How should the Cohen‑Mansfield Agitation Inventory be used to assess and guide treatment of agitation and psychosis in an older adult with dementia?

The Cohen-Mansfield Agitation Inventory (CMAI) in Dementia-Related Psychosis

The CMAI should be used as a quantitative assessment tool to measure agitation severity and track treatment response, with a clinically meaningful reduction of 20 points (range: 15-25 points) on the total score indicating significant improvement, but treatment decisions must prioritize nonpharmacologic interventions first, reserving antipsychotics only for severe, dangerous agitation or psychosis that fails behavioral approaches. 1

Understanding the CMAI as an Assessment Tool

The CMAI quantifies the frequency of agitation behaviors in elderly persons with dementia, with scores ranging from 29-203 (higher scores indicate more frequent agitation behaviors). 1 Recent research has established that:

• For minimal clinically important difference (MCID): A reduction of 4 points over shorter time periods (less than 1 month) or 11 points over 1-3 months represents minimal symptom improvement 2
• For meaningful within-patient change (MWPC): A reduction of 20 points (with threshold range of 15-25 points) represents clinically significant improvement over 12 weeks 1
• Severity matters: Greater baseline agitation severity correlates with higher MCID estimates, meaning more severely agitated patients require larger score reductions to demonstrate meaningful benefit 2

Treatment Algorithm Guided by CMAI Assessment

Step 1: Initial CMAI Assessment and Nonpharmacologic Intervention (ALWAYS FIRST)

Nonpharmacologic approaches must be prioritized as first-line treatment because antipsychotics and anxiolytics pose significant risks including respiratory depression, dysrhythmia, hypotension, and delirium exacerbation. 3

Implement these specific interventions:

• Person-centered communication and sensory therapy 3
• Behavior modification, scheduled toileting, and prompted voiding for agitation related to incontinence 4
• Low lighting levels, music, and simulated nature sounds to create calming environments 4
• Group cognitive stimulation therapy for mild-to-moderate dementia 5
• Psychoeducational interventions for caregivers to reduce caregiver distress (which directly impacts patient agitation) 5

Reassess CMAI at 2-4 weeks. If CMAI reduction is less than 4 points and dangerous behaviors persist, proceed to Step 2.

Step 2: Pharmacologic Intervention When Nonpharmacologic Approaches Fail

Only proceed to medications when:

• Dangerous agitation or psychosis threatens patient or staff safety 6
• Nonpharmacologic interventions have demonstrably failed 3
• The goal is minimal sedation allowing patient participation in care, NOT heavy sedation 3

For Agitation Without Prominent Psychosis:

First choice: Atypical antipsychotics (specifically quetiapine or risperidone)

• Quetiapine: Start 12.5 mg twice daily, maximum 200 mg twice daily. Quetiapine had the lowest frequency of adverse events (OR 0.27 compared to haloperidol) in emergency settings 3 and is generally well-tolerated 7
• Risperidone: Start 0.25 mg daily at bedtime, maximum 2-3 mg daily in divided doses. Use low dosages as extrapyramidal symptoms may occur at 2 mg daily 7. Risperidone reduces CMAI total scores by approximately 11.8 points versus 6.4 for placebo (p<0.001) 8

Atypical antipsychotics probably reduce agitation slightly (SMD -0.21) but increase risks: somnolence (RR 1.93), extrapyramidal symptoms (RR 1.39), serious adverse events (RR 1.32), and death (RR 1.36). 9

Alternative: SSRIs (citalopram or sertraline)

• Citalopram: Reduces agitation as measured by CMAI (MD -0.89) with better tolerability than typical antipsychotics 10
• Consider when antipsychotics are contraindicated or poorly tolerated 10

AVOID benzodiazepines: Midazolam poses excessive risk (OR 5.25 for adverse events versus haloperidol) with 53% adverse event rate. 3 Benzodiazepines can cause tolerance, addiction, depression, cognitive impairment, and paradoxical agitation in 10% of patients. 7

For Prominent Psychosis (Delusions/Hallucinations):

Atypical antipsychotics are indicated but have negligible effect on psychosis (SMD -0.11) 9

• Risperidone or olanzapine for control of problematic delusions, hallucinations, and severe psychomotor agitation 7
• Olanzapine: Start 2.5 mg daily at bedtime, maximum 10 mg daily in divided doses, generally well-tolerated 7

Typical antipsychotics (haloperidol) may improve psychosis slightly (SMD -0.29) but should be second-line due to significant extrapyramidal symptoms (RR 2.26), risk of tardive dyskinesia (50% after 2 years continuous use), and anticholinergic/cardiovascular side effects. 7, 9

Step 3: CMAI Monitoring and Treatment Adjustment

Reassess CMAI at 4 weeks, 8 weeks, and 12 weeks:

• If CMAI reduction <15 points at 12 weeks: Treatment is not providing meaningful benefit; consider discontinuation or alternative approach 1
• If CMAI reduction 15-25 points: Borderline meaningful benefit; assess caregiver distress/time resources impact 2
• If CMAI reduction ≥20 points: Clinically meaningful improvement achieved 1

Critical consideration: Treatment benefit occurring after 12 weeks or having no impact on caregiver/staff distress or time resources is not considered worthwhile by expert consensus. 2

Step 4: Route of Administration Matters for Safety

Oral administration is strongly preferred:

• Oral medications (particularly quetiapine) have the best safety profile 3
• Intravenous and intramuscular routes increase adverse drug event risk compared to oral administration 3

In emergency settings requiring rapid intervention:

• Avoid midazolam entirely 3
• Consider oral quetiapine or low-dose haloperidol if oral route feasible 3
• Dose-response relationship exists: escalating doses of lorazepam, droperidol, and quetiapine increase adverse event risk 3

Critical Pitfalls to Avoid

Do not use CMAI scores alone to justify antipsychotic initiation without documenting failed nonpharmacologic interventions and dangerous behaviors. 6

Do not continue antipsychotics indefinitely without reassessing necessity; benefits are small (at best) and risks are substantial including mortality. 6, 9

Do not ignore caregiver burden assessment: Treatment meaningfulness depends heavily on whether interventions reduce caregiver distress and time resources, not just CMAI scores. 2

Do not use typical antipsychotics as first-line given 50% risk of tardive dyskinesia after 2 years and significant extrapyramidal symptoms. 7

Do not prescribe anticholinergic medications (benztropine, trihexyphenidyl) for extrapyramidal symptoms in this population; instead reduce antipsychotic dose or switch agents. 7, 5