American Journal of Emergency Medicine (2008) 26, 5 – 9

Original Contribution

Authorship, collaboration, and predictors of extramural funding in the emergency medicine literature

Jaime S. Rosenzweig MD, Shawn K. Van Deusen MD, Okemefuna Okpara MD, Paris A. Datillo RN, William M. Briggs PhD, Robert H. Birkhahn MD*

Department of Emergency Medicine, New York Methodist Hospital, Brooklyn, NY 11215, USA

Received 22 September 2006; revised 8 February 2007; accepted 9 February 2007

Abstract

Objectives: The objectives of the study were to examine the last decade of general emergency medicine (EM) literature published in the United States for trends with regard to authorship and multidisciplinary collaboration and to estimate the effect on extramural funding.

Methods: Print articles published in the Academic Emergency Medicine, Annals of Emergency Medicine, Journal of Emergency Medicine, and American Journal of Emergency Medicine between 1994 and 2003 were reviewed. original research, case reports/series, and others (consensus/educational) were considered; abstracts, book reviews, and editorials were not. The author byline was reviewed for number, specialty, nationality, collaboration, and presence of extramural funding. Multidisciplinary collaboration was defined as authors from 2 or more specialties, whereas multi-institutional collaboration was defined as EM authors from more than one institution. Logistic regression was used to identify predictors of extramural funding from the variables collected.

Results: Of 5728 articles identified, there were 3278 (57%) original research, 1437 (25%) case reports/ series, and 975 (17%) classified as others. The percentage funded was 22% for all articles (32% for original research). The literature had at least one EM investigator as coauthor 84% of the time. Article location of origin was the United States (63%), foreign (15%), and combined (22%). Multidisciplinary collaboration increased overall from 33% in 1994 to a high of 43% in 2003. Multi-institutional collaboration also increased from 16% in 1994 to 26% in 2003. The percentage of articles having 6 or more authors increased from 12% to 18% over the decade. Of all variables studied, only article type (original research: odds ratio, 4.8; 95% confidence interval, 4.0-5.6) and foreign source (non-United States: odds ratio, 1.3; 95% confidence interval, 1.1-1.5) predicted extramural funding.

Conclusions: The number of authors per article in the EM literature has steadily increased over the last decade, as has evidence of collaboration with other specialties. This increase in collaboration and author number has not been associated with increased extramural funding in the general EM literature published in the United States.

D 2008

* Corresponding author. Tel.: +1 718 780 5040; fax: +1 718 780 3153.

E-mail address: [email protected] (R.H. Birkhahn).

Introduction

Science scarcely exists until it is published [1]. Scientific publication is both a mechanism for communicating new

0735-6757/$ – see front matter D 2008 doi:10.1016/j.ajem.2007.02.028

Fig. 1 Classification of articles appearing in the general EM literature published in the United States.

results and a contribution to a priceless repository of data that later may be retrieved and used in new ways that had not been considered at the time of the original publication [2]. Scientific research is essential for the development of any Medical specialty. It is especially true for emergency medicine (EM), which is a relatively new specialty. Research should not be considered something separate or expendable from the specialty; instead, it should be recognized as essential to the advancement of clinical EM [3].

In recent years, there has been an increase in Scientific publications on the whole. This increase has been especially well represented in the EM literature. With this increase in publication, there has also been a trend toward increased authorship and multidisciplinary collaboration. Has the EM literature in the past decade followed the trend toward increasing numbers of authors per article and increased collaboration with other disciplines? If so, what has the effect been on extramural funding? We sought to answer these questions by examining the 4 US peer-reviewed general EM journals published between 1994 and 2003. The author byline was reviewed for number, specialty, nation-

ality, and collaboration to detect trends of these variables and the effect on extramural funding in the literature.

Methods

Study design

This study involved a retrospective review of the 4 US peer-reviewed general EM journals. This protocol met the criteria for exemption from institutional review board review at our institution.

Selection and review methods

All articles published in the Academic Emergency Medicine, American Journal of Emergency Medicine, Annals of Emergency Medicine, and Journal of Emergency Medicine between January 1994 and December 2003 were reviewed. Published articles were classified as original research (author-reported), case reports/series, and others (consensus/educational) Abstracts, book reviews, and edi- torials were articles that were reviewed and classified but not included in this analysis.

Data sources

Two trained physicians reviewed and coded print versions of all 4 journals published within the study time frame. The abstracting physicians received a 30-minute didactic session reviewing article type and information to be obtained from the author byline in each journal. The reviewers then performed reviews of 100 articles with classification and data abstraction. These reviews were compared with reviews performed by 2 senior faculty. The intent was to use the reviews for additional education and to calculate a j statistic for the reviewers; but in practice, no discrepancies were found in the reviewed classifications. Coded data were transferred to an electronic database using double data entry for error checking.

Data extraction and classification

The author byline from the print version of each article was reviewed for number, specialty, nationality, collabora-

Table 1 Time trends for the decade with average change per year and 95% CIs; the P value for significance is also included
	1994	2003	Average change/y (95% CI)	Significance
Total no. of articles	581	566	–4.7 (–11 to 1.6)	.18
Original research	292	358	2 (–3.5 to 7.4)	.51
Case reports/series	164	128	–4.8 (–7.8 to –1.7)	.01
Other articles	125	80	–1.9 (–5.8 to 2)	.37
Mean number of authors	3.3	4.1	0.08 (0.05 to 0.10)	.0000003
Foreign articles (non-United States)	60	112	3.5 (1.3 to 5.7)	.02
Multidisciplinary collaboration	195	242	1.6 (–1.7 to 4.8)	.37
Intradisciplinary collaboration	233	287	5.2 (1.9 to 8.5)	.02
Extramural funding	100	153	2.2 (–7.4 to 5.2)	.18

tion, and extramural funding. Multidisciplinary collabora- tion was defined as authors noted from 2 or more specialties, whereas multi-institutional collaboration was defined as EM authors from more than one institution.

Data analysis

Data for time trends are reported at the beginning and end of the study period with average change per year and 95% confidence intervals (CIs). Authorship number was evaluated both as a continuous variable in the logistic regression model and as a categorical variable (summing authors numbering greater than 5 into a single group) for the purposes of reporting. Logistic regression using the presence of funding as a response variable was used to identify predictors of extramural funding from the variables collected.

Summary statistics, significance testing, and logistic regression were determined using R: A Language and Environment for Statistical Computing, version 2.0.1 (R Development Core Team, Vienna, Austria).

Results

Of 5728 articles identified, there were 3278 (57%)

original research, 1437 (25%) case reports/series, and 975 (17%) classified as others. Time trends for article types are shown in Fig. 1 using data from the 4 reviewed journals in aggregate.

The percentage funded was 22% for all articles in the period reviewed (32% for original research). The articles had at least one EM investigator as coauthor 84% of the time. Article location of origin was the United States (63%), foreign (15%), and combined (22%). Multidisciplinary collaboration increased overall from 33% in 1994 to a high

Fig. 2 Changes in number of authors per article over the last decade, for 1 through 4 and for 5 or more authors.

Fig. 3 The percentage of interdisciplinary and intradisciplinary and the percentage of funded research. All 3 variables have increased.

of 43% in 2003. Multi-institutional EM collaboration also increased from 16% in 1994 to 26% in 2003 (Table 1). The percentage of articles having 5 or more authors increased from 12% to 18% over the decade, with full time trends shown in Fig. 2.

The logistic regression model looking for predictors of extramural funding identified a strong association between article type and a weak association with foreign source of funding (original research: odds ratio, 4.8; 95% CI, 4.0-5.6) and foreign source (non-United States: odds ratio, 1.3; 95% CI, 1.1-1.5).

Discussion

In recent years, the concept of authorship has been the subject of great debate, mainly because of the proliferation of authors over the years [4]. One hundred years ago, 98% of the articles contained in the journal that became The New England Journal of Medicine were written by a single author; today, less than 5% are. In the past 20 years, the mean author number in that same journal has increased from 4 to more than 6 [5]. In our study, the percentage of articles having 6 or more authors increased from 12% to 18% over the decade. Multiple authorship of articles is now the norm [6]. This increase in authors is mainly due to inclusion of bgraftersQ (individuals listed as authors but who have made no contribution to the author), bghostsQ (unnamed authors who nevertheless actually held the pen), and bguestsQ (individuals who made no research or writing contribution to the manuscript but whose author- ship was awarded because of their position as department chair or grant awardee) [5]. It has been argued that this trend has not only led to a dilution of importance, but to a disconnect between credit and responsibility [1]. As a result, authorship disputes arise in which everyone takes credit and fewer take

responsibility [5]. The coin of publication has 2 sides: credit and accountability. The expansion in number of authors per article has tended to dilute accountability, while scarcely seeming to diminish credit [6].

In addition to the increase in authors and number of researchers, there has also been an increase in the intensity of competition among them [2]. As techniques become more specialized and problems multifaceted, researchers are relying more and more on collaborative efforts to remain competitive [7]. The nature of science and the growth of specialization inevitably led to collaboration [1]. The value of collaboration to produce high-quality clinical research has been well recognized. These include cross-fertilization of ideas, wider applicability and generalizability of the study, access to equipment or space, and recruitment of personnel with special expertise. Collaboration can occur with experts in disciplines outside EM or within (Fig. 3) [7]. In reviewing the literature, we found increasing trends in both multidisci- plinary and multi-institutional EM collaboration.

In the past, EM has missed out on collaborative opportunities mostly because of the newness and frag- mented presence of academic EM; however, the situation is beginning to change. As the specialty continues to grow in strength and stature, the impetus to include emergency physicians as collaborative investigators will increase. In addition to developing closer academic ties to other specialties, collaboration can improve clinical and adminis- trative relationships with other specialties; and it is virtually a necessity for studies requiring long-term follow-up. The value of collaboration is evident because collaborative studies are more likely to be funded and published in journals with broad readership [7].

Funding is vital to the proliferation of academic EM. Expertise among EM researchers is being increasingly recognized by potential funding agencies [7]. The current trend toward outcomes-based studies strongly favors the ED as a major site for clinical trials [7]. Yet, the majority of published EM research is unfunded [8]. Singer et al reported that research published in EM journals was less likely to have been funded than in non-EM journals [8]. Determi- nants of funding are multifactorial. According to our research, there does not seem to be a correlation with the increasing trends in EM of multiple authorship and collaboration on the impact of funding.

Limitations

The most important limitation to consider is that this study only reviewed the 4 general peer-reviewed EM journals published in the United States. The subSpecialty journals that round out our specialty as well as the ever- increasing EM journals published abroad were not consid- ered in this analysis. It has been shown that these 4 EM journals account for more than half of the articles originating from an EM unit (department/division/section)

[9]. It is important to note that the focus of this article was in response to the Macy report that was published in 1995 and set forth recommendations to further the development of research specific to the field of EM. A conference of

29 academic emergency physicians (Research Director’s Conference) was organized to address the Macy report and set goals for continued growth of the specialty and increased research productivity. The main recommendations were as follows: (1) enhance support for basic, clinical, and health services research pertinent to EM practice; (2) promote collaborative and interdisciplinary research within and across traditional and institutional boundaries; (3) develop new systems to manage clinical information; (4) develop new methods to assess the outcomes of emergency care; and

(5) seek and develop increased funding sources for EM research [10].

The use of an analysis directed at the end product of funding (published work) rather than the beginning of the research product (grants awarded) provides a different perspective that must be interpreted with caution. Although it is encouraging that collaborations of all types are increasing for our definition of EM, it should not be discouraging that the end product of the research effort has not shown a correlation between collaboration and funding given the amount of time necessary to go from initial funding to publication. It is encouraging to recognize that an increase in the collaborative efforts both within and outside of our specialty is moving in the right direction.

Conclusion

During the past decade, there has been an increase in author number and collaboration in research publications within and outside of EM. This trend will likely continue in the future, which gives rise to many questions as to what lies ahead and what impact, if any, it will have on extramural funding of EM research.

References

1. Rennie D. Who did what? Authorship and contribution in 2001. Muscle Nerve 2001;24:1274 – 7.
2. Schechter AN, Wyngaarden JB, Edsall JT, Maddox J, Relman AS, Marcia A, et al. Colloquium on scientific authorship: rights and responsibilities. FASEB J 1989;3(2):209 – 17.
3. Biros MH, Barson WG, Lewis RJ, Sanders AB. Supporting emergency medicine research: developing the infrastructure. Acad Emerg Med 1998;5(2):177 – 84.
4. Drenth JP. Multiple authorship: the contribution of senior authors. JAMA 1998;280(3):219 – 21.
5. Constantian MB. The Gordian knot of multiple authorship. Plast Reconstr Surg 1999;103(7):2064 – 6.
6. Rennie D, Yank V, Emanuel L. When authorship fails. A proposal to make contributors accountable. JAMA 1997;278(7):579 – 85.
7. Lewis LM, Callaham ML, Kellermann AL, Marx JA, White JD. Collaboration in emergency medicine research: a consensus statement. Ann Emerg Med 1998;31(2):161 – 5.
8. Singer AJ, Homan CS, Stark MJ, et al. Comparison types of research articles published in emergency medicine and non-emergency medicine journals. Acad Emerg Med 1997;4(12):1153 – 8.
9. Birkhahn RH, Van Deusen SK, Okpara OI, Datillo PA, Briggs WM, Gaeta TJ. Funding of original research in the emergency

medicine literature over the last decade. Acad Emerg Med 2006;13: 95 – 101.

1. Pollack CV, Hollander JE, O’Neil BJ, et al. Status report: development of emergency medicine research since the Macy report. Ann Emerg Med 2003;42(1):66 – 80.