Cognitive Assessment Tools

Cognitive assessment tools fall into four functional groups, and stating this classification first earns marks: - Delirium instruments — CAM, 3D-CAM, CAM-ICU, 4AT - Brief screening tools — Mini-Cog, AD8, AMT, GPCOG, MIS, six-item screener - Cognitive screening batteries — MMSE, MoCA, SLUMS, ACE-III, RUDAS - Detailed neuropsychological batteries and staging scales — RBANS, CVLT, Trail-Making, FAB, Boston Naming, Rey– Osterrieth; CDR and Reisberg GDS for staging

A second axis is the informant dimension — performance-based (patient does the task: MMSE, MoCA) versus informant-based (a relative reports change: AD8, IQCODE). Informant tools detect change from a personal baseline and are invaluable when education confounds performance tests.

The tool must match the question: Is this delirium? → CAM. Is there any cognitive problem? → Mini-Cog or AD8. How severe and in which domain? → MoCA, MMSE, SLUMS. Which dementia is it? → ACE-III with a full battery. How advanced is it? → CDR or Reisberg GDS.

CAM — sensitivity 94%, specificity 89%, high inter-rater reliability. Four features: acute onset and fluctuating course, inattention, disorganised thinking, altered level of consciousness (features 1 and 2 plus either 3 or 4).

Inattention testing at the bedside: months of the year backwards, days of the week backwards, digit span 3 or 4 backwards; or the examiner may simply observe inappropriate distraction or loss of the thread of conversation.

3D-CAM — three-minute structured version: sensitivity 96%, specificity 86% in patients with dementia; 93% and 96% without dementia. CAM-ICU for intubated patients uses the "SAVEAHAART" letter-A task and the RASS.

Mini-Cog — three-item recall plus clock drawing; screens memory and executive function in about three minutes and is relatively education-independent, making it the most practical screen in Indian outpatient settings.

AD8 — an eight-item informant questionnaire detecting change over time; useful when the patient's education level makes performance testing unreliable.

Abbreviated Mental Test (Hodkinson, 10 items) — the RCP/BGS-endorsed memory and cognition screen; extremely fast but crude.

Memory Impairment Screen and the five-word test assess anterograde verbal memory specifically at a first consultation.

MMSE (Folstein, 1975) — 30 points across orientation, registration, attention/calculation, recall, language and construction. Interpretation is complex because thresholds depend on education and age; it has a ceiling effect and tests executive function poorly. Severity bands: mild 20–24, moderate 10–19, severe below 9.

MoCA (Nasreddine, 2005) — 30 points; adds trail-making, cube copy, clock drawing, verbal fluency, abstraction and delayed recall, so it detects mild cognitive impairment and executive/subcortical dysfunction far better than the MMSE. Add 1 point if formal education is 12 years or less.

SLUMS (Saint Louis University Mental Status) — 30 points; free of charge, with separate cut-offs for those with and without a high-school education.

ACE-R / ACE-III — 100 points across attention/orientation, memory, verbal fluency, language and visuospatial function; ACE-R incorporates the MMSE and clock-face drawing, and the VLOM ratio helps separate Alzheimer's disease from frontotemporal dementia.

RUDAS (Rowland Universal Dementia Assessment Scale) — designed to be culturally and linguistically fair, needing no literacy; highly relevant to multilingual, variably literate Indian populations.

Domain mapping (Table 62.4): premorbid ability — NART; screening battery — MMSE, MoCA, RBANS, FAB; attention — digit span forward and backward; language — Boston Naming Test; memory — story recall, California Verbal Learning Test; executive — Trail-Making A and B, clock drawing; visuospatial — Rey– Osterrieth Complex Figure; mood — GDS-15.

Free and cued recall tests — the Grober and Buschke test and the CVLT — separate mild Alzheimer's disease from normal ageing with more than 90% precision, even when MMSE scores are identical, because they demonstrate a hippocampal amnestic pattern (cueing does not improve recall).

Staging scales: Clinical Dementia Rating (CDR) and the Reisberg Global Deterioration Scale combine cognition with ADL impact. The Severe Impairment Battery is used when impairment is too severe for standard testing.

Practical conduct: obtain consent, explain the purpose, ensure adequate lighting, hearing amplification, suitable seating and no distractions, introduce questions non-threateningly, phrase feedback positively, and give the patient a chance to comment on their own performance at the end. Assess premorbid intellectual status first, and always add a mood scale.

Two cardinal caveats: - Depression and delirium must be excluded before any score is attributed to dementia — screening tools become unreliable in delirium, and depression is the commonest reversible mimic. - No cognitive screening test is appropriate for determining decision-making capacity. Screens of executive function such as the clock-drawing test or the Executive Interview correlate better with decisional capacity than the MMSE, but capacity requires its own task-specific assessment.

2024 NIA-AA Revised Criteria define Alzheimer's disease biologically through Core 1 biomarkers — plasma p-tau217, amyloid PET, or CSF p-tau181/Aβ42, t-tau/Aβ42 or Aβ42/40. Cognitive tools now stage the disease rather than diagnose it.

Plasma p-tau217 is the practice-changing test. Models combining plasma p-tau217, age and ApoE genotype achieve AUC about 0.94. In the very elderly, p-tau181, p-tau217 and GFAP discriminate probable Alzheimer's from normal cognition, while Aβ42, Aβ40 and the Aβ42/40 ratio do not — so amyloid ratios should not be relied on in the oldest old.

Digital cognitive assessment. Mobile cognitive assessment systems show diagnostic equivalence to MMSE and MoCA in Alzheimer's screening. Future work is on composite scores integrating digital tests, wearables (for example sleep fragmentation) and plasma p-tau217, with cross-ethnic validation and regulatory frameworks still pending.

Mild behavioural impairment (MBI) — late-onset persistent neuropsychiatric symptoms — is now recognised as an early dementia-risk indicator that may precede measurable cognitive decline, and is being validated against plasma p-tau217.

Mandated inclusion in assessment frameworks. German S3 CGA guideline (2024): cognition and delirium is one of six core dimensions. WHO ICOPE: cognition is one of six intrinsic-capacity domains, with Mini-Cog, MoCA, MMSE and GPCOG as named step-2 tools. Age-Friendly 4Ms: cognition sits under Mentation. ASCO 2023 Practical Geriatric Assessment includes a cognitive domain for every cancer patient aged 65 and above.

Indian tools and data. LASI wave-1 found impaired cognition in 9.3% of those aged 60 and above. Because the MMSE penalises illiteracy heavily, Indian practice uses the Hindi Mental State Examination (HMSE), RUDAS, vernacular MoCA versions, the ICMR Neuro-Cognitive Tool Box, and the AIIMS Comprehensive Neuropsychological Battery in Hindi.