Premium
Statins in the prevention of dementia and Alzheimer's disease: A meta‐analysis of observational studies and an assessment of confounding
Author(s) -
Wong William B.,
Lin Vincent W.,
Boudreau Denise,
Devine Emily Beth
Publication year - 2013
Publication title -
pharmacoepidemiology and drug safety
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.023
H-Index - 96
eISSN - 1099-1557
pISSN - 1053-8569
DOI - 10.1002/pds.3381
Subject(s) - confounding , medicine , observational study , dementia , meta analysis , statin , relative risk , randomized controlled trial , disease , alzheimer's disease , random effects model , publication bias , pharmacoepidemiology , pharmacology , confidence interval , medical prescription
Background Studies demonstrate the potential for statins to prevent dementia and Alzheimer's disease (AD), but the evidence is inconclusive. Objective Conduct a meta‐analysis to estimate any benefit of statins in preventing dementia and examine the potential effect of study design and confounding on the benefit of statins in dementia. A secondary goal is to explore factors that may elucidate the mechanisms by which statins exert their potentially beneficial effect. Methods Performed systematic literature review to identify relevant publications. Relative risk (RR) estimates were pooled using both fixed and random effect models. Studies were stratified by study design and potential confounding factors. Results The pooled results for all‐type dementia suggest that use of statins is associated with a lower RR of dementia when compared to non‐statin users (random effects model: RR 0.82 (95%CI [0.69, 0.97]). The pooled results for AD also suggested a lower RR with statin user compared to non‐statin users in random effects models (RR: 0.70, 95% CI [0.60, 0.83]). Study design and methods used to address biases may influence the results. Conclusion These pooled results suggest that statins may provide a slight benefit in the prevention of AD and all‐type dementia. This benefit observed in both disease states should be interpreted with caution as observational studies are subject to bias, and it is possible that the slight benefit observed may disappear when these biases are addressed in a well‐designed randomized controlled trial. Copyright © 2012 John Wiley & Sons, Ltd.