“Big data” is the catchphrase of today. A simple search of titles in PubMed using the term big data yielded 1562 records as of February 28, 2018, of which 800 were published in the past 2 years and all were published within the past decade (2008-2018). As health care and research communities gain access to more complex and bigger data, it may help to remember key methodologic and statistical lessons gained from previous experience.
In this issue of JAMA Ophthalmology, Wu et al1 explore possible associations between a multitude of patient characteristics and receiving cataract surgery in 2 large patient populations: the US Medicare health insurance program and Veterans Health Administration. With more than 1 million patients in each data set, it is no surprise that nearly all associations investigated achieved statistically significant results with narrow 95% CIs. However, these statistically significant results require additional insight as to what may be meaningful. Thus, what can investigators using big data do to make us confident that these findings should be used to affect individual care or population-based policies?
Lindsley KB. Getting Back to Basics in a World of Data Overload: Characteristics Associated With Receipt of Cataract Surgery in 2 US Data Sets. JAMA Ophthalmol. 2018;136(7):745–746. doi:10.1001/jamaophthalmol.2018.1479
Customize your JAMA Network experience by selecting one or more topics from the list below.
Create a personal account or sign in to: