Arthroplasty and global research output: A bibliometric analysis

Abstract

Introduction

Peer-reviewed research helps to advance many aspects of medical and surgical practice. This paper determines the main contributors tos joint arthroplasty research in terms of quantity and quality.

Methods

A search of the Web of Science™ platform was conducted to identify arthroplasty articles published between 2001 and 2016. The number of articles and citations per article were analysed to assess the quantity and quality of research from individual countries. Results were standardised according to the country's populations and Gross Domestic Product (GDP).

Results

In total, 43,470 arthroplasty articles were published worldwide from January 2001 through December 2016. There was a 4.5-fold increase in global output during this time period. Twenty-two countries contributed at least 1% to the total number of publications. The United States of America published the most articles (35.40%), followed by England (10.31%) and Germany (10.03%). The USA had the highest absolute number of citations (50,777). Denmark had the highest average citation per item (8.76). When number of articles was normalized to population, Switzerland ranked the highest. When adjusted by GDP, Scotland ranked highest. When standardised according to GDP per capita, the People's Republic of China rated highest. The Journal of Arthroplasty produced the highest number of publications related to arthroplasty with 10.9% of total volume.

Conclusion

There has been a substantial increase in worldwide publications relating to arthroplasty. The USA has produced the largest volume whilst Denmark has produced the highest quality publications. When output was normalized according to population and GDP, Switzerland and Scotland ranked highest.

1. Introduction

Total joint arthroplasty (TJA) is one of the most cost-effective and successful surgical interventions. There are more than 1 million TJAs performed in the United States annually [Williams S.N., Wolford M.L., Bercovitz A. 2000. Hospitalization for total knee replacement among inpatients aged 45 and over: United States, 2000–2010 key findings.], [Wolford M.L., Palso K., Bercovitz A. 2000. Hospitalization for total hip replacement among inpatients aged 45 and over: United States, 2000–2010 key findings]. Total Joint Arthroplasty accounts for more Medicare spending than any other inpatient medical procedure, amounting to $5 billion of the Medicare budget in 2006. [Wilson NA, Schneller ES, Montgomery K, Bozic KJ. Hip and knee implants: current trends and policy considerations. Health Aff (Millwood) 2008; 27:1587–98. https://doi.org/10.1377/hlthaff.27.6.1587]

Demand for TJA is expected to increase substantially as the population of the United States ages. Projections indicate the number of TJA will rise to over 4 million per year by 2030 [Kurtz S., Ong K., Lau E., Mowat F., Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am. 2007; 89 (4):780]. This will raise Medicare's TJA spending to an estimated $50 billion annually [Wilson NA, Schneller ES, Montgomery K, Bozic KJ. Hip and knee implants: current trends and policy considerations. Health Aff (Millwood) 2008; 27:1587–98. https://doi.org/10.1377/hlthaff.27.6.1587]

Bibliometrics is a viable means to quantitatively and qualitatively evaluate trends in research activity over time using literature databases and literature metrology characteristics. It provides a way to grasp development in a certain field and evaluate academic groups and individual researchers.( Ekinci S, Agilli M, Ersen O et al. Letter to the editor regarding analysis of changing paradigms of management in 179 patients with spinal tuberculosis during a 12-Year period and proposal of a new management algorithm. World Neurosurg 2015; 84:2072–72.) Bibliometric studies can also provide supporting evidence for policy and decision making(Avcu G, Sahbudak Bal Z, Duyu M et al. Thanks to trauma: a delayed diagnosis of Pott disease. Pediatr Emerg Care 2015; 31:E17–E18.).

Ranking of countries according to their scientific literature contribution has been performed for areas of medicine including, anesthesia and critical care,^1,2 ophthalmology,³ emergency medicine,⁴ dermatology,⁵ endocrinology,⁶ rheumatology⁷ and oncological surgery.⁸ There have been a few publications investigating the worldwide research output in orthopaedic surgery. These include hand and wrist, foot and ankle, arthroscopy and spinal surgery.9, 10, 11, 12, 13, 14 There is however a gap in the literature pertaining to the international research output in TJA. The purpose of this paper is to determine which countries are the main contributors to joint arthroplasty research in terms of quantity and quality.

2. Materials and methods

A search of the Web of Science™ platform was conducted to identify arthroplasty articles. The search was performed on the 29th of April 2017. Search criteria included “Arthroplast*” in the topic field, over the time period “2001 to 2016”. Only original articles and reviews were included; editorials and article corrections were excluded. Publications in all languages were included. There were no other filters applied. Each publication was assigned a country of origin. This is indexed by the Web of Science™ using the corresponding authors address. The total number of articles published was used to represent quantity. The number of citations per article (citation index) was used as a measure of quality. The total number of publications and the citation index per country were collated.

Gross Domestic Product, GDP per capita and population size was sourced from the Work Bank.¹⁵ The results of the above search were then standardised according to the country's respective populations, GDP and GDP per capita. Only those countries contributing at least 1% of the total number of articles were included in this analysis. .

Output index was calculated to measure the relative change in research output over time. Calculations were based on each country's output in 2016 compared to 2001. The global output index was also calculated for the same time period. Journals were ranked according to volume of arthroplasty articles published over the given period.

3. Results

A total of 43,470 arthroplasty articles were published worldwide from 2001 to 2016. The United States published the most articles (15,384) accounting for 35.4% of the global output. The USA was followed by England (4,480), Germany (4,358), Canada (2,520), and Japan (2,236) (Fig. 1). Publications from the USA contributed the highest total number of citations with 50,777.

Fig. 1.

List of Countries according to % of total number of publications from 2001 to 2016, top ten countries. Total number of publications = 43,470.

Twenty-two countries contributed at least 1% to global output from 2001 to 2016. Of these 22 countries, Denmark had the highest average citation per item (8.76), followed by Canada (8.1) and the Netherlands (7.86) (Fig. 2).

Fig. 2.

List of Countries according to number of citations per publication from 2001 to 2016, top ten countries.

When normalized to population, Switzerland ranked the highest, followed by Denmark and Scotland (Fig. 3A). When adjusted by GDP, Scotland ranked first, followed by Switzerland and England (Fig. 3B). When standardised according to GDP per capita, People's Republic of China, followed by India followed by the USA ranked highest (Fig. 3C).

Fig. 3.

List of Countries according to number of publications A. per population and B. per GDP and C. per GDP per Capita, respectively, for the top ten countries of those countries publishing at least 1% of the global output from 2001 to 2016.

Ninety-four percent (41,045) of articles were published in the English language. The next most common language for publication was German (3.8%) followed by French (0.93%).

Sixteen journals published at least 1% of the total output. The ‘Journal of Arthroplasty’ was the most prolific with 4,719 articles accounting for 10.9%. This was followed by “Clinical Orthopaedics and Related Research’ who published 2,698 (6.2%) followed by ‘Journal of Bone and Joint Surgery (Am) 1,751 (4%) (Fig. 4).

Fig. 4.

List of Journals according to % of total number of publications from 2001 to 2016, top ten journals. Total number of publications = 43,470.

The global output index for arthroplasty publications was 4.55 when comparing 2016 with 2001. Countries with the greatest relative increase in research productivity are listed in Fig. 5.

Fig. 5.

Output index over time for the Top 5 countries producing output over 2001–2016. Output index equal to the number of journals published in 2016 divided by the number of journals published in 2001 by the respective countries.

Fig. 1. Countries ranked according to percentage of global arthroplasty research output.

Fig. 2. Countries ranked according to number of citations per publication.

Fig. 3. Countries ranked according to number of publications adjusted for A. population B. GDP and C. GDP per Capita.

Fig. 4. Journals ranked according to total number of arthroplasty publications 2001–2016.

Fig. 5. Output index showing the relative change in research output over time. Output index is equal to the number publications in 2016 divided by the number of publications in 2001.

4. Discussion

It is interesting and important to identify the trends in research productivity for sub-specialties in terms of medical advancement and appropriate resource management. This is the first publication identifying the global productivity in the vast, expanding area of joint arthroplasty. The USA emerged as the leader both in terms of absolute number of publications and absolute number of citations over the time-frame studied. However when the number of publications was standardised to parameters such as GDP and population size, other countries became leaders in the analysis. This paper reveals there has been a 4.55-fold increase in research productivity globally from 2001 to 2016. The increase in research output has been seen in other subspecialties of orthopaedics also but to a lesser extent.^9,10,12

In terms of quantity, i.e. number of publications, the USA were the leaders, publishing over one third of the global output over the time-frame studied here. In fact in absolute numbers, the USA is the most productive country in the majority of studies.⁸ This may be partially attributed to its large population size and English being its first language.⁸ These factors would all likely contribute to this prolific success.⁸ In terms of quality, i.e. citations per publication item, Denmark ranked highest, suggesting that Denmark has the highest quality research publications. A high citation per item rating could, of course, be due to an individual paper or small number of publications receiving a very high citation index, thus skewing the results. Self-citation was not investigated in this study. Self-citation could influence the citation index and magnify the importance or popularity of the publications.¹⁶ Further in-depth analysis of the individual publications contributing to the country's overall citation rating would be required to further interpret this quality ranking.

We will take a closer look here at two studies investigating the global distribution of research output in other orthopaedic subspecialties and discuss our findings in context of their results. Luo et al. determined the research output in the field of arthroscopy and found that the United States published the most articles and reviews (35.40%), followed by Germany (9.53%), the United Kingdom (6.80%), the Republic of Korea (5.45%), and Japan (4.76%).¹² We found a similar ranking where The United States published the most articles and reviews (35.40%), followed by England (10.31%), Germany (10.03%), the Canada (5.80%), and Japan (5.14%) from 2001 to 2016. Also, we investigated England, Scotland and Wales separately as that is how the Web of Science™ was set by default. It would be useful to have incorporated the three counties plus Northern Ireland as the UK to allow for comparison with other studies. A further reason for discrepancy between our results may be due to the fact that we reviewed a different time interval (2001–2016) and Luo et al. reviewed publications released between 1999 and 2013. Differences in our results may also be due to the fact that the two studies were pertaining to different subspecialties. Luo et al. found Sweden to have the highest mean citation whereas in our study, Denmark ranked first in this category.¹⁰ Luo et al. found that when normalized to population, Switzerland ranked the highest, whereas we found in the field of arthroplasty, Scotland ranked first, followed by Switzerland and England.¹⁰ When standardised according to GDP per capita, we found People's Republic of China, followed by India followed by the USA. Luo's study did not rank output according to this metric and this is a novel feature of this study. However, it should be noted that ratings from our study only included those countries publishing at least 1% of the total global number of publications and for a complete assessment of research output according to GDP per capita, we should include all countries in this analysis.

A similar bibliometric analysis investigating the international research output in the area of hand and wrist literature was performed by Mei et al., in 2016.⁹ This study largely agreed with both our study and the study by Luo et al. above. Mei's methodology was different in that they selected only research that was published in the top four highest cited journals from 2005 to 2014, whereas we considered all journals, and over the time period between 2001 and 2016. A total of 4,268 publications were identified in their study, compared with 43,470 studies in the field of arthroplasty. This may be partly explained by the authors limiting their study to the top four cited journals only or due to other factors such as private industry interests in arthroplasty and varying techniques and advancements in arthroplasty. The number of articles and reviews showed a significant increase of 2.10-fold between 2005 and 2014 (p = 0.0001). We found a greater increase of 4.55-fold between 2001 and 2016. The authors found that the United States published the most articles and reviews (53.89%), followed by United Kingdom (6.51%), Japan (6.14%), Canada (3.70%), and China (3.37%). The authors looked at 5-year impact factor, which is a metric we did not employ and thus could not use for comparison. When normalized to population, the United States ranked the first (7.16), followed by Sweden (6.53), and Netherlands (5.72) in hand and wrist research output. When normalized to population in our study, Switzerland ranked the highest, followed by Denmark and Scotland. Mei et al. found that Sweden had the highest average citations per article/review (11.38), followed by Germany (9.63), and Australia (9.08).⁹ These results were different to ours in that Denmark had the highest average citation per item (8.76) followed by Canada (8.1) and the Netherlands (7.86).

We examined the global output in terms of journals that were publishing the most number of articles and reviews in arthroplasty. The results inform us that the ‘Journal of Arthroplasty’ published the most number of articles/reviews, (10.9%) of the followed by the journal “Clinical Orthopaedics and Related Research’ (6.2%) ‘Journal of Bone and Joint Surgery - American Volume’ (4%), see Fig. 3. We also reviewed the trends in output for the top five countries to compare the rates at which the research output was increasing, (Fig. 4). It would be interesting to generate an output index for as a marker for trends in research productivity for all countries worldwide. From this we could estimate the relative growth in joint arthroplasty research for each country. This proposed study could provide a platform for a cause and effect analysis of the trends in international productivity with respect to research developments and funding.

From this study we can report that there is a significant increase in research output over the past fifteen years. This increase was most marked in the area of arthroplasty when compared with published bibliometric analysis in other orthopaedic specialties. The USA had the greatest number of publications and citations and this was seen in other orthopaedic subspecialties. However, when adjusted by citations per article/review, population, GDP, or GDP per capita, other countries such as Scotland and Switzerland came to the fore. Denmark were seen to produce the highest quality of research. The Journal of Arthroplasty has published the most number of articles and reviews, over 10% of the total number of articles and reviews from 2001 to 2016.

Discrepancies between our study and other orthopaedic subspecialties that were observed might be explained by differing centres within countries focussing on different subspecialties or the fact that there can exist a different prevalence and impact of different orthopaedic conditions in individual countries. As stated earlier, a high citation per item rating could be due to an individual paper or small number of publications receiving a very high citation index, or high levels of self-citation, thus skewing the results. Further in-depth analysis of the individual publications contributing to the country's overall citation rating would be required to further interpret ranking using this parameter. A further strength of our study is that it included studies published in all languages, and not only English. A potential limitation of this study may be that articles/reviews were not identified under the “orthopaedic” category specifically; our analysis may have included publications only loosely relevant to the subspecialty of arthroplasty and resulting analysis may not reflect the output of arthroplasty exactly. However, on balance, this approach to article/review selection was thought to be more likely to include all publications relevant to arthroplasty. Finally, the results obtained can only be as accurate as the record keeping accuracy of the database and journals and articles within. For an accurate picture of relevant literature, one would need to review the 47,000 publications or at least associated abstracts to assess their relevance. Future studies may include application of further filters on the database searches, for example selecting the subsection of orthopaedics only.

5. Conclusion

There has been a substantial increase in output in research output in arthroplasty, more marked than has been seen in other orthopaedic subspecialties. The USA had the largest output in terms of quantity. Denmark has the highest quality research publications. When output was normalized according to population and GDP, Switzerland and Scotland respectively scored highest. Ireland has had a significant increase in research output over the past sixteen years. This bibliometric analysis should be of interest to all healthcare and educational institutions, industries and government bodies involved in the ongoing advances and practices in the vast area of arthroplasty and orthopaedics.

Source of funding

None.

Conflicts of interest

None.