MMSE: Mini-Mental State Examination

What this assessment is

The Mini-Mental State Examination, or MMSE, is a brief clinician-administered cognitive screening examination.

The original form uses 11 task categories and produces a total from 0 to 30. Higher scores indicate better performance.

It samples orientation, registration, attention, recall, language, and visual construction.

The MMSE can provide a common snapshot of cognitive performance. It cannot diagnose dementia or identify the cause of a low score.

Use it carefully.

This page covers the original MMSE described by Folstein and colleagues. MMSE-2 is a later, distinct version.

Version matters.

Who it was designed for

The original MMSE publication introduced the examination in 1975 for grading cognitive state in clinical practice.

It is now used with adults in hospitals, clinics, research, and long-term care. Its accuracy depends on the group and purpose.

Age, education, literacy, first language, culture, hearing, vision, and movement can affect the result. A single threshold does not fit every person.

The official original MMSE record describes adult use and the standard 11 categories. It also distinguishes the original examination from MMSE-2.

Use the exact version and language named in the scoring materials. Do not label a local adaptation as the original MMSE.

How it is administered

A trained examiner gives the MMSE face to face. It usually takes about 5 to 10 minutes, although some settings allow longer.

The examiner asks spoken questions and gives tasks involving memory, attention, language, reading, writing, and drawing.

Standard administration matters. Extra prompts, repeated instructions, changed wording, or substituted tasks can alter the score.

Before testing, note factors that may reduce performance:

• hearing or vision difficulty
• limited proficiency in the test language
• speech, reading, writing, or movement limitations
• pain, fatigue, anxiety, or acute illness
• delirium, sleep loss, or medicine effects

Conditions matter.

Use the complete form and its official criteria. No general rule supports creating a partial MMSE total from missing tasks.

Do not guess.

What it measures

The original MMSE samples 11 categories:

• orientation to time
• orientation to place
• registration
• attention and calculation
• recall
• naming
• repetition
• comprehension
• reading
• writing
• drawing

These categories provide a brief performance sample. They do not cover executive function, social cognition, complex judgment, or everyday functioning in depth.

The ACE-III gives a broader 100-point profile across five cognitive domains. MoCA 8.1 and SLUMS also use 30-point totals but include different tasks and interpretation rules.

Do not compare their totals as if they measured the same thing.

How scoring works

The examiner awards points according to the standard criteria. Points across the 11 categories add to a total from 0 to 30.

Higher totals indicate better performance on the sampled tasks. Lower totals indicate more errors or incomplete responses.

The categories do not contribute equally. Orientation carries more points than some language tasks.

Do not divide the score by 30 and describe it as a percentage of cognitive ability. The MMSE is not a percentage scale.

Do not estimate missing responses, prorate the total, or replace a task that could not be completed unless the exact scoring manual provides a rule for that situation.

Record the version, language, date, testing conditions, and any barrier with the result.

How to interpret an MMSE score

A score of 24 or below is often cited as a possible impairment threshold. It is not a universal rule.

Context changes its meaning.

Cutoffs have varied across studies and services. Suitable interpretation depends on age, education, language, setting, disease prevalence, and whether the purpose is screening or follow-up.

A Cochrane review found substantial variation in how the MMSE was used to detect dementia. A threshold that is useful in one population can create too many false positive or false negative results in another.

Interpret the total with the pattern of responses and the person's previous level of function.

Look beyond the total.

A lower result may support further assessment. A higher result does not rule out early change, executive difficulty, or a condition that the MMSE samples poorly.

Patterns matter.

For repeat testing, use the same version and method. Practice, acute health changes, fatigue, and a different examiner can affect the score.

What the score cannot tell you

The MMSE cannot diagnose dementia, delirium, mild cognitive impairment, depression, or another condition by itself.

It cannot explain why performance was low. Hearing loss, language mismatch, limited schooling, pain, medicine effects, neurological illness, and acute medical problems can all contribute.

The score cannot establish decision-making capacity. It also cannot show whether someone can safely manage medication, money, transport, cooking, or personal care.

A high total does not prove that cognition is unchanged. A low total does not prove a progressive disease.

Do not make a diagnosis or major care decision from the number alone.

Use history, informant evidence, daily function, examination, appropriate laboratory testing, imaging when indicated, and follow-up.

Evidence and limitations

The MMSE is short, familiar, and widely studied. Those features support consistent communication across services.

Its brevity also creates gaps. The examination gives limited coverage to executive function and can show ceiling effects in people with subtle impairment or higher education.

Education and language can shift performance. Sensory and motor barriers can lower the total even when the underlying cognitive skill is stronger.

Repeated exposure can produce practice effects. Version changes can also make longitudinal comparison misleading.

Keep the score attached to the exact form, language, setting, and clinical question.

The MMSE is most useful as one structured observation within a wider assessment. It is not a diagnosis in 30 points.