TY - JOUR
T1 - Editorial Commentary
T2 - Sometimes You Don't Know What You've Got Until It's Gone—The Effect of Missing Data in “Big Data” Studies
AU - Dhawan, Aman
N1 - Funding Information:
The author reports the following potential conflicts of interest or sources of funding: A.D. is a consultant for Smith & Nephew and Avenue Therapeutics; is supported by grants or pending grants from Revotek, Department of Defense , National Institutes of Health , and Penn State University; receives payment for lectures including service on speakers’ bureaus from Smith & Nephew; is Associate Editor of Arthroscopy; is a committee member of the American Orthopaedic Society for Sports Medicine Publications Committee and Arthroscopy Association of North America Research Committee; and is on the editorial boards of the Orthopaedic Journal of Sports Medicine and Sports Medicine and Arthroscopy Review, outside the submitted work. Full ICMJE author disclosure forms are available for this article online, as supplementary material .
Publisher Copyright:
© 2020 Arthroscopy Association of North America
PY - 2020/5
Y1 - 2020/5
N2 - Big-data studies are powerful tools for comparative-effectiveness research, but because of the large number of included patients, they risk falsely identifying a difference when none exists because large sample sizes may result in statistically significant differences that have little clinical importance. Other limitations of big-data studies include lack of generalizability because of inclusion of only specific patient populations, lack of validated outcome measures, recording bias or clerical error, and vast troves of missing data. As such, the methods and results of big-data studies require careful scrutiny to ensure that the conclusions are correct.
AB - Big-data studies are powerful tools for comparative-effectiveness research, but because of the large number of included patients, they risk falsely identifying a difference when none exists because large sample sizes may result in statistically significant differences that have little clinical importance. Other limitations of big-data studies include lack of generalizability because of inclusion of only specific patient populations, lack of validated outcome measures, recording bias or clerical error, and vast troves of missing data. As such, the methods and results of big-data studies require careful scrutiny to ensure that the conclusions are correct.
UR - http://www.scopus.com/inward/record.url?scp=85083883201&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85083883201&partnerID=8YFLogxK
U2 - 10.1016/j.arthro.2020.02.024
DO - 10.1016/j.arthro.2020.02.024
M3 - Editorial
C2 - 32370886
AN - SCOPUS:85083883201
VL - 36
SP - 1240
EP - 1242
JO - Arthroscopy - Journal of Arthroscopic and Related Surgery
JF - Arthroscopy - Journal of Arthroscopic and Related Surgery
SN - 0749-8063
IS - 5
ER -