Medication Review and Deprescribing for Dementia Patients with Falls
In older adults with dementia experiencing falls, immediately review and deprescribe CNS-active medications—particularly psychotropics, antipsychotics, benzodiazepines, and antidepressants—as these medications substantially increase fall risk without reducing it through anti-dementia drugs.
Priority Medication Changes
Medications to Deprescribe or Minimize
The AGS Beers Criteria 1 explicitly recommends avoiding concurrent use of three or more CNS-active agents (antidepressants, antipsychotics, benzodiazepines, nonbenzodiazepine hypnotics, antiepileptics, and opioids) due to increased fall risk. Psychotropic medications increase fall rate by 41% (IRR 1.41) in community-dwelling adults with cognitive impairment 2.
Specific high-risk medications to target:
- Antipsychotics: Should be avoided due to increased mortality risk in older adults 1. Discontinuation rates of 44% were achieved safely in recent deprescribing studies 3
- Antidepressants: Increase fall rate by 54% (IRR 1.54), though this association is mediated by depressive symptoms themselves 2. Consider non-pharmacological management of depression first
- Benzodiazepines: High-priority for discontinuation per Beers Criteria 1, with 33% discontinuation rates demonstrated as feasible 3
- Tricyclic antidepressants: 55% discontinuation rate achieved safely 3
- Opioids: Particularly when combined with benzodiazepines or gabapentinoids 1
Medications That Do NOT Reduce Falls
Anti-dementia medications (cholinesterase inhibitors, memantine) do not reduce fall risk 2, so do not expect fall prevention benefit from these agents. However, the Beers Criteria notes that acetylcholinesterase inhibitors (AChEIs) themselves carry fall risk 1.
Systematic Medication Review Process
Identify all CNS-active medications: Count total number—if ≥3, prioritize immediate reduction 1
Assess each medication's risk-benefit:
Deprescribe systematically: Recent evidence shows 43.8% antipsychotic discontinuation, 54.5% tricyclic discontinuation, and 66.7% muscle relaxant discontinuation are achievable 3
Non-Pharmacological Interventions (Essential Component)
The USPSTF 5 provides a B-grade recommendation for exercise interventions in community-dwelling older adults at increased risk for falls. This is the strongest evidence-based intervention.
Specific Exercise Components
- Balance training
- Strength training
- Gait training
- Physical therapy referral for those with impaired gait speed, poor balance, or reduced lower limb function 2
Multifactorial Risk Assessment
The AGS recommends 5:
- Home environment modification
- Management of postural hypotension
- Foot problems and footwear assessment
- Functional assessment
- Environmental assessment
Critical Caveats
Depression management paradox: While antidepressants increase fall risk, untreated depressive symptoms also increase falls 2. Prioritize non-pharmacological approaches for depression (cognitive behavioral therapy, activity scheduling, social engagement) before pharmacological treatment.
Vitamin D: The USPSTF now recommends AGAINST vitamin D supplementation for fall prevention in community-dwelling older adults without known deficiency 5—this reversed their 2012 recommendation.
Dementia subtype matters: Fall incidence and predictors vary significantly by dementia type 6. Vascular dementia and mixed dementia show particularly high fall rates with impaired gait and ADL. Dementia with Lewy bodies requires specific attention to fluctuations and depression.
Implementation Algorithm
- Immediate: Review medication list for CNS-active agents—target ≥3 concurrent agents
- Within 1 week: Initiate deprescribing of highest-risk medications (antipsychotics, benzodiazepines, tricyclics)
- Within 2 weeks: Refer to physical therapy for exercise-based fall prevention
- Ongoing: Address depression with non-pharmacological approaches first; conduct home safety assessment; monitor for medication withdrawal effects (none to minimal expected based on evidence 3, 4)
Safety note: Deprescribing interventions show reduction or no change in mortality and hospital attendance, demonstrating safety 4. Medically treated falls remained stable (22% vs 21%) in deprescribing studies 3.