Abstract
This study compared Altmetric scores of neurological research as a marker of “buzz” with citations as a marker of academic attention.
The omnipresence of social media in modern life is undeniable but its value to society remains to be determined; thus, we aimed to investigate the recognition received by neurological research on social media. We used the Altmetric attention score (AAS), a weighted count of online attention received by an article, as a marker of the popularity (“buzz”) of neurological research in the public and compared it with the academic attention marker (citations).
Methods
The impact factor–based top 5 journals that publish research on all neurological subspecialties were identified using the Journal Citations Reports for 2016 (Table). All original research articles (no opinion/editorial and single-case reports/neuroimages) published in 2016 in these journals were extracted and tabulated. Dimensions (Digital Science & Research Solutions Inc), an online searchable platform that aggregates data on more than 99 million publications,1 was used to identify the AAS of the included publications (accessed from April 1, 2019 to April 7, 2019), their individual engagement on various social media platforms, and their open access status. The total citations accrued over the 3-year period was obtained through Web of Science2 (accessed from April 1, 2019 to April 7, 2019). Abstracts and full-text articles were reviewed to categorize the neurological subspecialty, article topic, and study design. The top 100 publications with the highest AAS were considered as receiving the highest recognition on social media (“buzz”) and analyzed separately. Institutional review board approval was waived because institutional policies did not consider this work as human research.
Table. Characteristics of All the Research Articles and the Top 100 AAS Articles Published in 2016.
| Characteristic | All Articles (N = 1050) | Top 100 Articles (based on AAS) |
|---|---|---|
| Journals, No. (%) | ||
| Neurology | 512 (48.8) | 45 |
| JAMA Neurology | 120 (11.4) | 28 |
| Brain | 215 (20.5) | 13 |
| The Lancet Neurology | 47 (4.5) | 9 |
| Annals of Neurology | 156 (14.9) | 5 |
| Altmetric score, median (IQR) | 18 (9-40) | 269 (197-441) |
| Citations, median (IQR) | 20 (11-35) | 33 (18-54) |
| Open access, median (%) | 769 (73.2) | 88 (88) |
| Social media mentions, median (IQR) | ||
| 12 (5-22) | 41 (21-75) | |
| News outlet | 1 (0-3) | 30 (21-46) |
| 1 (0-2) | 5 (2-10) | |
| Read on Mendeley | 40 (24-66) | 59 (39-111) |
| Neurological subspecialty, No. (%) | ||
| Cognitive neurology | 187 (17.8) | 27 |
| Cerebrovascular neurology | 189 (18.0) | 12 |
| Neuroimmunology | 123 (11.7) | 11 |
| Movement disorders | 128 (12.2) | 9 |
| Neuromuscular | 127 (12.1) | 9 |
| Epilepsy | 97 (9.2) | 6 |
| Neurotrauma | 23 (2.2) | 7 |
| Headache | 25 (2.4) | 4 |
| Pediatric | 36 (3.4) | 1 |
| Miscellaneous | 115 (11.0) | 14 |
| Study design, No. (%) | ||
| Observational studies | 895 (85.2) | 83 |
| Clinical trials | 63 (6.0) | 11 |
| Basic research | 92 (8.8) | 6 |
| Article topic, No. (%) | ||
| Neurocognition | 58 (5.5) | 14 |
| Lifestyle | 25 (2.4) | 14 |
| Pharmacological therapies | 96 (9.1) | 13 |
| Genetics | 126 (12.0) | 9 |
| Natural history of disease | 137 (13.0) | 7 |
| Non-pharmacological therapies | 58 (5.5) | 7 |
| Etiopathogenesis | 85(8.1) | 1 |
| Diagnosis | 90 (8.6) | 3 |
| Imaging | 100 (9.5) | 0 |
| Outcomes | 75 (7.1) | 2 |
| Epidemiology | 46 (4.4) | 9 |
| Brain networks | 45 (4.3) | 6 |
| Others/miscellaneous | 109 (10.4) | 15 |
Abbreviations: AAS, Altmetric attention score; IQR, interquartile range.
Categorical variables were described using percentages and continuous variables using medians and interquartile ranges (IQRs). The Mann-Whitney U test and the Spearman rank correlation coefficient were used for assessing the correlation of AAS with the open access status and the number of citations, respectively. The statistical analysis was performed using SPSS, version 20 (IBM Corporation).
Results
A total of 1050 articles were included in the final analysis. The median AAS was 18 (IQR, 9-40) and the median citations were 20 (IQR, 11-35) (Table). Of these, 493 articles (47%) achieved an AAS of 20 or higher. Open access was available for 1050 articles (73.2%), which was significantly associated with a higher AAS (median, 20 vs 15; P < .01). Among the articles with the top 100 AAS (JAMA Neurology = 28/120 [23.3%]; The Lancet Neurology = 9/47 [19.1%]; Neurology = 45/512 [8.9%]; Brain = 13/215 [6.0%]; and Annals of Neurology = 5/156 [3.2%]; Table), 27 belonged to cognitive neurology. The most talked about research topics included neurocognition along with lifestyle changes affecting brain health. There was a significant (P < .01), but weak correlation (r = 0.32) between the AAS and the number of citations garnered by all 1050 research articles over 3 years (Figure).
Figure. Correlation Between Altmetric Score and Number of Citations of All Research Articles Published in 2016 on a Nonlinear Scale.
Each point represents 1 of the 1050 articles published (r = 0.32; P < .01).
Discussion
There is a high degree of engagement with neurological research on social media. The median AAS for all the research articles combined was 18 (IQR, 9-40). For context, an AAS of 20 or higher corresponds to the top 5% of the entire research output3 (achieved by close to half of these articles). A similar study of cardiovascular research articles published in 2014 found a median AAS of 8 (IQR, 2-37).4 One of 4 articles that achieved a top 100 AAS came from the cognitive neurology field and the topic that gained the maximum attention in this subset was also neurocognition. This high public interest suggests possible social anxiety about the coming “dementia epidemic.”5 Unlike other specialties,3 open access status seems to correlate with a higher AAS in neurological research. There is a weak correlation between the attention generated on social media and the citations accrued by the articles over a 3-year period, which is similar to the association noted in cardiovascular research.4 The subspecialty, interspecialty, or journal-based differences in the proportion of articles achieving a high AAS or their association with citations may depend on several factors, including the online engagement of the individual members of a field and the efforts at the journal level to promote their output.
Citations are accrued over long periods and this may be a limitation of this study. However, the effect of a weak alignment in the perceptions about research among the public and academia deserves further investigation, along with the eventual effect of the “buzz” on social media about neurological research on clinical outcomes.
References
- 1.Digital Science & Research Solutions Inc Re-imagining discovery and access to research: grants, publications, citations, clinical trials, patents and policy documents in one place. https://www.dimensions.ai/. Accessed April 1, 2019.
- 2.Clarivate Analytics Web of Science: trust the difference. https://clarivate.com/products/web-of-science/. Accessed April 1, 2019.
- 3.Patel RB, Vaduganathan M, Bhatt DL, Bonow RO. Characterizing high-performing articles by Altmetric score in major cardiovascular journals [published online November 21, 2018]. JAMA Cardiol. doi: 10.1001/jamacardio.2018.3823 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Barakat AF, Nimri N, Shokr M, et al. . Correlation of Altmetric attention score with article citations in cardiovascular research. J Am Coll Cardiol. 2018;72(8):952-953. doi: 10.1016/j.jacc.2018.05.062 [DOI] [PubMed] [Google Scholar]
- 5.Larson EB, Yaffe K, Langa KM. New insights into the dementia epidemic. N Engl J Med. 2013;369(24):2275-2277. doi: 10.1056/NEJMp1311405 [DOI] [PMC free article] [PubMed] [Google Scholar]