The Mini-Mental State Examination (MMSE) is a clinician-administered, 30-point screen of global cognitive function. Across roughly ten short tasks it samples orientation to time and place, registration and delayed recall, attention and calculation, language (naming, repetition, comprehension, reading, writing) and visuoconstruction. It takes about 5-10 minutes and yields a single total score out of 30, with higher scores indicating better cognition.
For decades it has been the default bedside screen for dementia and delirium, and remains the metric in which much of the older dementia literature - including drug-trial eligibility bands - is expressed. It is a screening and staging aid, not a diagnostic instrument: it estimates the degree of global impairment and tracks gross change over time.
02 - Origin & purpose
Where it comes from.
The MMSE was published in 1975 by Marshal Folstein, Susan Folstein and Paul McHugh as a practical method for grading the cognitive state of patients on psychiatric wards, where full neuropsychological testing was impractical. Its brevity and face validity carried it far beyond its original setting, and it became the most widely used cognitive screen in the world.
Its later history is dominated by copyright. In 2001 the authors transferred rights to MiniMental LLC, with Psychological Assessment Resources (PAR) as exclusive publisher. Since then, reproduction of the items requires a licence and per-form fees apply, which has pushed many services towards free alternatives. This page therefore explains scoring and evidence but does not reproduce the instrument itself.
03 - Scoring & cutoffs
How scoring works.
The MMSE is scored 0-30. The traditional cutoff for significant cognitive impairment is 23/24 (scores of 23 or below screen positive). Commonly used severity bands are 24-30 no or questionable impairment, 19-23 mild, 10-18 moderate, and 9 or below severe impairment. Scores must be interpreted against education, age, language and sensory function: highly educated patients can score in the "normal" range despite early dementia, while low education, limited literacy or a different first language can depress scores in the cognitively well. Some services adjust cutoffs by education for this reason.
Score
Severity
Interpretation
0–9
Severe
Severe global cognitive impairment.
10–18
Moderate
Moderate global cognitive impairment.
19–23
Mild
Mild cognitive impairment range; 23 or below is the traditional positive screen.
24–30
No or questionable impairment
No or questionable cognitive impairment. Interpret against education and language background.
04 - Validation evidence
How well it performs.
A systematic evidence review for the US Preventive Services Task Force (Lin et al., 2013) pooled 14 studies and found good accuracy for detecting dementia at the standard cutoff. Accuracy is materially worse for mild cognitive impairment, where the MMSE lacks difficult enough items - a key reason the MoCA was developed. Reliability is adequate for a short screen, with internal consistency typically around 0.7-0.8 and strong short-interval test-retest stability in the original and subsequent studies.
88.3% / 86.2%
Screening for dementia
Sensitivity / specificity at the 23/24-25 cutoff; pooled across 14 studies - Lin et al. 2013
α ~0.76
Internal consistency
Varies with setting and population
r = 0.80-0.95
Test-retest reliability
Short intervals
5-10 min
Administration time
Clinician-administered
05 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
MMSE
~10 tasks / 30 pts
5-10 min
Licensed (PAR). Reach for it when continuity with older records or trial literature matters.
Free for clinical use but administration now requires paid certification training. Reach for it when mild cognitive impairment is the question - it is harder and more sensitive at the mild end.
SLUMS
30 pts
~7 min
Free. A close free substitute for the MMSE with better sensitivity to mild impairment in some studies.
Mini-Cog
3-word recall + clock draw
~3 min
Free for clinical use. Reach for it as an ultra-brief triage screen in primary care.
Free. Reach for it when low mood may explain poor cognitive performance in an older adult.
06 - When to use it
Right tool, wrong tool.
Reach for it when
-Services whose records, protocols or commissioning metrics are already anchored to MMSE scores.
-Staging established dementia and tracking gross decline.
-Interpreting older literature and drug-trial criteria expressed in MMSE bands.
-Settings that hold a valid PAR licence.
Reach for something else when
-Detecting mild cognitive impairment or early executive dysfunction - use the MoCA.
-Patients with low education, limited literacy or limited command of the testing language, without adjustment.
-Repeated testing at short intervals - practice effects inflate scores.
-Any service without a licence - use the SLUMS or the Mini-Cog instead.
-Diagnosis of dementia subtype, which needs full assessment.
07 - Confidence & precision
Reading the score with care.
Short-interval test-retest correlations of 0.80-0.95 imply that small score changes are unreliable: differences of 1-2 points are within measurement noise, and no consensus minimal clinically important difference exists. In practice, changes of 3 points or more, sustained across occasions, are more likely to reflect real change than measurement error - untreated Alzheimer's disease typically declines around 3 points per year on the MMSE. Practice effects inflate scores on repeat testing at short intervals; ceiling effects blunt precision in mild impairment, and floor effects in severe dementia.
08 - Limitations
What it cannot tell you.
Insensitive to mild cognitive impairment and to executive dysfunction, which it barely samples. Strong education, age, culture and language biases. Ceiling effects in intelligent, educated patients; practice effects with repetition. Scores cannot distinguish dementia from delirium or depression ("pseudodementia"). And the licensing itself: reproduction without a PAR licence is a copyright infringement, which constrains its use in digital tools and means this page cannot host the instrument.
09 - Licensing, explained
How licensing works.
The MMSE is copyrighted by MiniMental LLC and published exclusively by Psychological Assessment Resources (PAR), which has enforced the copyright since 2001. Reproduction of the items requires a licence, and per-form fees apply.
Because of this, Aisel cannot host the instrument or offer a scorable version. This page is an information resource: it explains scoring, cutoffs and evidence, but contains no MMSE items or task wording. Clinicians who wish to administer it should purchase official forms from parinc.com.
Free alternatives with comparable coverage include the SLUMS and the Mini-Cog.
See how Aisel removes friction where it costs most. A 20-minute walkthrough tailored to your clinic.
We value your privacy
We use cookies to analyse site usage and improve your experience. Analytics and embedded media (e.g. YouTube) only load if you accept. Read our cookie policy.