Statins and cognitive changes over time 

 
   Lead Investigator:    Kyle Steenland
   Institution      :    Emory
   E-Mail           :    nsteenl@emory.edu
   Proposal ID      :    207


   Publications:


1  Steenland K, Zhao L, Goldstein FC, Levey A
   Statins and cognitive decline in older adults with normal cognition or mild cognitive impairment
   Journal Article  J Am Geriatr Soc,61:1449-1455,2013,PMC3773248

   PUBLISHED





   Proposal Description:
   Recent reports have suggested memory problems resulting from statins. Although randomized trials have not noted any cognitive effects, sporadic reports of memory problems among statin users have prompted the FDA to issue a warning of possible memory problems to be added to sales of statins (NY Times, Feb 28, 2012). We propose to look at cognitive changes longitudinally in the NACC data base, looking at 10 cognitive tests over time, in relation to statin use. Subjects will be restricted to those diagnoses as 'normal', 'impaired not MCI', or 'MCI' at baseline. Covariates to be controlled include age, gender, race, education, depression, hypertension, and co-morbidities. Stroke patients will be excluded. We will consider whether there is an interaction between diabetes (another FDA noted side-effect) and statin use in cognitive changes. Depending on the frequencies of use, we will seek to compare 'always users' to 'never users' as the clearest comparison to detect any effects. We will take account of correlated data within a single person followed over time by specifying a correlation structure (eg, compound symmetry) for repeated measures on the same person (SAS PROC MIXED). We will run separate models for subjects with different baseline diagnoses. While our initial focus will be on analysis of cognitive tests as a continuous outcome, we may further explore the possible role of statins in progression from normal to MCI, or from MCI to AD, using Cox regression while controlling for relevant covariates.