Postconcussion statin use may protect against dementia

  • Redelmeier DA & al.
  • JAMA Neurol
  • 20 May 2019

  • International Clinical Digest
Access to the full content of this site is available only to registered healthcare professionals. Access to the full content of this site is available only to registered healthcare professionals.


  • Among older adults sustaining a concussion, those with a statin prescription in the first several postconcussion months had a 13% lower risk of developing dementia.

Why this matters

  • Lack of effective neuroprotective agents after traumatic brain injury (TBI).

Key results

  • 24.5% of patients had a statin prescription within 90 days after concussion.
  • During mean 3.9-year follow-up, 16.4% of study cohort developed dementia.
  • Relative to nonstatin users, statin users had lower dementia risk after adjustment for factors including other cardiovascular medications and neuropsychiatric medications (relative risk, 0.87; P<.001>
  • Reduction was similar across subgroups, increased over time, was distinct from depression risk.
  • No significant protective effect of statins seen after ankle sprain or of nonstatin lipid-lowering drugs.
  • Only other cardiovascular medication class independently protective was angiotensin II receptor blockers (relative risk, 0.88; 95% CI, 0.79-0.98).

Expert comment

  • In an editorial, Rachel A. Whitmer, PhD, writes, “While this study cannot infer causality, it is a first step in accumulating evidence for possible future therapeutic interventions post-TBI and a call for further pharmacoepidemiologic harnessing of integrated health care systems to discover other potential protective or risk factors for dementia after brain injury.”

Study design

  • Canadian population-based cohort study of 28,815 older adults (median age, 76 years) with a concussion diagnosis, excluding those hospitalized because of severity, those with prior dementia or delirium, those dying within 90 days.
  • Main outcome: dementia.
  • Funding: Canada Research Chair in Medical Decision Sciences; Canadian Institutes of Health Research; others.


  • Potential unmeasured, residual confounding.
  • Lack of power to assess timing of statin prescription.
  • Possible underestimation of dementia.
  • Limited generalizability.