Benzodiazepines and Dementia: The Evidence
Benzodiazepines are associated with cognitive impairment and functional decline in older adults, but the evidence linking them to true dementia risk is confounded by reverse causality—meaning the anxiety, insomnia, and prodromal dementia symptoms that prompt benzodiazepine prescribing likely explain most of the observed association rather than the drugs themselves causing dementia. 1
What the Guidelines Tell Us
The 2021 Mayo Clinic guidelines are unequivocal: benzodiazepines cause cognitive impairment, reduced mobility, unsafe driving, functional decline, falls, fractures, and addiction in older adults 1. Current consensus guidelines advise benzodiazepine use solely on a short-term basis 1. The 2019 AGS Beers Criteria classify high-potency, long-acting, or prolonged benzodiazepine use as high-risk in older adults 2.
Critically, these guidelines emphasize that benzodiazepines cause acute and ongoing cognitive impairment—this is established fact, not controversial 1. The question is whether they cause permanent dementia.
The Research Evidence: A Confusing Picture That Clears Up
The Most Recent and Highest Quality Studies (2024-2025)
The 2025 Canadian case-control study found benzodiazepine exposure associated with dementia (OR 1.65), with higher risk for long half-life agents (OR 2.81). However, the critical finding: chronic use was linked to dementia only within four years before diagnosis, strongly suggesting confounding by indication or reverse causality 3. This means people were likely prescribed benzodiazepines because they had early dementia symptoms (anxiety, sleep problems), not that benzodiazepines caused the dementia.
The 2025 Manitoba retrospective cohort study examined up to 22 years of follow-up and found that when properly controlling for confounding through restriction to persons with depression/anxiety or using high-dimensional propensity scores, the association between benzodiazepines and dementia disappeared entirely 4. The modest increase seen in crude analyses (aHR 1.33 in youngest group) was driven by higher rates of diabetes, cardiovascular disease, depression, and anxiety among users.
The 2025 systematic review and meta-analysis of chronic benzodiazepine use found no significant association with dementia (HR 1.17,95% CI: 0.96-1.43) or Alzheimer's disease specifically (HR 1.00,95% CI: 0.87-1.15) 5.
Supporting Evidence from 2024
The 2024 Rotterdam Study (population-based, 5,443 participants, 11.2 years follow-up) found overall benzodiazepine use was not associated with dementia risk (HR 1.06,95% CI: 0.90-1.25). However, anxiolytics at high cumulative doses showed some association (HR 1.33), while sedative-hypnotics did not (HR 0.92). Importantly, current benzodiazepine use was associated with smaller hippocampal and amygdala volumes and accelerated hippocampal atrophy on MRI 6.
The 2024 community-based cohort study found benzodiazepine use associated with developing MCI (not dementia), suggesting benzodiazepines may be a risk factor for mild cognitive impairment but not progression to dementia 7.
The Pattern That Emerges
When studies use longer lag periods (5+ years between exposure and outcome) and better control for confounding (especially psychiatric conditions that prompt benzodiazepine prescribing), the dementia association weakens or disappears 4, 5. When associations persist, they're strongest in the prodromal period (2-4 years before diagnosis) when early dementia symptoms likely prompted prescribing 3.
Clinical Bottom Line
What We Know for Certain:
- Benzodiazepines cause reversible cognitive impairment while being taken 1
- They cause falls, fractures, functional decline, and increased mortality in older adults 1, 8
- Long half-life benzodiazepines and Z-drugs appear riskier than shorter-acting agents 3, 9
- They show structural brain changes (hippocampal atrophy) on imaging 6
What Remains Uncertain:
- Whether benzodiazepines cause permanent, irreversible dementia after discontinuation
- The evidence suggests they probably don't, with observed associations explained by confounding and reverse causality 4, 5
Practical Recommendations
Avoid benzodiazepines in older adults whenever possible 1, 2. The established harms (falls, cognitive impairment, functional decline) are reason enough, regardless of dementia risk.
When Benzodiazepines Must Be Used:
- Use short-term only (consensus guideline recommendation) 1
- Prefer short-to-intermediate half-life agents over long-acting ones 3, 9
- Avoid cumulative doses >180 defined daily doses 9
- Women appear at higher risk for adverse cognitive effects 9
Deprescribing Strategy:
- Gradual taper over many weeks, not abrupt discontinuation 1
- Consider 25% dose reduction weekly, though slower may be needed 1
- Use patient education about risks (EMPOWER trial approach: 27% vs 5% discontinuation success) 1
- Substitute with safer alternatives: antidepressants, CBT, psychotherapy, or integrative approaches 1
Key Caveat
The 2025 hospice study found benzodiazepine initiation associated with 41% increased 180-day mortality in nursing home residents with dementia (HR 1.41) 8. This reinforces that even if benzodiazepines don't cause dementia, they substantially increase mortality risk in patients who already have it.