Dementia Screening and Diagnosis
Introduction to Screening Tools
- A Mini-Cog score of 4 out of 5 is above the threshold for concern, as scores less than 3 are considered concerning for possible dementia, according to the Mayo Clinic Proceedings 1, 2
- The Mini-Cog has 76% sensitivity and 89% specificity for identifying dementia, meaning it can miss approximately 24% of dementia cases, as reported by the Mayo Clinic Proceedings 2, 3
Interpreting Screening Results
- Screening tools provide probability estimates, not diagnoses—they cannot definitively rule out or rule in dementia, as stated by the American Academy of Neurology 4
- When screening tests show conflicting results or borderline scores, comprehensive neuropsychological assessment is indicated, according to the Mayo Clinic Proceedings 1, 2
- An abnormal or borderline score on any screening test must be followed by: neurologic examination, multidomain mental status testing, appropriate laboratory studies, and imaging studies, as recommended by the Mayo Clinic Proceedings 1, 2
Next Steps in Diagnosis
- Obtain collateral history from family members or caregivers regarding functional decline in activities of daily living, behavioral changes, and medication adherence, as suggested by the Mayo Clinic Proceedings 3, 4
- Screen for dementia mimics: depression (using PHQ-2/PHQ-9) and delirium (using Confusion Assessment Method), according to the Mayo Clinic Proceedings 1, 2
- Consider administering a more comprehensive multidomain test such as the Montreal Cognitive Assessment (MoCA), which takes 10-15 minutes and may provide additional diagnostic clarity, as recommended by the American Academy of Neurology 2, 4
Avoiding Common Pitfalls
- Do not assume normal cognition based on a single screening test, especially when another test suggests impairment, as warned by the American Academy of Neurology 4
- Education level, language barriers, and cultural factors significantly affect test performance and must be considered in interpretation, according to the Mayo Clinic Proceedings 2, 3