The Very Large Database of Lipids (VLDL) is a recently developed ongoing project designed to leverage industry derived data from a leading clinical laboratory for the independent use by academic investigators for high statistical power hypothesis generation, modeling, epidemiologic analyses, and data mining relevant to medicine. Data sets of this size and scope are not feasible to develop using conventional methods of grant funded prospectively developed epidemiologic cohort study and provide opportunities for discovery complementary to traditional approaches based on cohort studies.
In its initial harvest of data over the interval 2009-2011 the data set included extensive lipid and biomarker data from 1.3 million unique individuals, representing a sample of approximately 1:180 Americans. The second harvest is in progress now, representing 4.4 million unique individuals, tripling the size of the data set. With the second harvest (VLDL 2.0), we will have survival data on a subset of approximately 1.7 million individuals using CDC National Death Index matching.
Several high impact analyses have been published using the VLDL data which have allowed needed scrutiny of the Friedewald method routinely used across medical practice for estimation of serum levels of low density lipoprotein cholesterol (LDL-C) as well as subsequent development of a statistically robust novel method for high precision estimation of this clinical parameter. The VLDL project is registered on ClinicalTrials.gov and invites interested investigators to contact the VLDL lead investigators Steven Jones, firstname.lastname@example.org or Seth Martin, email@example.com concerning opportunities for collaboration or proposals for original research.
ClinicalTrials.gov registration: http://www.clinicaltrials.gov/ct2/show/NCT01698489?term=vldl&rank=1