Confusion Assessment Method (CAM) for Delirium Assessment
Core Diagnostic Algorithm
The CAM requires the presence of Feature 1 (acute onset and fluctuating course) AND Feature 2 (inattention) AND either Feature 3 (disorganized thinking) OR Feature 4 (altered level of consciousness) to diagnose delirium. 1, 2
The four cardinal features are:
- Feature 1: Acute onset and fluctuating course - Document when symptoms began and whether they vary throughout the day 2, 3
- Feature 2: Inattention - Patient has difficulty focusing attention, is easily distractible, or has difficulty keeping track of what is being said 2, 3
- Feature 3: Disorganized thinking - Conversation is rambling or irrelevant, unclear or illogical flow of ideas, or unpredictable switching between subjects 2, 3
- Feature 4: Altered level of consciousness - Any level other than alert (vigilant, lethargic, stupor, or coma) 2, 3
Practical Bedside Assessment Steps
Obtain collateral history first - Before examining the patient, speak with family or nursing staff to establish baseline cognitive function and the exact timeline of mental status changes 1, 4
Administer brief cognitive tests including:
- Months of the year backwards (MOTYB) to assess attention 2, 1
- Short Orientation Memory Concentration Test (SOMCT) for orientation and memory 2, 1
Assess each CAM feature systematically - The complete assessment can be performed in less than 5 minutes by trained staff 1, 3
Repeat assessments every 8-12 hours - Given the fluctuating nature of delirium, screening should occur at least once per nursing shift 1
Diagnostic Performance
The CAM demonstrates sensitivity of 82-100% and specificity of 89-99% when compared to psychiatrist diagnosis using DSM criteria 1, 3, 5
Critical Clinical Pitfalls
Hypoactive delirium is frequently missed - This is the most common subtype, particularly in cancer patients and older adults, characterized by psychomotor slowing and sedated appearance rather than agitation 1, 4
The CAM performs best as a screening tool - When CAM is positive, confirm the diagnosis using formal DSM-IV criteria, as specificity can be lower in real-world settings (63-84%) compared to research conditions 5
Training is essential - Standardized training programs with supervised practice sessions and ongoing inter-rater reliability checks are necessary to maintain diagnostic accuracy across multiple assessors 6
Bedside nursing assessments may underdiagnose delirium - Studies show routine CAM-ICU use by bedside nurses without intensive training may detect fewer cases than unstructured clinical assessments by experienced ICU nurses 7