Niranjan J Sathianathen1, Robert Lane2, Benjamin Condon3, Declan G Murphy4, Nathan Lawrentschuk5, Christopher J Weight2, Alastair D Lamb6. 1. Department of Urology, University of Minnesota, Minneapolis, MN, USA. Electronic address: niranjan19@gmail.com. 2. Department of Urology, University of Minnesota, Minneapolis, MN, USA. 3. Department of Urology, University of Minnesota, Minneapolis, MN, USA; Department of Surgical Oncology, Peter MacCallum Cancer Centre, Melbourne, Australia; Sir Peter MacCallum Department of Oncology, University of Melbourne, Parkville, Australia; Department of Surgery, Austin Health, University of Melbourne, Parkville, Australia; Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK; Department of Urology, Churchill Hospital Cancer Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, UK. 4. Department of Surgical Oncology, Peter MacCallum Cancer Centre, Melbourne, Australia; Sir Peter MacCallum Department of Oncology, University of Melbourne, Parkville, Australia. 5. Department of Surgical Oncology, Peter MacCallum Cancer Centre, Melbourne, Australia; Department of Surgery, Austin Health, University of Melbourne, Parkville, Australia. 6. Nuffield Department of Surgical Sciences, University of Oxford, Oxford, UK; Department of Urology, Churchill Hospital Cancer Centre, Oxford University Hospitals NHS Foundation Trust, Oxford, UK.
Abstract
BACKGROUND: The scientific impact of published articles has traditionally been measured as citation counts. However, there has been a shift in academia to a digitalized age in which research is widely read, disseminated, and discussed online. As part of this shift, each published article has a digital footprint. OBJECTIVE: To develop a urology social media score (#UroSoMe_Score) to predict citation counts from measures of online attention for urological articles. DESIGN, SETTING, AND PARTICIPANTS: We included articles published between June 2016 and June 2017 in the top ten highest-impact urology journals. We obtained data on the online attention received by each of these articles from Altmetric Explorer and 2-yr citation counts from Scopus. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: We created a multivariable linear model using the forward stepwise regression method based on the Akaike information criterion to determine the best-fitting model using online sources of attention to predict 2-yr citation count. RESULTS AND LIMITATIONS: We included a total of 2033 urology articles. The median weighted Altmetric score for the articles included was 4 (interquartile range [IQR] 2-11). The median number of citations for all articles included was 7 (IQR 3-14). There was an association between Altmetric score and 2-yr Scopus citation count (p < 0.001) but the adjusted R2 value for this model was only 0.013. Our stepwise regression model revealed that citations could be predicted from a model comprising the following sources of online attention: policy documents, Google+, blogs, videos, Wikipedia, Twitter, and Q&A. The adjusted R2 value for the #UroSoMe_Score model was 0.14, which is superior to the full Altmetric score. CONCLUSIONS: The #UroSoMe_Score can be used to predict 2-yr citation counts for urological publications on the basis of online metrics. PATIENT SUMMARY: Online measures of attention can be used to predict citation counts and thus the scientific impact of an article. Our #UroSoMe_Score can be used in such a manner specifically for the urological literature. Outliers may still be present especially for popular topics that receive online attention but are not heavily cited.
BACKGROUND: The scientific impact of published articles has traditionally been measured as citation counts. However, there has been a shift in academia to a digitalized age in which research is widely read, disseminated, and discussed online. As part of this shift, each published article has a digital footprint. OBJECTIVE: To develop a urology social media score (#UroSoMe_Score) to predict citation counts from measures of online attention for urological articles. DESIGN, SETTING, AND PARTICIPANTS: We included articles published between June 2016 and June 2017 in the top ten highest-impact urology journals. We obtained data on the online attention received by each of these articles from Altmetric Explorer and 2-yr citation counts from Scopus. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS: We created a multivariable linear model using the forward stepwise regression method based on the Akaike information criterion to determine the best-fitting model using online sources of attention to predict 2-yr citation count. RESULTS AND LIMITATIONS: We included a total of 2033 urology articles. The median weighted Altmetric score for the articles included was 4 (interquartile range [IQR] 2-11). The median number of citations for all articles included was 7 (IQR 3-14). There was an association between Altmetric score and 2-yr Scopus citation count (p < 0.001) but the adjusted R2 value for this model was only 0.013. Our stepwise regression model revealed that citations could be predicted from a model comprising the following sources of online attention: policy documents, Google+, blogs, videos, Wikipedia, Twitter, and Q&A. The adjusted R2 value for the #UroSoMe_Score model was 0.14, which is superior to the full Altmetric score. CONCLUSIONS: The #UroSoMe_Score can be used to predict 2-yr citation counts for urological publications on the basis of online metrics. PATIENT SUMMARY: Online measures of attention can be used to predict citation counts and thus the scientific impact of an article. Our #UroSoMe_Score can be used in such a manner specifically for the urological literature. Outliers may still be present especially for popular topics that receive online attention but are not heavily cited.