The Laryngoscope C 2014 The American Laryngological, V
Rhinological and Otological Society, Inc.
A Century of Citation Classics in Otolaryngology–Head and Neck Surgery Journals Revisited Daniel H. Coelho, MD; Luke W. Edelmayer, BS; John E. Fenton, FRCSI, FRCS(ORL-HNS) Objectives/Hypothesis: Citation classics have traditionally been defined in the smaller medical specialties as any article published in a peer-reviewed journal that has received 100 or more citations from other articles also published in peerreviewed journals. This study aimed to determine patterns of citation classics changes in the medical field otorhinolaryngology and head and neck surgery (OHNS) over the past decade and serves as a follow-up to an original study published in 2002, “A Century of Citation Classics in Otolaryngology–Head & Neck Surgery.” Study Design: Bibliometric analysis. Methods: Using the Journal Citation Reports and Web of Science, OHNS journals were selected and assessed for the content of citation classics. Results: Nine-hundred five citation classics were found, over 11-fold more than 1 decade prior. Other significant changes were seen in country of origin, decade of publication, number of authors per article, subspecialty of article, and most frequently discussed topics. Conclusions: The dramatic rise in quantity and nature of citation classics in the past decade may be due to unprecedented advancements in information technology and communication, allowing studies and experiments to be performed, written, reviewed, published, and cited at rapid rates. Key Words: Citation analysis, otorhinolaryngology, bibliometrics, citation classics, publication. Level of Evidence: NA Laryngoscope, 124:1358–1362, 2014
INTRODUCTION The term citation classic was first coined in the early 1970s by the founder and chairman emeritus of Thomson Reuters, Eugene Garfield. Dr. Garfield proposed that one of the purposes of recognizing citation classics was to “provide a kind of living history,” identifying those authors and articles most influential in their respective field.1 Traditionally, a citation classic in the smaller specialties is an article in a peer-reviewed journal that has received 100 or more citations from other peer-reviewed publications. (Of note, the recognition that an article gives to another is considered a reference; the recognition that an article receives from another is considered a citation).2 In 2001, Fenton et al. analyzed otolaryngology–head and neck surgery (OHNS) citation classics from 1900 to 1999, giving particular insight into medical advances and historical developments over the preceding century. However, the past decade has led to From the Department of Otolaryngology–Head and Neck Surgery (D.H.C., L.W.E.), Virginia Commonwealth University, Richmond, Virginia, U.S.A.; and the Department of Otolaryngology–Head and Neck Surgery (J.E.F.), University Hospital Limerick and Graduate Entry Medical School, University of Limerick, Limerick, Ireland. Editor’s Note: This Manuscript was accepted for publication December 24, 2013. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Daniel Coelho, MD, Department of Otolaryngology–Head and Neck Surgery, Virginia Commonwealth University, PO Box 980146, Richmond, VA 23298-0146. E-mail: [email protected] DOI: 10.1002/lary.24573
Laryngoscope 124: June 2014
1358
rapid changes in information technology, and its effects on study reading, writing, publication, and citation have been significant. We hypothesize that the nature of citations has changed markedly in the years since Fenton et al.’s original publication, and as such has implications for the way we as a field read articles, write articles, and ultimately manage patients.
MATERIALS AND METHODS The 2009 Journal Citation Reports (JCR) online database was accessed. Access was obtained from Virginia Commonwealth University library. Thirty-six OHNS journals with impact factors were identified. As in the original study, the multidisciplinary journal Dysphagia was removed from the rankings, as it is chiefly a gastroenterology publication. Name changes of journals and merging of journals were taken into consideration and assessed and consolidated. The Web of Science database was accessed on or before June 11, 2013, and OHNS journals were selected from the year 1965 (or the date of initial journal publication) to the end of 2009. Seven articles that were identified in the previous study and not located in the Web of Science database due to date exclusion (published before 1965) were ultimately located through the Web of Knowledge or through Google Scholar. Both databases were crossreferenced, and duplicate articles were removed. Only articles written in English were included. Again, articles cited 100 or greater times were considered citation classics and ranked in order according to the number of citations received (the citation score). Using the Web of Science database, the citation classics were further identified by country of origin, decade of publication, and subspecialty of the article. The findings were then compared to those of the original 2001 study.
Coelho et al.: Citation Classics Revisited
RESULTS Out of the 36 OHNS journals identified by both JCR and Web of Science, 28 journals published at least one citation classic. In our study, 905 articles were identified as OHNS citation classics, compared with 80 in 2001. The most cited article received 1,277 citations, which in 2001 was also the top-cited article with 408 citations.3,4 There were 143 articles with 200 or more citations, compared with only 10 in 2002. A list of the journals that published citation classics and the number of classics published is shown in Table I. Nearly 70% of all classic articles were published in only five journals. The earliest recorded article with citation classic status was published in 1933, and the most recently
TABLE I. Otolaryngology–Head and Neck Surgery Journals That Published Articles Identified as Classic Citations and the Number of Classic Articles Identified in Each Journal.
Journal
No. of Citation Classics
The Laryngoscope* (United States)
173
Archives of Otolaryngology–Head & Neck Surgery* (United States)
154
Hearing Research (the Netherlands) Annals of Otology Rhinology and Laryngology* (United States) Otolaryngology–Head and Neck Surgery* (United States)
152 90 56
Acta Oto-laryngologica* (United Kingdom)
46
Head & Neck–Journal for the Sciences and Specialties of the Head and Neck* (United States) Journal of Speech and Hearing Disorders (United States)
45 34
Ear and Hearing (United States)
33
International Journal of Audiology* (United Kingdom) European Archives of Oto-Rhino-Laryngology (Germany)
22 16
Audiology and Neurotology (Switzerland)
12
Otology & Neurotology (United States) Journal of Voice (United States)
11 10
Journal of Laryngology and Otology (United Kingdom)
8
Otolaryngologic Clinics of North America (United States) American Journal of Otolaryngology (United States)
7 6
JARO–Journal of the Association for Research in Otolaryngology (United States)
6
Clinical Otolaryngology* (United Kingdom) Journal of Otolaryngology–Head & Neck Surgery (Canada) ORL-Journal for Oto-Rhino-Laryngology and Its Related Specialties (Switzerland)
5 5
Decade
3
Auditory Neuroscience (United States) HNO (Germany)
2 2
Advances in Oto-Rhino-Laryngology (Switzerland)
1
American Journal of Rhinology (United States) Folia Phoniatrica Et Logopaedica (Switzerland)
1 1
Journal of Vestibular Research–Equilibrium & Orientation (the Netherlands)
1
Laryngoscope 124: June 2014
TABLE II. The Decade of Publication of Classic Citations and the Number of Articles Published per Decade.
3
Rhinology (the Netherlands)
Total
published articles were in 2007. In the 2001 study, the most recent citation was from 1993. One hundred thirtyfour articles published in OHNS journals since 1999 have become citation classics, whereas 80 articles published in OHNS journals during the entire 20th century became citation classics. The decade in which citation classics were published and the number represented in each decade are presented in Table II. The year with the highest number of citation classics was 1990 with 46, whereas in the previous analysis it was 10 in 1985. There were 422 authors with two or more citation classic authorships, compared with 24 in the 2001 analysis. There were 166 articles with single authors and 223 articles with two authors, compared with 51 articles in 2001 containing either one or two authors. There were 516 articles with three or more authors. The most authors for one citation classic was 28, compared with 11 in 2001.5 Twenty-two citation classics had 10 or more authors, and there were seven articles with group/committee authorship. The author with the most number of citation classics is MC Liberman with 12. In the current study, the 905 classic articles were divided by subspecialty: 468 (51.7%) otology/lateral skull base surgery, 179 (19.8%) benign head and neck surgery/ laryngology, 163 (18%) head and neck oncology, 95 (10.5%) rhinology/anterior skull base surgery, and no academic otolaryngology articles. In the earlier study, the 80 classic articles as per subspecialty were calculated as 49% otology/lateral skull base surgery, 17.5% benign head and neck surgery/laryngology, 17.5%, rhinology/anterior skull base surgery, 16% head and neck oncology, and again no academic otolaryngology articles (Fig. 1). Table III displays the countries of origin. In the present study, there were 34 different countries represented (counting West Germany, Germany, and Federal Republic of Germany as one country), compared to 10 countries in 2001. The United States had the highest number of citations with 639 articles, and 27 other countries had two or more citations classics. In the prior study, the United States also had the highest number citations at 67, or 84% of the total, and only three other countries had two or more citation classics. Sixty-nine citation classics had two or more countries in which the
905
No. of Citation Classics
1930–1939
3
1940–1949 1950–1959
0 2
1960–1969
31
1970–1979 1980–1989
145 290
1990–1999
300
2000–2009
134
Coelho et al.: Citation Classics Revisited
1359
Fig. 1. Topics most frequently associated with citation classics presented in percentage of total. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
research was conducted, and the most countries collaborating on one citation classic was six (Austria, Belgium, France, Germany, Italy, and Norway).6
DISCUSSION The importance in defining citation classics may have been best explained by Dubin et al. The authors state that the acknowledgement of the most read articles in a certain field not only emphasizes the impact of the work presented by colleagues by distinguishing pivotal advances in the respective field, but fulfills our intrigue into the historical developments of that field.7 In their original study, Fenton and colleagues painted an unique perspective of how our specialty had evolved from 1900 to 1999. However, the data of this current study indicate that rather than medical advances alone, it is writing, publishing, and citation practices that may have had the most dramatic impact in the past 10 years. In comparing 1 decade ago, there were 11-fold more publications cited more than 100 times, an increase of over 1,000%. Articles with 200 or more citations are more than 13 times more cited than in 2001. The most cited article had over three times as many citations compared to the 408 in 2001. The 905 articles originated in 28 journals, compared with eight journals in 2001. The reasons for this are probably multifactorial including the increased number of articles per journal and the overall expansion in the numbers of journals, thus increasing the number of articles being published, referenced, and cited. For example, one leading journal, The Laryngoscope, now publishes an average of 18.3 more articles per issue than in 1993 and 28.2 more articles when comLaryngoscope 124: June 2014
1360
pared to 1933 (Fig. 2). According to the Web of Science, approximately 105,000 OHNS articles were published at the time of the original 2001 study. That number only 10 years later is roughly 144,000, indicating a 37% increase in 1 decade. Such an increase in articles published annually is by no means limited to a single journal and could readily explain the observed geometric increases. The duration from publication to citation classic has also decreased. In 2001, the most recently published classic was 1993, 8 years prior. In the current study, the most recent citation classic was from 2007, 6 years earlier. However, as the initial study was performed 1 year following the study period of 1900 to 1999, and ours 3 years (2000–2009), the significance of this finding is questionable. Perhaps more interestingly, the year with the most citations did not advance 12 years (time interval between initial 2001 and subsequent 2013 analyses) as might be expected. In the original study, 1985 (6 years prior to analysis) had the most classic publications, whereas in the current study it is 1990, 13 years prior to analysis. It would appear that although the number of citations classics has risen dramatically, the amount of time to reach classic status has not changed significantly. This would suggest that as a result of more overall articles published and more on similar themes, there are a greater variety of potential articles to cite. A well-known trend in publication has been the increase in the average number of authors per publication.8 Correspondingly, the number of articles per authorship has increased from about 1.8 to about 3.9. This is likewise observed in the OHNS classic citation literature. For the period of 1900 to 1999, 64% (51/80) of all classic articles had only one or two authors. When Coelho et al.: Citation Classics Revisited
TABLE III. Country of Author Origin and Associated Number of Articles Identified as Citation Classics. Country of Origin
No. of Citation Classics
United States
639
United Kingdom Germany
54 42
Canada
41
Australia Sweden
38 35
Switzerland
22
The Netherlands Japan
20 16
Finland
12
Austria France
11 11
Belgium
8
Denmark Italy
6 6
Israel
5
Scotland Cuba
4 2
Egypt
2
Estonia Hong Kong
2 2
Norway
2
Poland South Korea
2 2
Spain
2
Taiwan Thailand
2 2
Argentina
1
Czechoslovakia Greece
1 1
Hungary
1
Portugal Slovak Republic
1 1
South Africa Total
1 997
reanalyzed in 2013, that number dropped to 43% (389/ 905). The number of authors per article has increased from about 1.8 to about 3.7 in the period studied. Certainly, increased multidisciplinary and multicenter collaboration could explain this trend. Likewise, less stringent requirements for inclusion for authorship may be responsible, despite the trend of journals to require documentation of authorship roles. Perhaps one of the strongest indications of how our field has changed comes when analyzing the subspecialty breakdown. As might be expected, the overall distribution of citation classics among fields within OHNS has not changed significantly. There was a slight decrease in the percentage of rhinology/anterior skull base articles, though the reasons for this are unclear. Laryngoscope 124: June 2014
As in the original study, that vast majority of publications come from United States authors and institutions, though the percentage has dropped slightly. Certainly, representation from other countries, some considered “developing nations,” is encouraging. However, this reanalysis again identifies a common practice in the world of journal publication, wherein there is bias toward local journals and articles among US and UK health professionals.9 Reviewers from the United States and outside the United States have a preference for and favor articles submitted by US authors.10 The prior study noted a trend for non-US or UK authors to submit to and publish in US and UK journals, potentially leading to the relegation of European journals to “third rate.” With the inclusion of several European journals newly publishing citations classics, our data would suggest a possible reversal of this trend, or possibly more Europeans are submitting articles to their more local journals. Though citation analysis is frequently used to establish citation classics, one must always remember that this bibliometric system is also a sociometric method, and thus it is in no way a predictor of the significance of an article. The significance of an article can only be measured by evaluation through time and assessment by peers to give any meaning or value to the individual’s or journal’s contribution to that field.6 Many important articles to our specialty are not included in this list, as the presented data may have been deemed incontrovertible and therefore absorbed into established knowledge. Certainly, other limitations to this type of analysis exist as citation analysis itself is inherently biased. These studies were performed for English-language articles only, and English-speaking authors have a tendency to read, and therefore cite, English-language publications. Authors tend to cite themselves, potentially dramatically overinflating citations for more prolific authors. Moreover, authors likely tend to cite those articles that everyone else is citing, resulting in a self-sustaining cycle irrespective of scientific merit. Textbooks and monographs were not included, and although often not a source of original material, they are frequently cited. This study also did not include OHNS subject articles in non-OHNS journals, a potential source of numerous highly cited sources. As mentioned earlier, the duration of time between study periods and analysis (1 year for the original study and 3 years for the current study) may influence the outcomes. So although the trends are certainly worthy of attention, no serious weight should be given to the absolute numbers provided. Ultimately, beyond the intellectual value these data provide, the merit of this article does not come from merely learning how we as a specialty write, cite, reference, and publish. Rather, the pragmatic application has greatest influence on how we read articles and apply knowledge. As with reading any article, a critical eye is essential. For all the reasons listed above, the reader should be especially cautious not to blindly accept the findings of a citation classic as gospel. Rather, readers are encouraged to look at the reference list of the mostcited articles.11 The explosion of publications and citations prompts the reexamination of the term citation classics. These Coelho et al.: Citation Classics Revisited
1361
Fig. 2. Average number of articles per issue in The Laryngoscope, the journal with the most citation classics. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
data suggest that a benchmark of 100 citations may not be indicative of the most read and/or influential articles, and top 100 cited articles had at least 221 citations at the time this study was conducted. This represents
Rhinological and Otological Society, Inc.
A Century of Citation Classics in Otolaryngology–Head and Neck Surgery Journals Revisited Daniel H. Coelho, MD; Luke W. Edelmayer, BS; John E. Fenton, FRCSI, FRCS(ORL-HNS) Objectives/Hypothesis: Citation classics have traditionally been defined in the smaller medical specialties as any article published in a peer-reviewed journal that has received 100 or more citations from other articles also published in peerreviewed journals. This study aimed to determine patterns of citation classics changes in the medical field otorhinolaryngology and head and neck surgery (OHNS) over the past decade and serves as a follow-up to an original study published in 2002, “A Century of Citation Classics in Otolaryngology–Head & Neck Surgery.” Study Design: Bibliometric analysis. Methods: Using the Journal Citation Reports and Web of Science, OHNS journals were selected and assessed for the content of citation classics. Results: Nine-hundred five citation classics were found, over 11-fold more than 1 decade prior. Other significant changes were seen in country of origin, decade of publication, number of authors per article, subspecialty of article, and most frequently discussed topics. Conclusions: The dramatic rise in quantity and nature of citation classics in the past decade may be due to unprecedented advancements in information technology and communication, allowing studies and experiments to be performed, written, reviewed, published, and cited at rapid rates. Key Words: Citation analysis, otorhinolaryngology, bibliometrics, citation classics, publication. Level of Evidence: NA Laryngoscope, 124:1358–1362, 2014
INTRODUCTION The term citation classic was first coined in the early 1970s by the founder and chairman emeritus of Thomson Reuters, Eugene Garfield. Dr. Garfield proposed that one of the purposes of recognizing citation classics was to “provide a kind of living history,” identifying those authors and articles most influential in their respective field.1 Traditionally, a citation classic in the smaller specialties is an article in a peer-reviewed journal that has received 100 or more citations from other peer-reviewed publications. (Of note, the recognition that an article gives to another is considered a reference; the recognition that an article receives from another is considered a citation).2 In 2001, Fenton et al. analyzed otolaryngology–head and neck surgery (OHNS) citation classics from 1900 to 1999, giving particular insight into medical advances and historical developments over the preceding century. However, the past decade has led to From the Department of Otolaryngology–Head and Neck Surgery (D.H.C., L.W.E.), Virginia Commonwealth University, Richmond, Virginia, U.S.A.; and the Department of Otolaryngology–Head and Neck Surgery (J.E.F.), University Hospital Limerick and Graduate Entry Medical School, University of Limerick, Limerick, Ireland. Editor’s Note: This Manuscript was accepted for publication December 24, 2013. The authors have no funding, financial relationships, or conflicts of interest to disclose. Send correspondence to Daniel Coelho, MD, Department of Otolaryngology–Head and Neck Surgery, Virginia Commonwealth University, PO Box 980146, Richmond, VA 23298-0146. E-mail: [email protected] DOI: 10.1002/lary.24573
Laryngoscope 124: June 2014
1358
rapid changes in information technology, and its effects on study reading, writing, publication, and citation have been significant. We hypothesize that the nature of citations has changed markedly in the years since Fenton et al.’s original publication, and as such has implications for the way we as a field read articles, write articles, and ultimately manage patients.
MATERIALS AND METHODS The 2009 Journal Citation Reports (JCR) online database was accessed. Access was obtained from Virginia Commonwealth University library. Thirty-six OHNS journals with impact factors were identified. As in the original study, the multidisciplinary journal Dysphagia was removed from the rankings, as it is chiefly a gastroenterology publication. Name changes of journals and merging of journals were taken into consideration and assessed and consolidated. The Web of Science database was accessed on or before June 11, 2013, and OHNS journals were selected from the year 1965 (or the date of initial journal publication) to the end of 2009. Seven articles that were identified in the previous study and not located in the Web of Science database due to date exclusion (published before 1965) were ultimately located through the Web of Knowledge or through Google Scholar. Both databases were crossreferenced, and duplicate articles were removed. Only articles written in English were included. Again, articles cited 100 or greater times were considered citation classics and ranked in order according to the number of citations received (the citation score). Using the Web of Science database, the citation classics were further identified by country of origin, decade of publication, and subspecialty of the article. The findings were then compared to those of the original 2001 study.
Coelho et al.: Citation Classics Revisited
RESULTS Out of the 36 OHNS journals identified by both JCR and Web of Science, 28 journals published at least one citation classic. In our study, 905 articles were identified as OHNS citation classics, compared with 80 in 2001. The most cited article received 1,277 citations, which in 2001 was also the top-cited article with 408 citations.3,4 There were 143 articles with 200 or more citations, compared with only 10 in 2002. A list of the journals that published citation classics and the number of classics published is shown in Table I. Nearly 70% of all classic articles were published in only five journals. The earliest recorded article with citation classic status was published in 1933, and the most recently
TABLE I. Otolaryngology–Head and Neck Surgery Journals That Published Articles Identified as Classic Citations and the Number of Classic Articles Identified in Each Journal.
Journal
No. of Citation Classics
The Laryngoscope* (United States)
173
Archives of Otolaryngology–Head & Neck Surgery* (United States)
154
Hearing Research (the Netherlands) Annals of Otology Rhinology and Laryngology* (United States) Otolaryngology–Head and Neck Surgery* (United States)
152 90 56
Acta Oto-laryngologica* (United Kingdom)
46
Head & Neck–Journal for the Sciences and Specialties of the Head and Neck* (United States) Journal of Speech and Hearing Disorders (United States)
45 34
Ear and Hearing (United States)
33
International Journal of Audiology* (United Kingdom) European Archives of Oto-Rhino-Laryngology (Germany)
22 16
Audiology and Neurotology (Switzerland)
12
Otology & Neurotology (United States) Journal of Voice (United States)
11 10
Journal of Laryngology and Otology (United Kingdom)
8
Otolaryngologic Clinics of North America (United States) American Journal of Otolaryngology (United States)
7 6
JARO–Journal of the Association for Research in Otolaryngology (United States)
6
Clinical Otolaryngology* (United Kingdom) Journal of Otolaryngology–Head & Neck Surgery (Canada) ORL-Journal for Oto-Rhino-Laryngology and Its Related Specialties (Switzerland)
5 5
Decade
3
Auditory Neuroscience (United States) HNO (Germany)
2 2
Advances in Oto-Rhino-Laryngology (Switzerland)
1
American Journal of Rhinology (United States) Folia Phoniatrica Et Logopaedica (Switzerland)
1 1
Journal of Vestibular Research–Equilibrium & Orientation (the Netherlands)
1
Laryngoscope 124: June 2014
TABLE II. The Decade of Publication of Classic Citations and the Number of Articles Published per Decade.
3
Rhinology (the Netherlands)
Total
published articles were in 2007. In the 2001 study, the most recent citation was from 1993. One hundred thirtyfour articles published in OHNS journals since 1999 have become citation classics, whereas 80 articles published in OHNS journals during the entire 20th century became citation classics. The decade in which citation classics were published and the number represented in each decade are presented in Table II. The year with the highest number of citation classics was 1990 with 46, whereas in the previous analysis it was 10 in 1985. There were 422 authors with two or more citation classic authorships, compared with 24 in the 2001 analysis. There were 166 articles with single authors and 223 articles with two authors, compared with 51 articles in 2001 containing either one or two authors. There were 516 articles with three or more authors. The most authors for one citation classic was 28, compared with 11 in 2001.5 Twenty-two citation classics had 10 or more authors, and there were seven articles with group/committee authorship. The author with the most number of citation classics is MC Liberman with 12. In the current study, the 905 classic articles were divided by subspecialty: 468 (51.7%) otology/lateral skull base surgery, 179 (19.8%) benign head and neck surgery/ laryngology, 163 (18%) head and neck oncology, 95 (10.5%) rhinology/anterior skull base surgery, and no academic otolaryngology articles. In the earlier study, the 80 classic articles as per subspecialty were calculated as 49% otology/lateral skull base surgery, 17.5% benign head and neck surgery/laryngology, 17.5%, rhinology/anterior skull base surgery, 16% head and neck oncology, and again no academic otolaryngology articles (Fig. 1). Table III displays the countries of origin. In the present study, there were 34 different countries represented (counting West Germany, Germany, and Federal Republic of Germany as one country), compared to 10 countries in 2001. The United States had the highest number of citations with 639 articles, and 27 other countries had two or more citations classics. In the prior study, the United States also had the highest number citations at 67, or 84% of the total, and only three other countries had two or more citation classics. Sixty-nine citation classics had two or more countries in which the
905
No. of Citation Classics
1930–1939
3
1940–1949 1950–1959
0 2
1960–1969
31
1970–1979 1980–1989
145 290
1990–1999
300
2000–2009
134
Coelho et al.: Citation Classics Revisited
1359
Fig. 1. Topics most frequently associated with citation classics presented in percentage of total. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
research was conducted, and the most countries collaborating on one citation classic was six (Austria, Belgium, France, Germany, Italy, and Norway).6
DISCUSSION The importance in defining citation classics may have been best explained by Dubin et al. The authors state that the acknowledgement of the most read articles in a certain field not only emphasizes the impact of the work presented by colleagues by distinguishing pivotal advances in the respective field, but fulfills our intrigue into the historical developments of that field.7 In their original study, Fenton and colleagues painted an unique perspective of how our specialty had evolved from 1900 to 1999. However, the data of this current study indicate that rather than medical advances alone, it is writing, publishing, and citation practices that may have had the most dramatic impact in the past 10 years. In comparing 1 decade ago, there were 11-fold more publications cited more than 100 times, an increase of over 1,000%. Articles with 200 or more citations are more than 13 times more cited than in 2001. The most cited article had over three times as many citations compared to the 408 in 2001. The 905 articles originated in 28 journals, compared with eight journals in 2001. The reasons for this are probably multifactorial including the increased number of articles per journal and the overall expansion in the numbers of journals, thus increasing the number of articles being published, referenced, and cited. For example, one leading journal, The Laryngoscope, now publishes an average of 18.3 more articles per issue than in 1993 and 28.2 more articles when comLaryngoscope 124: June 2014
1360
pared to 1933 (Fig. 2). According to the Web of Science, approximately 105,000 OHNS articles were published at the time of the original 2001 study. That number only 10 years later is roughly 144,000, indicating a 37% increase in 1 decade. Such an increase in articles published annually is by no means limited to a single journal and could readily explain the observed geometric increases. The duration from publication to citation classic has also decreased. In 2001, the most recently published classic was 1993, 8 years prior. In the current study, the most recent citation classic was from 2007, 6 years earlier. However, as the initial study was performed 1 year following the study period of 1900 to 1999, and ours 3 years (2000–2009), the significance of this finding is questionable. Perhaps more interestingly, the year with the most citations did not advance 12 years (time interval between initial 2001 and subsequent 2013 analyses) as might be expected. In the original study, 1985 (6 years prior to analysis) had the most classic publications, whereas in the current study it is 1990, 13 years prior to analysis. It would appear that although the number of citations classics has risen dramatically, the amount of time to reach classic status has not changed significantly. This would suggest that as a result of more overall articles published and more on similar themes, there are a greater variety of potential articles to cite. A well-known trend in publication has been the increase in the average number of authors per publication.8 Correspondingly, the number of articles per authorship has increased from about 1.8 to about 3.9. This is likewise observed in the OHNS classic citation literature. For the period of 1900 to 1999, 64% (51/80) of all classic articles had only one or two authors. When Coelho et al.: Citation Classics Revisited
TABLE III. Country of Author Origin and Associated Number of Articles Identified as Citation Classics. Country of Origin
No. of Citation Classics
United States
639
United Kingdom Germany
54 42
Canada
41
Australia Sweden
38 35
Switzerland
22
The Netherlands Japan
20 16
Finland
12
Austria France
11 11
Belgium
8
Denmark Italy
6 6
Israel
5
Scotland Cuba
4 2
Egypt
2
Estonia Hong Kong
2 2
Norway
2
Poland South Korea
2 2
Spain
2
Taiwan Thailand
2 2
Argentina
1
Czechoslovakia Greece
1 1
Hungary
1
Portugal Slovak Republic
1 1
South Africa Total
1 997
reanalyzed in 2013, that number dropped to 43% (389/ 905). The number of authors per article has increased from about 1.8 to about 3.7 in the period studied. Certainly, increased multidisciplinary and multicenter collaboration could explain this trend. Likewise, less stringent requirements for inclusion for authorship may be responsible, despite the trend of journals to require documentation of authorship roles. Perhaps one of the strongest indications of how our field has changed comes when analyzing the subspecialty breakdown. As might be expected, the overall distribution of citation classics among fields within OHNS has not changed significantly. There was a slight decrease in the percentage of rhinology/anterior skull base articles, though the reasons for this are unclear. Laryngoscope 124: June 2014
As in the original study, that vast majority of publications come from United States authors and institutions, though the percentage has dropped slightly. Certainly, representation from other countries, some considered “developing nations,” is encouraging. However, this reanalysis again identifies a common practice in the world of journal publication, wherein there is bias toward local journals and articles among US and UK health professionals.9 Reviewers from the United States and outside the United States have a preference for and favor articles submitted by US authors.10 The prior study noted a trend for non-US or UK authors to submit to and publish in US and UK journals, potentially leading to the relegation of European journals to “third rate.” With the inclusion of several European journals newly publishing citations classics, our data would suggest a possible reversal of this trend, or possibly more Europeans are submitting articles to their more local journals. Though citation analysis is frequently used to establish citation classics, one must always remember that this bibliometric system is also a sociometric method, and thus it is in no way a predictor of the significance of an article. The significance of an article can only be measured by evaluation through time and assessment by peers to give any meaning or value to the individual’s or journal’s contribution to that field.6 Many important articles to our specialty are not included in this list, as the presented data may have been deemed incontrovertible and therefore absorbed into established knowledge. Certainly, other limitations to this type of analysis exist as citation analysis itself is inherently biased. These studies were performed for English-language articles only, and English-speaking authors have a tendency to read, and therefore cite, English-language publications. Authors tend to cite themselves, potentially dramatically overinflating citations for more prolific authors. Moreover, authors likely tend to cite those articles that everyone else is citing, resulting in a self-sustaining cycle irrespective of scientific merit. Textbooks and monographs were not included, and although often not a source of original material, they are frequently cited. This study also did not include OHNS subject articles in non-OHNS journals, a potential source of numerous highly cited sources. As mentioned earlier, the duration of time between study periods and analysis (1 year for the original study and 3 years for the current study) may influence the outcomes. So although the trends are certainly worthy of attention, no serious weight should be given to the absolute numbers provided. Ultimately, beyond the intellectual value these data provide, the merit of this article does not come from merely learning how we as a specialty write, cite, reference, and publish. Rather, the pragmatic application has greatest influence on how we read articles and apply knowledge. As with reading any article, a critical eye is essential. For all the reasons listed above, the reader should be especially cautious not to blindly accept the findings of a citation classic as gospel. Rather, readers are encouraged to look at the reference list of the mostcited articles.11 The explosion of publications and citations prompts the reexamination of the term citation classics. These Coelho et al.: Citation Classics Revisited
1361
Fig. 2. Average number of articles per issue in The Laryngoscope, the journal with the most citation classics. [Color figure can be viewed in the online issue, which is available at wileyonlinelibrary.com.]
data suggest that a benchmark of 100 citations may not be indicative of the most read and/or influential articles, and top 100 cited articles had at least 221 citations at the time this study was conducted. This represents
