of more trauma-informed cancer care [1,2,38]. The 100 most-cited journal articles were all published by 2016 and the ten most-cited articles averaged 21.8 years since publication, indicating that the research exists along a developmental trajectory whereby time impacts on citation count and subsequent influence. The majority of research originated from the United States (66%). The vast majority of research publications were original articles (80%). Observational research study designs represented the majority of studies (37%). Mixed cancer population research studies represented the largest cancer research population (36%).
Our analysis revealed that positive psychology topics and clinical psychosocialoncology topics represented the most prolific proportion of included studies. This finding reflects one of the most fundamental questions that psycho-oncology seeks to understandhow do people with cancer feel? The global subject topics included in our analysis reflect the targets of previous narrative reviews of psycho-oncology [38]. Other subject topics included in our review explored parallel psychological and physical health outcomes, symptom prevalence including pain and cancer-related fatigue, health promotion and cancer prevention research including self-monitoring behaviour, genomic testing and survivorship intervention studies, patient-physician communication and patient communication needs, patient treatment choices including complementary and alternative medicine, palliative care research, psychoneuroimmunology, survivorship, and family system outcomes. Our analysis highlights the psychosocial transitory nature of cancer, which presents the potential for both positive and negative outcomes [62]. Findings reflect increased recognition for the "people part" of cancer care and the sixth vital sign in medicine-distress [63]. Enhanced patient participation and increased patient-physician communication in treatment decisions have been described in recent reviews of the field [3]. Beyond this, the analysis emphasises the impact of psychosocial factors in physical health and the growing attention that psychoneuroimmunology research has gained [64]. A paucity of highly cited research on adherance to cancer treatment was identified. Given the value of research on this subject topic for MDTs, this factor represents a deficit among included study topics.
Journal and author contributions were widely heterogeneous in nature. Our analyses revealed 91 first authors contributors across the 100 included studies. Notably, self-citations represented a very small percentage of citations (4.4%). A previous review of self-citations in research indicated that self-citations typically account for an average 10-20% of citation counts [65]. The 100 included articles were published in 46 journals, where 17 journals represented more than one study. Included studies were published in high-impact factor journals. Our analysis of highly cited journal articles reflects the interdisciplinary nature of psychosocial-oncology, which demonstrated the interfacing and overlapping boundaries with general medicine, oncology, psychiatry, pain medicine, health, and social psychology [2]. In keeping with this finding, interdisciplinary researcher and Chair of the Interdisciplinary Department of Medical Social Sciences at Northwestern University, Prof. David Cella, was the most prolific author) [40,52,[57][58][59][60][61]. Additional analysis of global publication trends within Scopus indicated that the term "psychosocial oncology" precedes "psycho-oncology". Although the percentage of publication increased over time, a noticeable peak in publications was observed in 2018 for both search terms, clearly demonstrating the dynamic evolution of the discipline. In addition to time, other secular trends such as increased capacity of search engines and access to research articles online positively impact citation count.

Strengths and Limitations
Beyond its novel contribution, this bibliometric analysis was strengthened by the use of two search methods. The keyword search enabled the identification of publication trends for psychosocial oncology in addition to psycho-oncology. This methodological consideration enriches the tapestry of the findings as psychosocial oncology terminology precedes psycho-oncology in the evolution of the field [1]. Additionally, the review was strengthened by its adherence to bibliometric technical methods [66,67]. A further strength of this study is the assessment of the prevalence of self-citations. Inclusion of this analyses explores academic biases which can artificially inflate citation impact rate by objectively assessing the impact of 'other-driven' citations [5,68]. Inversely, this bibliometric review is not without its own limitations. Specifically, a publication bias may have been induced by the methodological limitation to only include English language publications. This limitation may explain why the study origins of the leading contributing counties were Anglophonic countries, namely the USA, UK, and Canada, because seminal articles in other languages were not included.
Though comprehensive, our analysis was limited in that research influence was operationalised using a citation-driven approach. Indeed each metric has its own limitations that need to be considered when selecting an appropriate metric for evaluation. Given the advantages and disadvantages of citation counts, our analysis should be interpreted with caution [5,68]. In academia, it is a common misconception that citation counts provide a benchmark for the impact of research. It should be noted that citation-driven bibliometric analyses neglect to consider the influence of landmark conceptual research journal articles. Further, our analysis does not assess the quality of the research presented. Quality appraisal of the findings was not possible, given the heterogeneity of the resulting output. It is important to consider that citation count fails to represent the quality of the research. Our analysis cannot identify with any authority the key conceptual journal articles that have shaped the trajectory and development of the field. This shortcoming serves as a rationale to support the investigation of conceptually-driven influential psychosocial-oncology research in future. However, it can be noted that previous review articles have reflected on key conceptual developments [1,3]. Finally, the search was limited to the Scopus electronic database. While the inclusion of one electronic database is standard practice in bibliometric analyses [14,16,17,30,31,33], it is important to critique any outcome metric provider. Key to database selection, Scopus provides tools for citation overview including self-citation analysis. However, highly cited articles in journals not indexed in Scopus may not have been captured in the findings. Further, citation count varies between databases [5,68]. For this reason, the ranking of included articles should be interpreted with caution.

Implications for Psycho-Oncology Practice and Research
Given the extensive remit of the multidisciplinary field of oncology, a bibliometric review of the psycho-oncology literature may prove a helpful introduction for multidisciplinary teams working in cancer care. This review offers a broad overview of seminal research in the field. It also honours the key contributors to the field by identifying work that has been frequently cited by other researchers. Clinicians new to the field may perceive psycho-oncology to solely encompass the psychological health of oncology patients. It is important to educate new clinicians to routinely and sensitively consider the individual and systemic level psychological, social, behavioural, and ethical aspects of cancer, since they can substantially influence the outcome of treatment. This review provides health professions with an educational resource that compounds our understanding of the mind-body interaction that continues to challenge a mechanical model of cancer.
This study generates knowledge regarding the intricacies of psycho-oncology clinical practice and research work and emphasises the need for compassionate collaborative, cross-disciplinary cancer care. It is important to acknowledge the need for translation beyond citation into interdisciplinary practice.

Future Directions
This bibliometric review provides a situational analysis of the field of psycho-oncology in the present, as opposed to a view of the future of the field. As discussed previously, it is important to note that our analyses offers a snapshot of highly cited seminal research in psycho-oncology at one point in time. Our analysis is best viewed as a live document responding to the evolving priorities of the field. We recommend replication studies at regular intervals to update the findings in order to maximise educational value. The nature of our review offers a broad scope of the field; future research could consider a more introspective bibliometric analysis of Psycho-Oncology exclusively [69]. Bibliometric reviews of single journals help chart the developmental growth and trajectory of a journal by identifying research trends, areas of research neglect, and disparities in academic publishing. Findings may offer editorial boards insight to help close gaps in research and help support funded external research grant calls [69].
Such an approach would offer increased insight and further support the maturity of the discipline, educational materials, and journal editorial strategy.
Relatively few high-quality RCT studies were included. This deficit identified in our analysis generates greater understanding of one of the pervading gaps in the research field. Our analysis underscores the critical need to enhance the science of psycho-oncology. Greater emphasis on high-quality methodological research is needed. This finding serves as a specific area of research opportunity to greater align future research to the needs of the field.

Conclusions
Psycho-oncology is a vast subspecialty of oncology encompassing diverse areas of clinical practice and research, focusing on the humanism in cancer-prevention, treatment, and aftercare. Given the evolution of the field from a place where the word 'cancer' was stigmatised and the feelings of cancer patients were not acknowledged, a bibliometric review which reflects on almost fifty years since the formal foundation of the field is overdue. This bibliometric review identifies the most frequently cited psycho-oncology journal articles published across all journals listed in Scopus. The results identified in this study are landmark papers that have contributed greatly to the field. This review denotes the growing nature of the discipline, which continues to advance. As the discipline has become increasingly established, there has been a simultaneous increase in research publications. While this study is not without its limitations, it is hoped that identification of seminal research publications will help inform future research contributions. This analysis should serve to support the routine consideration of the psychosocial aspects of cancer care. It may provide a useful educational tool for interdisciplinary clinicians. It is hoped that it will encourage considered compassionate care for cancer patients.

Informed Consent Statement: Not applicable.
Data Availability Statement: Data sharing is not applicable to this article as no new data were created or analysed in this study.

Conflicts of Interest:
The authors declare no conflict of interest.