Answers are grounded in this paper and cite their sources.
There are several problems unique to large data sets. Large amounts of biased data are still biased and clinical significance is not always the same as statistical significance. Large number of predictors of outcome can confound conclusions, but there are several ways to manage wide ranging data sets including matching, regression, propensity scores, and randomization.
No takes yet. Share an insight, caveat, or question.
Mitchell Maltenfort (2015) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: