Abstract
Background
To use the top 100 articles pertaining to total shoulder arthroplasty (TSA) to understand the impact that social media platforms have on the dissemination of shoulder research while highlighting bibliometric factors associated with Altmetric Attention Score (AAS) to offer insight into the impact that social media platforms have on the dissemination, attention, and citation of shoulder research publications.
Methods
In June 2023, the Altmetric database was searched using the following PubMed MeSH terms: “total shoulder arthroplasty” or “TSA.” Articles with the highest AAS were screened to exclude other topics unrelated to TSA. The top 100 articles that met inclusion criteria were used in the final analysis. Bibliometric factors pertaining to each study were collected for further analysis of article characteristics in accordance with prior studies.
Results
The Altmetric Database query yielded 1,283 studies. After applying our inclusion criteria, the top 118 articles with the highest AAS were identified. The mean AAS was 29.14 ± 42.35, with a range of 13 to 402. The included articles represented 27 journals, with 70 articles attributed to 2 journals: Journal of Shoulder and Elbow Surgery (JSES; 43%) and the Journal of Bone and Joint Surgery (JBJS; 16%). There was a significant increase in AAS for the presence of a conflict of interest (p = 0.042) and open access status (p < 0.01), but no association between the score and citation rate (p > 0.05).
Conclusions
Top articles on TSA, as defined by high AAS, mostly comprise original clinical research performed in the United States or Europe. The presence of a conflict of interest and open access status is associated with an increase in AAS, but there was no association between AAS score and citation rate.
Keywords: Altmetric, Shoulder arthroplasty, Social media, Total shoulder arthroplasty
Traditional evaluations of the impact of research is based on author-level bibliometrics such as the Hirsch index (h-index) as well as journal-level metrics such as impact factor (IF).1,2) These measures are utilized to quantify research productivity, determine grant funding, and climb academic standings.3,4) However, recent works have questioned the generalizability of such measures, which commonly limit their focus on citation rates within the scientific community while failing to consider the impact of research through nontraditional channels; this includes the increase in open-access journals and social media dissemination of research.5,6) As such, researchers and readers are unable to fully understand the influence of a publication using simple bibliometrics. To better ascertain the online impact of an article, the Altmetric Attention Score (AAS) was created using a weighted algorithm based on the amount of social media attention of a given research work.7,8) The score tabulates the attention of research publications on sources such as X (Twitter), Facebook, and LinkedIn and weighs each with differing importance in the AAS. Previous work has validated the score throughout the medical literature and has demonstrated a positive correlation with citation rates, particularly in orthopedic works.9,10,11)
The AAS has been utilized to identify the most impactful articles in shoulder and elbow surgery, including the rotator cuff and the medial ulnar collateral ligament.12,13) However, no prior studies have utilized AAS to evaluate the social media attention garnered by articles pertaining to shoulder arthroplasty. As such, the purpose of the current article was to use the top 100 articles pertaining to shoulder arthroplasty to understand the impact that social media platforms have on the dissemination of total shoulder arthroplasty (TSA) research while highlighting bibliometric factors associated with AAS to offer insight into the impact that social media platforms have on the dissemination, attention, and citation of shoulder research publications.
METHODS
Article Selection Criteria
On June 16, 2023, the Altmetric Explorer database (Digital Science, Holtzbrinck Publishing) was searched for articles pertaining to TSA. The PubMed MeSH term “total shoulder arthroplasty” OR “TSA” was entered into the Altmetric Explorer. After stratification by descending AAS, title and abstract review was performed with articles unrelated or not focused on the TSA procedure excluded. The 100 articles pertaining to TSA with the highest AAS scores were included for analysis. The overall AAS score, calculated and updated in real-time, is determined by averaging relatively weighted social media activity scores that an article receives as well as the number of cumulative citations an article receives (Table 1). The AAS includes activity from multiple online platforms, including Facebook, Twitter, and LinkedIn. There is no limit to the AAS an article receives as a study can continue to accumulate mentions over time. Articles that have received no online attention or mentions obtain an AAS of 0. All Altmetric variables from Table 1 were considered in the analysis.
Table 1. Altmetric Attention Score Calculation.
| Social media outlet | Weight |
|---|---|
| News | 8.0 |
| Blog | 5.0 |
| Policy document | 3.0 |
| Patent | 3.0 |
| Wikipedia | 3.0 |
| X (Twitter) | 1.0 |
| Peer review (Publon, PubPeer) | 1.0 |
| Google+* | 1.0 |
| F1000 | 1.0 |
| Weibo† | 1.0 |
| Syllabi (Open Syllabus) | 1.0 |
| 0.5 | |
| 0.25 | |
| 0.25 | |
| YouTube | 0.25 |
| 0.25 | |
| Q&A (Stack Overflow) | 0.25 |
| Mendeley readers | 0 (Tracked but not considered in calculation) |
*Not trackable since 2019, but historical data kept. The total score is calculated as a function of the social media weights listed in the table. †Not trackable since 2015, but historical data kept.
Secondary bibliometric factors pertaining to each study were collected in accordance with prior studies.10,11,12,13,14) These article characteristics included journal, article title, date of publication, highest degree obtained by the first author, number of authors, number of academic institutions, geographic region of origin of the publication, open access status of the publication, disclosure of any conflict of interest (presence or absence of general self-reported conflict of interest), study design, subject of study (anatomic TSA, reverse TSA, or other), and the number of referenced studies. Similar to prior Altmetric investigations, study design was classified as either original clinical research (prospective/retrospective cohort, case-control, case series/report, or randomized controlled trial), systematic review or meta-analysis, review articles, basic science or biomechanics, or other. The institutional affiliation of the corresponding author was utilized to identify the geographic origin of the article from either the United States, Europe, or other.
Statistical Analysis
All statistical analyses were performed with Stata version 16.1 (StataCorp.). Normality was verified with the Shapiro-Wilks test and was not achieved for any of the included variables. Bibliometric and Altmetric characteristics were summarized by descriptive statistics using means with standard deviations for continuous variables and frequencies with percentages for categorical variables. A multivariable linear regression analysis was conducted to determine the association between article characteristics and AAS. Statistical significance was defined as p < 0.05.
RESULTS
The Altmetric Database query yielded 1,283 studies published between 1995 and 2022. After screening for articles that fit our inclusion criteria, this investigation utilized AAS to sequentially identify the 100 most impactful articles generating online attention that pertained to TSA. While applying our inclusion criteria sequentially starting at the highest AAS article, a total of 153 articles were excluded from the search, including 128 publications that were totally unrelated to TSA and 25 that did not focus on TSAs. As a result, a total of 118 articles were included in our final analysis.
The median number of years since publication was 5 years (interquartile range [IQR], 1–10 years). The overall mean and standard deviation for the AAS was 29.14 ± 42.35, with a range of 13 to 402 (median, 17). The included articles represented 27 journals, with 70 articles attributed to 2 journals and their open-access companions: Journal of Shoulder and Elbow Surgery (JSES; 43% [51 articles]) and the Journal of Bone and Joint Surgery (JBJS; 16% [19 articles]). The top 2 articles identified in this study had AAS scores that were substantially higher (AAS > 200) than all other included articles, which were substantially lower (AAS < 110). Both articles were population-based studies using national registries to identify the risk of serious adverse events following TSAs.15,16) One of the articles was published in British Journal of Sports Medicine and the other in Journal of the American Association of Orthopaedic Surgeons. A complete summary of all included bibliometric and Altmetric characteristics can be found in Tables 2 and 3, respectively.
Table 2. Bibliometric Characteristics of Included Articles.
| Bibliometric characteristics of included articles | Value (n = 118) | |
|---|---|---|
| Journal | ||
| Journal of Bone & Joint Surgery | 19 (16.0) | |
| Journal of Shoulder and Elbow Surgery | 51 (43.0) | |
| Other | ||
| Year of publication | 48 (40.8) | |
| Before 2017 | 50 (42.4) | |
| After 2017 inclusive | ||
| Open access | ||
| No | 68 (57.6) | |
| Yes | 82 (74.0) | |
| Number of authors | 36 (26.0) | |
| Mean ± SD | 5.96 ± 2.51 | |
| Median (range) | 6.00 (1.00–18.0) | |
| First author’s degree | ||
| MD/DO | 91 (77.1) | |
| Other | 27 (22.9) | |
| Number of institutions | ||
| Mean ± SD | 2.84 ± 2.32 | |
| Median (range) | 2.00 (1.00–12.00) | |
| Continent of origin | ||
| North America | 89 (75.4) | |
| Europe | 22 (18.6) | |
| Asia | 2 (1.7) | |
| South America | 4 (3.4) | |
| Africa | 0 | |
| Australia | 1 (0.8) | |
| Conflict of interest | ||
| No | 27 (22.9) | |
| Yes | 91 (77.1) | |
| Study topic | ||
| Basic science and biomechanics | 29 (24.6) | |
| Diagnostic/imaging | 33 (28.0) | |
| Clinical outcomes | 56 (47.5) | |
| Number of citations | ||
| Mean ± SD | 77.8 ± 122 | |
| Median (range) | 41 (0–867) | |
| Number of references | ||
| Mean ± SD | 32.8 ± 15.3 | |
| Median (range) | 31.5 (2.00–126) | |
Values are presented as number (%) unless otherwise indicated.
SD: standard deviation.
Table 3. Altmetric Characteristics of Included Articles.
| Altmetric characteristics of included articles | Value (n = 118) | |
|---|---|---|
| Altmetric Attention Score | ||
| Mean ± SD | 29.1 ± 42.3 | |
| Median (range) | 17.0 (13.0–402) | |
| Twitter mentions | ||
| No | 21 (17.8) | |
| Yes | 97 (82.2) | |
| Number of Twitter mentions | ||
| Mean ± SD | 12.2 ± 19.2 | |
| Median (range) | 5.50 (0–150) | |
| Facebook mentions | ||
| No | 78 (66.1) | |
| Yes | 40 (33.9) | |
| Number of Facebook mentions | ||
| Mean ± SD | 0.63 ± 1.24 | |
| Median (range) | 0 (0–7.00) | |
| Blog mentions | ||
| No | 105 (89.0) | |
| Yes | 13 (11.0) | |
| Number of blog mentions | ||
| Mean ± SD | 0.127 ± 0.404 | |
| Median (range) | 0 (0–3.00) | |
| News mentions | ||
| No | 32 (27.1) | |
| Yes | 86 (72.9) | |
| Number of news mentions | ||
| Mean ± SD | 2.89 ± 5.85 | |
| Median (range) | 1.00 (0–50.0) | |
| Mendeley readership | ||
| Mean ± SD | 91.8 ± 78.2 | |
| Median (range) | 77.5 (1.00–533) | |
| Policy mentions | ||
| No | 97 (82.2) | |
| Yes | 21 (17.8) | |
| Number of policy mentions | ||
| Mean ± SD | 0.229 ± 0.546 | |
| Median (range) | 0 (0–3.00) | |
| Patent mentions | ||
| No | 110 (93.2) | |
| Yes | 8 (6.8) | |
| Number of patent mentions | ||
| Mean ± SD | 0.32 ± 1.55 | |
| Median (range) | 0 (0–11.0) | |
| Google Plus mentions | ||
| No | 112 (94.9) | |
| Yes | 6 (5.1) | |
| Number of Google Plus mentions | ||
| Mean ± SD | 0.102 ± 0.605 | |
| Median (range) | 0 (0–6.00) | |
Values are presented as number (%) unless otherwise indicated.
SD: standard deviation.
The most common article type was original clinical research (82%), with the most common subgroup being retrospective cohort studies (n = 54; 56% of original research articles). The most common article topics were diagnostics/imaging (28%) and outcomes following TSA (47%). Of the included studies, 76% originated from within the USA, 19% originated from Europe, and 5% were published outside of the USA and Europe. A large portion of the included articles focused on both anatomic and reverse TSAs (62%).
A total of 1,436 Twitter mentions (median, 5.5; IQR, 1–21) and 74 Facebook mentions (median, 0; IQR, 0–1) were identified from the 118 included articles. A multivariable linear regression model incorporating all bibliometric characteristics collected from included studies was constructed to determine the influence of each factor on the AAS (Table 4). This model demonstrated presence of a conflict of interest and open access status was statistically significant and associated with the AAS, with an additional mean increase in the AAS of 17.17 (95% confidence interval [CI], 2.72–31.62; p = 0.042) and 3.61 (95% CI, 1.33–5.81; p < 0.01), respectively. The regression analysis demonstrated that the citation rate was not significantly associated with the AAS (p > 0.05).
Table 4. Multivariate Linear Regression Model for Key Publication Characteristics, Citations, and the Altmetric Attention Score.
| Variable | Regression coefficient | 95% Confidence interval | p-value | |
|---|---|---|---|---|
| Reference: subscription access | ||||
| Open Access | 3.61 | 1.33 to 5.81 | < 0.01* | |
| Reference: No general COI | ||||
| COI | 17.17 | 2.72 to 31.62 | 0.042* | |
| Reference: Non-MD/DO† | ||||
| MD or DO | –4.64 | –15.15 to 43.18 | 0.654 | |
| Reference: Not from North America | ||||
| Continent of origin: North America | 5.78 | –16.66 to 35.36 | 0.546 | |
| Reference: Before 2017 | ||||
| Year: after and including 2017 | 13.53 | –24.21 to 28.39 | 0.140 | |
| Number of authors | –3.39 | –8.24 to 3.53 | 0.106 | |
| Number of institutions | 2.22 | –1.53 to 19.78 | 0.335 | |
| Number of references | 0.22 | –0.29 to 0.79 | 0.417 | |
| Number of citations | –0.02 | –0.06 to 0.08 | 0.601 | |
COI: conflict of interest.
*Indicates statistical significance at p < 0.05 level. †Other author degrees include BS, MS, PhD, MD PhD, MD MBA, MD MS, MD MPH, and PT.
DISCUSSION
Our study showed that the 100 most impactful articles on shoulder arthroplasty came mainly out of Europe and the USA, were mostly original clinical research, and focused mostly on both anatomic and reverse TSAs. Our findings also showed that the JSES and the JBJS contributed the highest number of articles in our list and that the presence of conflicts of interest and open access status were positively associated with higher AAS impact.
The vast majority of articles in our study came out of the USA and Europe. That comes as no surprise, as the vast investments in research and healthcare seen in the West rendered these countries pioneers in surgical and medical research worldwide.17) As such, it was found that the most impactful articles on shoulder arthroplasty were conducted by American or European surgeons. Previous research by Kunze et al found that publications from North America were associated with higher AAS when looking at articles published in 5 English-language orthopedic journals.18) That being said, however, our manuscript did not find any significant associations between the country of origin and AAS, probably due to the fact that our study did not pose limitations on the journals included.
World-leading journals in the field of orthopedics often produce more impactful research and have high publication standards, and thus, it comes as no surprise that JSES and JBJS had the highest number of contributions from our list. JSES and JBJS are ranked 9th and 15th out of 306 journals in the field of Orthopedics and Sports Medicine, and are as such, considered elite journals, especially in the field of shoulder surgery.19) Social media promotion of such publications is often superior to those at low-tier journals and that could be due to the greater marketing prowess of such journals, the high standard of the articles published in them, and the high interest generated from their target audience. As a matter of fact, this trend has been mirrored by findings of other similar papers in the literature that explored social media interest in other orthopedics-related topics.11,13) For example, in an Altmetric analysis of top 100 cited papers on anterior cruciate ligament, Civilette et al.11) demonstrated that around two-thirds of the articles were published in the American Journal of Sports Medicine (AJSM), British Journal of Sports Medicine, and the Journal of Orthopaedic and Sports Physical Therapy.11) Another similar study by Kirloskar et al.13) explored medial ulnar collateral ligament and found that 72% of the articles were published in AJSM, JSES, or Orthopaedic Journal of Sports Medicine. Hence, toptier journals often have higher promotive force on social media, which can translate into higher AAS and social media interactions. That being said, even though journal IF, a highly utilized metric for measuring journal impact and significance, has been associated with AAS in other similar orthopedic studies,18,20) our study did not show any significant association.
The majority of articles in our study were clinical research studies (82%), and this falls in accordance with other similar studies in literature.11,13) Clinical research studies found in our study can provide valuable evidence-based information related to shoulder arthroplasty, like data on diagnostics and patient outcomes.21) Such studies can impact patient management and care and are thus bound to generate much more interest to their target audience than other article types like review articles or basic science, especially in the field of shoulder arthroplasty. Moreover, while citation rate has been found to be significantly associated with AAS for the general orthopedic literature and hand surgery, no association was found in our study.10,14,22) This can be explained by the fact that these studies excluded the articles published in the last 2.5–4 years in order to have more representative data on citation counts. Our study included all articles published up to the point of data collection, including 25 articles that were published within the last 3 years. As such, the more recent articles probably did not have enough time to accrue a fair number of citations so far, and this may have affected the association between citation count and AAS.
In our study, a significant association was found between AAS scores and open access status. This comes as no surprise, as open access articles are freely accessible and available to the scientific community, leading to higher readership and consequently higher interaction on social media platforms.23) In addition, the presence of conflict of interest among authors was also associated with higher AAS scores. This can be explained by the fact that conflicted authors may have their research promoted by their industry partners, and this can help expand the reach of the published study, leading to more social media interactions and higher AAS scores.
Our study is the first to explore the 100 shoulder arthroplasty publications with the highest AAS scores. Nevertheless, several limitations exist. First, there is an inherent bias for greater online attention for more recent publications, as social media platforms may not have been available for older articles. Second, several journals and authors use Instagram to promote their research online, but Instagram is not part of the AAS scoring system. Finally, while AAS can be an indicator of research impact and significance, it is important for the reader to know that AAS is more a measure of popularity rather than quality. Hence, readers should always be critical of the research at hand and not conflate AAS scores with quality scores.
Top articles on TSA, as defined by high AAS, mostly comprise original clinical research performed in the USA or Europe. The presence of a conflict of interest and open access status was associated with an increase in AAS, but there was no association between AAS score and citation rate.
Footnotes
CONFLICT OF INTEREST: JAA would like to disclose royalties from DJO Global, Zimmer-Biomet, Smith and Nephew, Stryker, Globus Medical, Inc.; research support as a PI from Lima Corporation-Italy, Orthofix, Arthrex, OREF; royalties, financial or material support from Wolters Kluwer; and board member/committee appointments for American Shoulder and Elbow Society, Pacira.
References
- 1.Hirsch JE. An index to quantify an individual's scientific research output. Proc Natl Acad Sci U S A. 2005;102(46):16569–16572. doi: 10.1073/pnas.0507655102. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Fenner M. Google Scholar citations, Researcher Profiles, and why we need an Open Bibliography. Front Matter; 2011. [Google Scholar]
- 3.Chapman CA, Bicca-Marques JC, Calvignac-Spencer S, et al. Games academics play and their consequences: how authorship, h-index and journal impact factors are shaping the future of academia. Proc Biol Sci. 2019;286(1916):20192047. doi: 10.1098/rspb.2019.2047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hicks D, Wouters P, Waltman L, de Rijcke S, Rafols I. Bibliometrics: The Leiden Manifesto for research metrics. Nature. 2015;520(7548):429–431. doi: 10.1038/520429a. [DOI] [PubMed] [Google Scholar]
- 5.Moorhead SA, Hazlett DE, Harrison L, Carroll JK, Irwin A, Hoving C. A new dimension of health care: systematic review of the uses, benefits, and limitations of social media for health communication. J Med Internet Res. 2013;15(4):e85. doi: 10.2196/jmir.1933. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Kunze KN, Vadhera A, Purbey R, Singh H, Kazarian GS, Chahla J. Infographics are more effective at increasing social media attention in comparison with Original Research articles: an Altmetrics-based analysis. Arthroscopy. 2021;37(8):2591–2597. doi: 10.1016/j.arthro.2021.03.056. [DOI] [PubMed] [Google Scholar]
- 7.Elmore SA. The Altmetric Attention Score: what does it mean and why should i care? Toxicol Pathol. 2018;46(3):252–255. doi: 10.1177/0192623318758294. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kwok R. Research impact: Altmetrics make their mark. Nature. 2013;500(7463):491–493. doi: 10.1038/nj7463-491a. [DOI] [PubMed] [Google Scholar]
- 9.Djulbegovic M, Kalahasty K, Watane A, Jabori SK, Al-Khersan H, Sridhar J. Correlation between Altmetric Attention Scores and citations for articles published in high-impact factor ophthalmology journals from 2018 to 2019. JAMA Ophthalmol. 2022;140(6):623–627. doi: 10.1001/jamaophthalmol.2022.0858. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kunze KN, Richardson M, Bernstein DN, Premkumar A, Piuzzi NS, McLawhorn AS. Altmetrics Attention Scores for randomized controlled trials in total joint arthroplasty are reflective of high scientific quality: an Altmetrics-based methodological quality and bias analysis. J Am Acad Orthop Surg Glob Res Rev. 2020;4(12):e20.00187. doi: 10.5435/JAAOSGlobal-D-20-00187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Civilette MD, Rate WR, Haislup BD, et al. The top 100 most impactful articles on the anterior cruciate ligament: an Altmetric analysis of online media. SAGE Open Med. 2022;10:20503121221111694. doi: 10.1177/20503121221111694. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Haislup BD, Rate WR, Civilette MD, Cohen AS, Bodendorfer BM, Gould HP. The 100 most impactful articles on the rotator cuff: an Altmetric analysis of online media. J Exp Orthop. 2022;9(1):92. doi: 10.1186/s40634-022-00530-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kirloskar KM, Civilette MD, Rate WR, et al. The 50 most impactful articles on the medial ulnar collateral ligament: an Altmetric analysis of online media. SAGE Open Med. 2022;10:20503121221129921. doi: 10.1177/20503121221129921. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Yoshimura R, Grant MC, Gardiner MD, Wade RG. Disseminating hand surgery research using social media: the relationship between Altmetrics and citations. J Hand Surg Am. 2021;46(9):740–747. doi: 10.1016/j.jhsa.2021.03.028. [DOI] [PubMed] [Google Scholar]
- 15.Craig RS, Lane JC, Carr AJ, Furniss D, Collins GS, Rees JL. Serious adverse events and lifetime risk of reoperation after elective shoulder replacement: population based cohort study using hospital episode statistics for England. BMJ. 2019;364:l298. doi: 10.1136/bmj.l298. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Cancienne JM, Brockmeier SF, Werner BC. Association of perioperative glycemic control with deep postoperative infection after shoulder arthroplasty in patients with diabetes. J Am Acad Orthop Surg. 2018;26(11):e238–e245. doi: 10.5435/JAAOS-D-16-00784. [DOI] [PubMed] [Google Scholar]
- 17.Philipson L. Medical research activities, funding, and creativity in Europe: comparison with research in the United States. JAMA. 2005;294(11):1394–1398. doi: 10.1001/jama.294.11.1394. [DOI] [PubMed] [Google Scholar]
- 18.Kunze KN, Polce EM, Vadhera A, et al. What is the predictive ability and academic impact of the Altmetrics Score and social media attention? Am J Sports Med. 2020;48(5):1056–1062. doi: 10.1177/0363546520903703. [DOI] [PubMed] [Google Scholar]
- 19.SJR. Scimago Journal & Country Rank [Internet] SJR; 2023. [cited 2023 Feb 1]. Available from: https://www.scimagojr.com/journalrank.php?category=2732 . [Google Scholar]
- 20.Kunze KN, Vadhera AS, Polce EM, et al. The Altmetric Attention Score is associated with citation rates and may reflect academic impact in the total joint arthroplasty literature. HSS J. 2023;19(1):37–43. doi: 10.1177/15563316221115723. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Keller RB, Rudicel SA, Liang MH. Outcomes research in orthopaedics. J Bone Joint Surg. 1993;75(10):1562–1574. [Google Scholar]
- 22.Khalid MA, Verma A, Kazan O, Chen J. Twitter use among orthopedic surgery journals correlates with increased citation rates. Orthopedics. 2023;46(1):e38–e44. doi: 10.3928/01477447-20221024-02. [DOI] [PubMed] [Google Scholar]
- 23.Davis PM. Open access, readership, citations: a randomized controlled trial of scientific journal publishing. FASEB J. 2011;25(7):2129–2134. doi: 10.1096/fj.11-183988. [DOI] [PubMed] [Google Scholar]