Original article 355

Relationship between category size and journals’ impact factor: implications for emergency medicine journals and researchers Òscar Miróa, Anthony F.T. Brownc, Colin A. Grahamd, James Ducharmee, Francisco J. Martin-Sanchezb and David C. Conef Objective We assessed the relationship between the size of the 39 Journal Citation Reports (JCR) medical categories and impact factor (IF) of journals in these categories, and the implications that it might have for emergency medicine (EM) journals. Materials and methods Using the 2010 JCR database, we calculated the mean IF, 5-year IF (5y-IF), Eigenfactor (EF), and Article Influence (AI) scores including all journals for each category. We also calculated a ‘weighted IF’ for all journals by dividing each journal IF by the mean IF of its category. We ranked EM journals according to IF and ‘weighted IF’ into all the journals included in the 39 categories. We assessed the relationship between category size and bibliometric scores by linear regression. Results Category size varied from 252 journals (Pharmacology and Pharmacy) to 14 (Primary Healthcare), EM category occupying the 36th position (23 journals). The mean IF of EM category ranked in 34th position, 5-yIF in 32nd, EF in 34th, and AI in 34th position. Category size had a direct and significant association with mean IF, 5y-IF, and AI but not with mean EF. When the EM journals were ranked among all the journals according to their IF, only two (9%) were placed into the first quartile and raised up to eight (35%) when ‘weighted IF’ was considered.

Introduction Emergency medicine (EM) as a specialty historically involves extensive medical activity and a large number of doctors. However, its recognition as an independent specialty came later than many other specialties and in many countries, certainly in Europe, Asia, and Africa, EM specialty and residency programs continue to await formal approval. In addition, research in EM has suffered from delays with respect to other disciplines [1–3]. One reflection of this could be the low number of EM journals currently indexed by the Thomson Reuters Journal Citation Reports (JCR). Noticeable too is their relatively low impact factors (IF) in comparison with other categories of biomedical research. Two decades ago, Gallagher and Barnaby suggested that these apparent low scores could not be exclusively attributed to qualitative and quantitative aspects of EM research itself, but also to a methodological bias inherent in the derivation of the IF [4,5]. In this sense, recent articles that analyzed the position of EM journals in 0969-9546 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.

Conclusion There is a negative relationship between JCR size category and IF achieved by the journals. This places EM journals at a clear disadvantage because they represent one of the smallest clinical medical research disciplines. European Journal of Emergency Medicine 22:355–362 Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved. European Journal of Emergency Medicine 2015, 22:355–362 Keywords: bibliometry, emergency medicine, Journal Citation Reports a Editor-in-Chief, Emergencias, Department of Emergency Medicine, Hospital Clinic de Barcelona, Barcelona, bAssociate Editor, Emergencias, Department of Emergency Medicine, Hospital Clínico San Carlos, School of Medicine, Universidad Complutense, Madrid, Spain, cFormer Editor-in-Chief, Emergency Medicine, Department of Emergency Medicine, Royal Brisbane and Women’s Hospital, School of Medicine, University of Queensland, Brisbane, Queensland, Australia, dEditor-in-Chief, European Journal of Emergency Medicine, Accident and Emergency Medicine Academic Unit, The Chinese University of Hong Kong, Shatin, Hong Kong, eEditor-in-Chief, Canadian Journal of Emergency Medicine, Department of Medicine, Division of Emergency Medicine, McMaster University, Hamilton, Ontario, Canada and fEditor-in-Chief, Academic Emergency Medicine, Yale University School of Medicine, New Haven, Connecticut, USA

Correspondence to Òscar Miró, MD, PhD, Emergency Department, Hospital Clínic, Villarroel 170, 08036 Barcelona, Catalonia, Spain Tel: + 34 227 98 33; fax: + 34 93 227 56 93; e-mail: [email protected] Received 21 January 2014 Accepted 26 June 2014

the JCR database showed that, despite an increasing number of international EM journals being included over the past 10 years, EM continues to be a category with fewer journals indexed [6]. In addition, although every EM journal exhibited an increasing IF trend, the gap between EM journals’ IFs has widened in the past 10 years [7]. Thus, EM journals rank low in IF summary statistics and in growth trends among the 31 medical and surgical specialties analyzed in a recent paper [6]. This underscoring of EM journals could negatively impact on emergency physician community in general, as well as on the EM specialty itself [8]. Specifically, it could affect emergency physician researchers when they apply for academic positions or general grants, as, despite its limitations, the IF of the journals that authors have contributed to continues to be taken into account [9,10]. For example, two competing doctors performing research in two different specialties who have published a similar DOI: 10.1097/MEJ.0000000000000184

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

356 European Journal of Emergency Medicine 2015, Vol 22 No 5

number of papers in the best journals of their corresponding specialty could hugely differ in their cumulated IFs due to the mere fact that the number of journals indexed in their specialties is very dissimilar. In the present paper, we have assessed the relationship between the size of the JCR categories and the IF achieved by those journals, as well as the implications that this might have for EM journals. We postulate that one of the main barriers to achieving a higher IF for EM journals is the number of JCR indexed journals in the EM category. Bearing in mind that only the citations appearing in the articles published in the journals indexed by the JCR count for the IF calculation, it seems clear that the larger the number of journals indexed in a particular category, the higher the possibility for each journal included in that particular category to get citations, and consequently to increase its IF. If that is true, research and researchers included in a category in which the JCR indexes a smaller number of journals, such as EM category, are at a disadvantage with respect to those of other specialties.

Methods This exploratory study was performed using the 2010 JCR database. This was chosen because it was the most recent complete update available to the authors in Excel sheet format, as well as being used in the last papers published in this field [6,7]. The database was obtained from FECyT website (Spanish Ministry of Health) [10] and corresponds to that published in the Web of Knowledge by the JCR Science Citation Index edition [11]. Among the 179 categories within which JCR distributes the 7797 journals indexed in the 2010 report, we selected the 39 categories related to clinical medicine research. The number of journals included in each category was compiled. For each category, we calculated the mean IF, 5-year IF (5y-IF), Eigenfactor (EF), and Article Influence (AI) scores including all journals of such a category, as well as using only the journals ranked in the first decile of each category according to their IFs. The IF is among the most classic bibliometric scoring systems. It was formulated by Garfield [12] and is annually published by the JCR. For a particular journal and for a particular year, the IF is calculated as the number of citations received by papers published in the 2 preceding years divided by the number of citable articles published during those two preceding years. The JCR also includes other bibliometric parameters. The 5y-IF relies on the same concept and formulation as the IF but enlarges the period of citation and citable papers published to 5 years instead of 2. The EF score is also a rating of the overall importance of a scientific journal but has a more complex calculation. It is measured using the current JCR year citations to citable items from the 5 previous years [13]. However, the IF weighs each citation to a journal equally, whereas the EF assigns a greater weight to those citations coming from the most

influential journals, allowing these journals to exert greater influence in the determination of the rank of any journal that they reference. In addition, the EF does not count journal self-citations. The sum of EF for all journals included in JCR is 100, and each journal’s EF is a part of this total [14]. Journals generating a higher impact to the field have larger EF scores. Finally, the AI measures the relative importance of the journal on a perarticle basis. It is calculated by dividing the journal’s EF by the fraction of articles published by the journal. That fraction coming from all journals is then normalized; hence, the mean AI of all indexed journals is 1.00. A score greater than 1.00 indicates that articles in that journal have above-average influence; a score less than 1.00 indicates that articles in the journal have a below-average influence [14]. The IF, 5y-IF, EF, and AI were individually extracted for each of the 23 journals indexed in the EM category in the 2010 JCR database. We then ranked the IF of the EM journals into all the journals included in the 39 categories evaluated in our study. We also calculated a ‘weighted IF’ for all journals included in the present study. We defined a weighted IF as dividing each journal’s IF by the mean IF of its category. The EM journals were then ranked again according to weighted IF. We further ascertained the relationship between the number of journals indexed in a category with the bibliometric scores achieved by the journals included in that category and the mean bibliometric scores achieved by journals situated in the first decile of each category. This was performed by means of a linear regression model. R2 (degree of determination) was used to estimate how well data points fitted the line and how much the proportion of the total variation in bibliometric scores was explained by the variation in the category size where the journal was indexed. Values of P less than 0.05 were considered statistically significant.

Results The number of journals contained in the 39 categories analyzed in the present study ranged from 252 journals in the category of Pharmacology and Pharmacy to 14 in the category of Primary Healthcare. The EM category occupied the 36th position with 23 journals, the same number as Critical Care Medicine (Table 1). For the EM journals, mean IF ranked them in the 34th position of 39, mean 5-yIF in the 32nd position, mean EF in the 34th position, and mean AI in the 34th position. When the mean bibliometric scores of the journals in the first decile of each category were taken into account, the EM category ranked in the 34th, 33rd, 29th, and 30th positions, respectively. The size of a category had a direct and statistically significant association with the mean IF, 5y-IF, and AI of the category, irrespective of whether these scores were

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

EM journals and impact factors Miró et al. 357

Table 1 Categories related to biomedicine research in 2010 Journal Citation Reports database ordered by the number of journals that each category includes and some descriptive metrics of each category For journals into the first decile of the category according to the impact factor

For all journals into the category

JCR category Pharmacology and pharmacy Neurosciences Surgery Clinical neurology Oncology Medicine, general and internal Immunology Psychiatry Endocrinology and metabolism Cardiac and cardiovascular systems Radiology, nuclear medicine and medical imaging Pediatrics Microbiology Nursing Toxicology Obstetrics and gynecology Pathology Gastroenterology and hepatology Nutrition and dietetics Peripheral vascular disease Urology and nephrology Hematology Orthopedics Infectious diseases Ophthalmology Dermatology Respiratory system Geriatrics and gerontology Rehabilitation Otorhinolaryngology Virology Parasitology Rheumatology Anesthesiology Transplantation Emergency medicine Critical care medicine Neuroimaging Primary healthcare

Journals included

Mean impact factor

Mean 5-year impact factor

Mean Eigenfactor

Mean Article Influence

Mean impact factor

Mean 5-year impact factor

Mean Eigenfactor

Mean Article Influence

252 237 187 185 182 153 133 128 115

2.766 3.560 1.611 2.572 4.012 2.526 4.174 2.568 3.499

3.040 4.000 1.971 2.903 4.426 3.421 4.328 3.142 3.847

0.0094 0.0187 0.0094 0.0109 0.0208 0.0171 0.0229 0.0102 0.0149

0.8628 1.4641 0.5714 0.9145 1.4915 1.2335 1.6808 0.9731 1.2168

8.904 11.728 4.317 7.617 17.256 15.214 17.184 7.978 10.644

8.964 12.512 4.406 7.699 15.540 15.304 16.401 8.411 10.323

0.0269 0.0780 0.0389 0.0401 0.1051 0.1394 0.0879 0.0412 0.0478

2.8568 5.4518 1.3716 2.6792 5.9604 6.1419 8.0368 2.9070 3.8191

114

2.709

3.110

0.0157

1.0200

8.415

7.987

0.0832

2.6878

113

2.175

2.400

0.0106

0.7112

5.233

5.399

0.0407

1.6901

107 103 89 83 77 75 72

1.612 3.300 0.968 2.584 1.781 2.430 2.816

1.893 3.742 1.456 2.748 2.094 2.671 3.149

0.0074 0.0188 0.0016 0.0071 0.0079 0.0073 0.0150

0.5834 1.3415 0.3572 0.7631 0.6034 0.8605 0.9354

3.749 11.788 1.803 6.913 4.563 7.654 8.784

4.078 12.850 2.300 7.114 4.648 7.100 8.344

0.0322 0.0434 0.0033 0.0154 0.0329 0.0245 0.0737

1.3494 2.2740 0.5856 2.4046 1.3729 2.7634 2.5754

70 68 68 66 60 57 56 55 45 43 43 41 33 31 29 26 25 23 23 14 14

2.334 2.661 2.293 3.306 1.400 3.123 1.784 1.874 2.597 2.575 1.402 1.149 2.968 0.570 3.018 2.190 2.526 1.428 2.873 2.101 1.539

2.829 2.917 2.435 3.294 1.912 3.284 2.093 2.208 2.899 2.909 1.866 1.423 3.137 2.561 3.212 2.260 2.651 2.094 3.229 2.391 2.113

0.0088 0.0189 0.0116 0.0271 0.0084 0.0191 0.0097 0.0059 0.0158 0.0068 0.0036 0.0052 0.0212 0.0091 0.0150 0.0101 0.0147 0.0060 0.0195 0.0188 0.0041

0.7964 0.8916 0.7272 1.0391 0.5847 1.1144 0.6791 0.5674 0.9036 0.8828 0.5110 0.4589 1.0465 0.8081 0.9655 0.6139 0.7453 0.5722 1.0129 0.8023 0.6774

6.071 8.481 6.461 9.758 3.343 8.765 5.326 4.776 7.289 6.829 2.959 2.488 7.009 6.246 8.759 5.421 6.128 4.160 8.223 5.937 4.457

6.963 9.117 5.887 9.856 3.913 8.460 5.238 4.556 7.138 6.759 3.679 2.769 7.581 6.603 8.065 5.785 5.246 4.414 8.236 6.821 4.969

0.0254 0.1085 0.0490 0.1930 0.0234 0.0874 0.0427 0.0289 0.0730 0.0165 0.0061 0.0076 0.0561 0.0371 0.0866 0.0454 0.0360 0.0286 0.0958 0.1535 0.0108

2.1694 3.1582 1.9063 3.4995 1.0998 3.1712 1.8284 1.3232 2.3298 2.2805 1.0663 1.0590 3.0430 2.4813 2.4135 1.6890 1.3635 1.6310 2.8200 5.9370 1.9210

JCR, Journal Citation Reports. Bold figures correspond to the emergency medicine category.

compared with all of the journals indexed in the category or to the journals ranked in the first decile (Fig. 1). This relationship was not observed for the mean EF scores. Of note, the EM journal category scores were always placed below the regression line of the model. The 23 EM journals included in the 2010 JCR database along with their bibliometric parameters are presented in Table 2. The highest IF and EF were accorded to Resuscitation, whereas the highest 5y-IF and AI were accorded to Annals of Emergency Medicine. In the light of the definitions of the bibliometric parameters described previously, the journal Resuscitation with the highest EF contributed with a 0.334 thousandths to the whole impact of all journals indexed by the JCR, and only two journals,

Resuscitation and Annals of Emergency Medicine, had AI scores above 1, which means that they were the only EM journals ranked above the mean AI for all EM journals indexed by the JCR. When the EM journals were ranked among all the journals analyzed in the present study according to their IF, only two of 23 (9%) were placed into the first quartile. When the IFs of journals are weighted by the mean IF of their category, up to eight of them (35%) would be allocated into the first quartile (Table 3).

Discussion The results of this study support our hypothesis that there is a negative relationship between the size of a JCR

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

358 European Journal of Emergency Medicine 2015, Vol 22 No 5

3 2 1 0

Mean Eigenfactor of journals in the category

Mean 5-year impact factor of journals in the category

0

50

100

150

200

5

250

300

R2 = 0.1755 P = 0.008

4 3 2 1 0

0

50

100

150

200

0.03

250

300

R2 = 0.0204 P = 0.385

0.025 0.02 0.015 0.01 0.005 0

Mean article influence of journals in the category

0 2

50

100

150

200

250

300

R2 = 0.1804 P = 0.007

1.5

1

0.5

0

0

50 100 150 200 250 Number of journals in the JCR category

300

Mean impact factor of journals in the first decile of the category

4

20

R2 = 0.2263 P < 0.001

15 10 5 0 0

Mean 5-year impact factor of journals in the first decile of the category

R2 = 0.1461 P = 0.02

20

Mean Eigenfactor of journals in the first decile of the category

5

0.25

50

100

150

200

250

300

R2 = 0.2358 P = 0.002

15

10

5

0

0

50

100

150

200

250

300

R2 = 0.0039 P = 0.71

0.2 0.15 0.1 0.05 0 0

Mean Eigenfactor of journals in the first decile of the category

Mean impact factor of journals in the category

Fig. 1

50

100

150

10

200

250

300

R2 = 0.1415 P = 0.02

8 6 4 2 0 0

50 100 150 200 250 Number of journals in the JCR category

300

Relationship between the number of journals indexed in each category of 2010 Journal Citation Reports and the mean (for all journals, left; for only the first decile journals, right) of some bibliometric parameters achieved by each category. White biggest dots correspond to the EM category. EM, emergency medicine.

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

EM journals and impact factors Miró et al. 359

Table 2

Emergency medicine journals included in 2010 JCR database

JCR category

Impact factor

Five-year impact factor

Eigenfactor

Article Influence

Elsevier Scientific Publishers Mosby-Elsevier Sanidad Ediciones Elsevier Science Ltd Blackwell Science Publishers Biomed Central Ltd W B Saunders Co-Elsevier Inc. Informa Healthcare Elsevier North Holland Decker Publishing Inc. W B Saunders Co-Elsevier Inc. B M J Publishing Group Blackwell Science Publishers Lippincott Williams & Wilkins Lippincott Williams & Wilkins Springer Elsevier North Holland Springer Springer Heidelberg

4.177 4.142 3.085 2.269 2.197 2.176 1.994 1.889 1.552 1.515 1.294 1.269 1.089 0.901 0.803 0.675 0.429 0.400 0.257

4.297 4.531 – 2.435 2.536 – 1.794 – 1.325 – 1.237 1.442 – – 0.954 0.741 0.617 – 0.299

0.0334 0.0238 0.0007 0.0168 0.0170 0.0006 0.0103 0.0055 0.0066 0.0025 0.0021 0.0081 0.0025 0.0021 0.0054 0.0022 0.0017 0.0002 0.0004

1.5490 1.7130 – 0.6850 0.9330 – 0.6180 – 0.4640 – 0.3890 0.4350 – – 0.3170 0.1830 0.1940 – 0.0700

Turkish Association of Trauma Emergency Surgery Pharmamed Mado Ltd Medcom Ltd Georg Thieme Verlag

0.211



0.0003



0.200 0.167 0.154

0.194 – –

0.0001 0.0001 0.0000

0.0880 – –

Publisher

Resuscitation Annals of Emergency Medicine Emergencias Injury Academic Emergency Medicine Scandinavian Journal of Trauma Resuscitation American Journal of Emergency Medicine Prehospital Emergency Care Journal of Emergency Medicine Canadian Journal of Emergency Medicine Emergency Medicine Clinics of North America Emergency Medicine Journal Emergency Medicine Australasia European Journal of Emergency Medicine Pediatric Emergency Care Unfallchirurg Journal of Emergency Nursing Notfall Rettungsmed European Journal of Trauma and Emergency Surgery Ulusal Travma ve Acil Cerrahi Dergisi Signa Vitae Hong Kong Journal of Emergency Medicine Notarzt JCR, Journal Citation Reports.

category and the IF achieved by the journals contained in that category. This would place EM journals at a clear disadvantage given that they – as a whole – represent one of the smallest clinical medical research disciplines. This means that, when EM research output is compared with that from other disciplines or when emergency physicians engaged in research are compared with researchers in other disciplines, not all the differences found in the IF of the respective journals are related to the true value of the research or the quality of the researchers themselves. Accordingly, our finding could be a solid argument to face comments or statements regarding the low quality of EM research, especially if these positions are taken based on the low IF of papers published by emergency physicians.

hand, emergency physicians are able to produce highquality research, although in many cases these best EM research papers are published in non-EM journals, particularly internal medicine journals [16–18]. In contrast, and, according to our data, we observed that, after obtaining a weighted IF, which should reduce the demonstrated effects of the category size on the IF, 35% of EM journals would be in the first quartile of all examined journals. This equates to a 10% above the 25% that a first quartile mathematically implies. Although 5y-IF was not examined in the present study in depth, it seems to behave similarly to IF and also is influenced by the category size. This is not surprising, as the 5y-IF has a similar basis as IF.

What determines the number of journals indexed in a category and what drives the inclusion of new journals into the JCR are not clear. It is not stated whether the journal category coverage by the JCR is based on research quality or quantity within a particular field. It is also unclear if both criteria are used, based on a proportional mix. The 23 indexed journals into the EM category are a small number compared with other disciplines, especially if the bulk of activity performed in the ED is taken into account. One could argue that there is not enough research activity justifying additional inclusions in this category. Although scientific publication rates in EM have traditionally been lower than those of other medical specialties, steady progress has been made since 1995, with rates of increase exceeding that observed in other medical disciplines [1,15]. It could also be argued that research performed in EM does not reach the required quality. Again, this seems not to be the case. On the one

The EF is intended to measure the importance of a journal to the scientific community, by considering the origin of the incoming citations, and is thought to reflect how frequently an average researcher would access content from that journal. The EF approach is more robust than the IF metric, which simply counts incoming citations without recognizing the significance of those citations. From the perspective of our study, the EF seems to be less influenced than IF and 5y-IF by the category size, as we were unable to uncover any relationship. However, the EF score is influenced by the size of the journal, such that the score doubles when the journal doubles in size (measured as published articles per year). To correct this bias, the AI score measures the average influence of articles in the journal [14]. Again, we have observed that AI is to some extent affected by category size, underestimating the real value of journals of disciplines with a

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

360

European Journal of Emergency Medicine 2015, Vol 22 No 5

Table 3

Ranking of the emergency medicine journals categorized in the 2012 Journal Citation Reports

JCR category Resuscitation Annals of Emergency Medicine Emergencias Injury Academic Emergency Medicine Scandinavian Journal of Trauma Resuscitation American Journal of Emergency Medicine Prehospital Emergency Care Journal of Emergency Medicine Canadian Journal of Emergency Medicine Emergency Medicine Clinics of North America Emergency Medicine Journal Emergency Medicine Australasia European Journal of Emergency Medicine Pediatric Emergency Care Unfallchirurg Journal of Emergency Nursing Notfall Rettungsmed European Journal of Trauma and Emergency Surgery Ulusal Travma ve Acil Cerrahi Dergisi Signa Vitae Hong Kong Journal of Emergency Medicine Notarzt

Position/quartile according the weighted IF (into all biomedical journals)

Impact factor

Position/quartile according to IF (into its category)

Position/quartile according to IF (into all biomedical journals)

Weighted IF

4.177 4.142 3.085 2.269 2.197 2.176

1/1 2/1 3/1 4/1 5/1 6/2

483/1 495/1 865/2 1417/2 1479/2 1498/2

2.925 2.900 2.160 1.588 1.538 1.523

115/1 116/1 239/1 514/1 545/1 554/1

1.994

7/2

1645/2

1.396

686/1

1.889 1.552 1.515

8/2 9/2 10/2

1735/3 2010/3 2046/3

1.322 1.086 1.060

771/1 1063/2 1097/2

1.294

11/2

2246/3

0.906

1406/2

1.269 1.089

12/3 13/3

2266/3 2452/4

0.888 0.762

1444/2 1711/3

0.901

14/3

2606/4

0.631

2034/3

0.803 0.675 0.429 0.400 0.257

15/3 15/3 17/3 18/4 19/4

2685/4 2795/4 2996/4 3020/4 3112/4

0.562 0.472 0.300 0.280 0.180

2222/3 2434/3 2774/4 2808/4 3001/4

0.211

20/4

3149/4

0.147

3052/4

0.200 0.167

21/4 22/4

3160/4 3179/4

0.140 0.116

3064/4 3098/4

0.154

23/4

3189/4

0.107

3116/4

Weighted impact factors have been achieved by dividing every impact factor by the mean impact factor of the journal category (shadows in some columns of the table represent quartiles: the darker, the higher). IF, impact factor; JCR, Journal Citation Reports.

small presence in the JCR database. For example, only two of 13 EM journals with AI (15%) ranked above the mean of all JCR indexed journals. The role of journals and editorial committees in promoting and spreading research is critical. Some editors feel that a new journal in their category is a threat to their own journal, because it competes for the same articles, authors, and audiences. This viewpoint should actively be discouraged, as our study suggests that the greater the number of journals in a JCR category, the higher the IF of all the journals within this category. Collaboration among different journal editors themselves can accelerate the spread of knowledge, scope, and the reputation of research performed in their field [19–23]. This is not a contradiction to maintaining the best quality in papers published in their own journals and is in keeping with the strict ethical commitments that lead editors’ decisions [24,25]. Undoubtedly, there are multiple pathways to improve the quality and quantity in EM research that would favor a greater impact of the EM journals. Thus, initiatives should be addressed to improve research skills in

emergency physicians, such as involving residents in research projects developed in the ED [26–28]. The inclusion of a mandatory research education rotation into resident programs including critical appraisal, workshop seminars, and integration into the faculty research projects would lead to an increase in scholarly output. Evaluations have indicated generally high satisfaction throughout this type of rotation. The experience of the Royal College of Physicians and Surgeons (Emergency Medicine) in Canada has been highly satisfactory [26]. Another opportunity for improving research output is to raise the number of conference abstract presentations being submitted as completed papers for publication. During the last 10 years, ∼ 35% of American, Canadian, British, Australian, or international EM meeting and conference abstracts have been submitted for publication. This is lower than that reported by other established specialties [29–32]. Further exploration into the barriers preventing the publication of studies presented at these meetings is required [21]. Another potential pathway that EM researchers have to explore and foster is the collaboration among EM research groups as well as with other specialties, both locally and internationally. The creation

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

EM journals and impact factors Miró et al. 361

of stable networks of research centers and researchers in EM should be a key target for EM societies (and EM journals) worldwide [33–37]. The evaluation of the prestige of the EM journals does not directly translate to the quality of EM researchers. Thus, the IF is not an adequate parameter to measure a researchers’ value; this is better profiled by the h index. This bibliometric score was introduced by Hirsch [38] in 2005, and, since then, it has gained greater acceptance with the research community. Few studies, however, have been conducted looking at EM researchers. De Luca et al. [39] reported that 59% of academic emergency physicians in the USA had an h-index of 0, and 85% had an h-index less than 6. When they analyzed the most productive American EM researchers, they found that a quotient higher than 0.5 between h-index and years since the first publication defined the most productive researchers. Interestingly, the h-index achieved by year 12 was strongly related to the likely future research productivity. A similar hypothesis has been suggested from an analysis of 24 Spanish emergency physicians involved in research. Here, it was possible to model changes in the output and IF of these 24 researchers. These models can be used in the future to compare the behavior of these research indicators with those of other Spanish research groups or with those of emergency physicians engaging in research in other countries [40,41]. In addition, it could serve to predict further productivity in those researchers fitting the model. However, as h-index is an evaluation tool that is also based on the number of citations achieved by authors, it appears that the effect of an under-representation of EM journals could again negatively influence the projection of EM research and researchers. Our work has some limitations. It has been restricted to 39 categories related to clinical medical research; hence, a broader analysis might have demonstrated a different result. On the other hand, the weighted IF was a construct created by the authors. Finally, the ratio between the number of ‘research active’ clinicians and raw number of clinicians in the specialties (which have not been calculated for any of the analyzed categories) has to be taken into account when considering the pertinence of a particular category size. For example, if there were only 100 academic emergency physicians, they would need fewer journals than 1000 academic cardiologists. Then, this uninvestigated factor has to be considered when interpreting our results. However, this relationship is not always present; it is interesting to note that primary healthcare category scored very low, despite having a large number of academic primary care physicians. Despite the above limitations, we believe that it is in the best interest of EM research and researchers and ultimately of the patients suffering from urgent and emergency processes to increase the number of journals included in the JCR category of EM and to address the under-representation of EM journals as soon as possible.

Acknowledgements Conflicts of interest

The authors have editorial roles as listed below: OM: Editor-in-Chief, Emergencias. AFTB: Former Editor-in-Chief, Emergency Medicine Australasia. CAG: Editor-in-Chief, European Journal of Emergency Medicine. JD: Editor-in-Chief, Canadian Journal of Emergency Medicine. FJMS: Associate Editor, Emergencias. DCC: Editor-in-Chief, Academic Emergency Medicine.

References 1 2 3

4 5

6

7

8 9 10

11

12 13 14

15

16

17

18

19 20 21 22

Miró O, Burillo-Putze G. Research in emergency medicine in Europe. Eur J Emerg Med 2012; 19:63–68. Graham CA. Impact factor, change and academic emergency medicine. Eur J Emerg Med 2009; 16:169. Miró O, Valcárcel De La Iglesia MA, Cremades Pallas RM, Burillo-Putze G, Julián Jiménez A, Martín Sánchez FJ. Productivity of Spanish emergency physicians in the 5-year periods from 2005 to 2009 and 2000–2004. Emergencias 2012; 24:164–174. Gallagher EJ, Barnaby DP. Evidence of methodologic bias in the derivation of the Science Citation Index impact factor. Ann Emerg Med 1998; 31:83–86. Barnaby DP, Gallagher EJ. Alternative to the Science Citation Index impact factor as an assessment of emergency medicine’s scientific contributions. Ann Emerg Med 1998; 31:78–82. Reynolds JC, Menegazzi JJ, Yealy DM. Emergency medicine journal impact factor and change compared to other medical and surgical specialties. Acad Emerg Med 2012; 19:1248–1254. Lee CH, Shih CP, Chang YC, Chaou CH. The evolution of academic performance in emergency medicine journals: viewpoint from 2000 to 2009 journal citation reports. Acad Emerg Med 2011; 18:898–904. Kumar V, Upadhyay S, Medhi B. Impact of the impact factor in biomedical research: its use and misuse. Singapore Med J 2009; 50:752–755. Miró O. The new impact factor. Emergencias 2012; 24:262–264. Portal to Web of Science. Fundación Española para la Ciencia y la Tecnología (FECyT). Available at: http://www.accesowok.fecyt.es [Accessed 17 September 2013]. ISI Web of Knowledge. Journal Citation Reports. Available at: http://adminapps.webofknowledge.com/JCR/JCR?SID = R24vzCQI5KzTHE5lqHj [Accessed 17 September 2013]. Garfield E. Citation indexes to science: a new dimension in documentation through association of ideas. Science 1955; 122:108–111. Bergstrom CT, West JD, Wiseman MA. The Eigenfactor Metrics. J Neurosci 2008; 28:11433–11434. Clemson University Libraries. Journal citation reports (JCR): definitions/ terminology. Available at: http://clemson.libguides.com/content.php?pid = 164111&sid = 1384479 [Accessed 07 August 2013]. Miró O, Salgado E, Gonzalez-Duque A, Tomas S, Burillo-Putze G, Sanchez M. Scientific production of Spanish emergencists over the last 30 years (1975–2004): a comparison with the production of other specialties in Spain and with that of emergencists abroad. Emergencias 2007; 19:59–64. Litt HI, Gatsonis C, Snyder B, Singh H, Miller CD, Entrikin DW, et al. CT angiography for safe discharge of patients with possible acute coronary syndromes. N Engl J Med 2012; 366:1393–1403. CRASH-2 trial collaborators. Shakur H, Roberts I, Bautista R, Caballero J, Coats T, Dewan Y, et al. Effects of tranexamic acid on death, vascular occlusive events, and blood transfusion in trauma patients with significant haemorrhage (CRASH-2): a randomised, placebo-controlled trial. Lancet 2010; 376:23–32. Armstrong PW, Gershlick AH, Goldstein P, Wilcox R, Danays T, Lambert Y, et al. STREAM Investigative Team. Fibrinolysis or primary PCI in ST-segment elevation myocardial infarction. N Engl J Med 2013; 368:1379–1387. Graham CA. Emergencias: important research questions for the future. Emergencias 2010; 22:404–405. Brown AFT. What is takes to have a good journal. Emergencias 2009; 21:3–4. Cone D. The growth and maturity of Emergencias as a scientific journal. Emergencias 2010; 22:161–162. Cone DC, Miró O, Javier Martín-Sánchez F. Launching Spanish abstracts. Acad Emerg Med 2012; 19:E1085.

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

362 European Journal of Emergency Medicine 2015, Vol 22 No 5

23 24 25 26

27

28 29 30

31

32

Miró O, Martín-Sánchez FJ, Cone DC. The Spanish language in North American emergency medicine. Emergencias 2012; 24:341–342. Graham CA. Progress and publication ethics. Eur J Emerg Med 2011; 18:243. Martín-Sánchez FJ, Miró O. Reflections on the peer review of manuscripts sent to Emergencias. Emergencias 2012; 24:81–83. Abu-Laban RB, Jarvis-Selinger S, Newton L, Chung B. Implementation and evaluation of a novel research education rotation for royal college of physicians and surgeons emergency medicine residents. CJEM 2013; 15:233–236. Holmes JF, Sokolove PE, Panacek EA. Ten-year experience with an emergency medicine resident research project requirement. Acad Emerg Med 2006; 13:575–579. Rosell-Ortiz F, Mateos Rodríguez A, Miró O. Research in emergency medicine and prehospital emergency care. Emergencias 2012; 24:3–4. Korn CS, Henderson SO. Publication of research abstracts presented at four emergency medicine research forums. Acad Emerg Med 2000; 7:534. Walby A, Kelly AM, Georgakas C. Abstract to publication ratio for papers presented at scientific meetings: how does emergency medicine compare? Emerg Med (Fremantle) 2001; 13:460–464. Macmillan CD, Moore AK, Cook RJ, Pedley DK. Abstract-to-publication ratio for papers presented at scientific meetings: a quality marker for UK emergency medicine research. Emerg Med J 2007; 24:425–426. Chan JW, Graham CA. Full text publication rates of studies presented at an international emergency medicine scientific meeting. Emerg Med J 2011; 28:802–803.

33 34

35

36

37 38 39

40

41

Piñera Salmerón P. Spanish research networks in emergency medicine. Emergencias 2012; 24:234–237. Newgard CD, Beeson MS, Kessler CS, Kuppermann N, Linden JA, Gallahue F, et al. Establishing an emergency medicine education research network. Acad Emerg Med 2012; 19:1468–1475. Mintegi S. Research in pediatric emergency medicine: the research network of the Spanish Society of Pediatric Emergencies. Emergencias 2012; 24:238–240. Klassen TP, Acworth J, Bialy L, Black K, Chamberlain JM, Cheng N, et al. Pediatric emergency research networks: a global initiative in pediatric emergency medicine. Eur J Emerg Med 2010; 17: 224–227. Great Network. Global research on acute conditions team. Available at: http://www.greatnetwork.org/int/. [Accessed 17 September 2013]. Hirsch JE. An index to quantify an individual’s scientific research output. Proc Nat Acad Sci USA 2005; 102:16569–16572. De Luca LA Jr, St John A, Stolz U, Matheson L, Simpson A, Denninghoff KR. The distribution of the H-index among academic emergency physicians in the United States. Acad Emerg Med 2013; 20:997–1003. Burbano P, Martín-Sánchez FJ, Burillo-Putze G, Fernández Pérez C, Del Prado González N, Miró O. Research output and impact of a group of emergency physicians selected according to research career criteria. Emergencias 2013; 25:245–254. Aleixandre-Benavent R. Model of the productivity and scientific impact of Spanish emergency physicians. Emergencias 2013; 25:243–244.

Copyright r 2015 Wolters Kluwer Health, Inc. All rights reserved.

Relationship between category size and journals' impact factor: implications for emergency medicine journals and researchers.

We assessed the relationship between the size of the 39 Journal Citation Reports (JCR) medical categories and impact factor (IF) of journals in these ...
198KB Sizes 0 Downloads 4 Views