Abbreviated Mental Test Score Calculator
Abbreviated Mental Test Score Calculator
Ten questions, one point each, from Hodkinson’s 1972 paper. The published cut-offs disagree by two points, which leaves a band one point wide where a patient is abnormal on one source and normal on the other.
Abbreviated Mental Test Score (AMTS)
10 items, 0 to 10Age, time, year, place, two people, date of birth and monarch all correct; the recalled address, the First World War year and counting backwards all wrong
Scoring
Published cut-offs: 6 or less, and below 8
- two cut-offs, two points apart
- the British Geriatrics Society’s reproduction gives “6 or less suggests delirium or dementia”. The validity study of 168 consecutive acute admissions found the best cut-off to be 8, with scores below 8 suggesting abnormal cognition. Published cut-offs for this instrument range from 6 to 10 across the literature. A score of 7 is normal on one and abnormal on the other
- no sensitivity and specificity printed here
- deliberately. The validity study’s abstract reports the best cut-off and a seven-item short form and gives NO numeric sensitivity or specificity; its full text is subscriber-only and was not read. A figure of 92 per cent sensitivity and 87 per cent specificity circulates for a culturally adapted version, from a single secondary summary, and this page will not print a performance figure on that basis. For a worked example of what sensitivity and specificity do to predictive value, see the Short Blessed Test page
- the items are not fixed
- the First World War item and the monarch item are general-knowledge questions tied to a time and a country, and reproductions substitute the Second World War, a president or a prime minister. A substitution makes a different instrument, and no published cut-off was derived on it. Shortened forms exist too — AMT4, AMT5 and AMT7 — with their own cut-offs of 3 or 4, 4, and 5 respectively
- what moves the score other than cognition
- education, first language, deafness, visual impairment, aphasia, pain, opioids and the physical environment. Two of the ten items are general knowledge and one depends on there being a clock-free room and two identifiable people present. None of this is adjustable: the instrument has no norms by age or education and this page offers none
- derived from
- the 26-item Blessed Dementia Scale, by removing 16 items. The AMT4 subset — age, date of birth, place and current year — is in turn the second item of the 4AT delirium assessment
- licensing
- ten short factual questions published in a journal in 1972 and reproduced in British geriatric practice ever since. No proprietor, no licence, no registration and no training requirement, and no permission notice on either independent reproduction read for this page, one of them the British Geriatrics Society’s own
Worked example
Age, time, year, place, two people, date of birth and monarch all correct; the recalled address, the First World War year and counting backwards all wrong
Seven items correct = 7 points
7 is the one-point band where the two published cut-offs disagree: normal at “6 or less is abnormal”, and abnormal at “below 8 is abnormal”
Look at which three items failed rather than at the total. Delayed recall of the address and counting backwards from 20 are the memory and attention items; the First World War year is general knowledge. A patient who fails the two cognitive items and one knowledge item is a different patient from one who fails three knowledge items for the same total
Score the First World War item correct instead and the total is 8, normal on both cut-offs, from a general-knowledge question about 1914
Score the address recall correct as well and the total is 9
A normal total does not exclude delirium. The AMTS has no item for acute change and no item for fluctuation, which is what the 4AT adds
The ten items, and what each one actually tests
| Item | Domain | Caution |
|---|---|---|
| Own age | Orientation to self | Cross-check against the date of birth item |
| Time to the nearest hour | Orientation to time | Remove the clock first; fails in a windowless bay |
| Recall of a five-element address | Delayed recall | The most informative item, and the one skipped when the address is not given at the start |
| The year | Orientation to time | Preserved late; less sensitive than it looks |
| Name of this place | Orientation to place | Fails after repeated transfers |
| Recognises two people | Recognition | Unscoreable if nobody identifiable is present |
| Own date of birth | Remote memory | The most robustly preserved item |
| Year the First World War began | General knowledge | Culturally and generationally specific; often substituted |
| Name the present monarch | General knowledge | Country-specific; often substituted |
| Counts backwards from 20 to 1 | Attention | The item delirium disturbs first |
Published cut-offs and where they came from
| Cut-off | Source | Population |
|---|---|---|
| 6 or less suggests delirium or dementia | British Geriatrics Society reproduction, citing Hodkinson 1972 | British geriatric practice; no cohort stated |
| Below 8 suggests abnormal cognitive function | Jitapunkul, Pillay and Ebrahim 1991 | 168 consecutive acute admissions to a department of health care of the elderly, 58 (34%) with abnormal cognition against the final clinical diagnosis |
| Reported cut-offs range from 6 to 10 | Secondary summary of the wider literature | Various, including culturally adapted versions |
Ten questions, two cut-offs, and a band one point wide
Hodkinson built the Abbreviated Mental Test in 1972 by cutting sixteen items out of the twenty-six-item Blessed Dementia Scale, leaving ten questions that could be asked at a bedside in two minutes. Each correct answer scores a point. It is still the default cognitive screen in British geriatric and orthopaedic practice, it is recorded in national audits, and it has the virtues of being short, free of any proprietor and memorable enough to be used.
What it does not have is a single cut-off. The British Geriatrics Society’s own reproduction says that six or less suggests delirium or dementia. The 1991 validity study of 168 consecutive acute admissions, in which thirty-four per cent had abnormal cognition against the final clinical diagnosis, found the best cut-off to be eight, with scores below eight abnormal. Cut-offs from six to ten appear in the wider literature. A score of seven is therefore normal on one published reading and abnormal on another, and no amount of arithmetic settles it — which is why this page has a band one point wide and labels it that way.
Nor does it have a sensitivity and a specificity this page is willing to print. The validity study’s abstract reports the cut-off and a seven-item short form and gives no numeric accuracy figures; its full text is behind a subscription. A pair of figures for a culturally adapted version circulates in secondary sources and is not reproduced here. The reason that matters is arithmetic: the useful question about a screening test is not its sensitivity but the probability that a positive result is right, and that depends on how common the condition is in the population screened. Worked through on an instrument whose figures ARE published, on the Short Blessed Test page, a test with 88 per cent sensitivity and 94 per cent specificity gives a positive predictive value of 86 per cent where three in ten are affected and 44 per cent where one in twenty are. The same test, the same thresholds, and half the positives wrong.
Two practical cautions. The items are not fixed: the First World War question and the monarch question are general knowledge tied to a country and a generation, and reproductions substitute the Second World War, a president or a prime minister — which makes a different instrument with no validated threshold. And the total hides the pattern. Failing delayed recall and counting backwards is not the same as failing two general-knowledge items, and the AMTS has no item at all for acute change or fluctuation, which is what a delirium instrument is for.
A screening score is not a diagnosis: a published sensitivity is a property of the instrument in the population it was validated in, not a statement about this patient. This page reports published figures and recommends no action. Every weight, cut-off and outcome figure here comes from a named derivation cohort, and cohorts differ in case mix, era, coding and outcome definition; where your own institution’s protocol or analysis plan differs, it takes precedence.
Frequently asked questions
What is an abnormal Abbreviated Mental Test Score?
It depends which source you follow, and they differ by two points. The British Geriatrics Society’s reproduction gives six or less as suggesting delirium or dementia; the 1991 validity study in 168 acute admissions found the best cut-off to be eight, with below eight abnormal. A score of seven is normal on the first and abnormal on the second.
Can the AMTS diagnose dementia?
No. A screening score is not a diagnosis: a published sensitivity is a property of the instrument in the population it was validated in, not a statement about this patient. It is insensitive to mild cognitive impairment, it cannot distinguish dementia from delirium — it has no item for acute change or fluctuation — and two of its ten items are general knowledge rather than cognition.
Can I substitute the First World War or monarch questions?
People do, and it makes a different instrument. Both are general-knowledge items tied to a country and a generation, and reproductions outside Britain substitute the Second World War or a head of state. The published cut-offs were derived on the original ten items, so a locally adapted version has no validated threshold. If you substitute, record that you did.
What is the AMT4, and how does it relate to this?
A four-item subset — age, date of birth, place and current year — with a cut-off of three or four. It is the second item of the 4AT delirium assessment. AMT5 and AMT7 short forms also exist, with cut-offs of four and five; the 1991 study developed the AMT7 and found it matched the full test on validity and internal consistency.
Is the Abbreviated Mental Test Score copyrighted?
No proprietor, licence, registration or training requirement could be found. It was published in Age and Ageing in 1972 and is reproduced without a permission notice in British geriatric medicine documents, including the British Geriatrics Society’s own. That is unusual among cognitive instruments: the Mini-Mental State Examination is sold under licence and the Montreal Cognitive Assessment requires registration and training, and neither is reproduced anywhere on this site.
Related calculators
References
- Hodkinson HM. Evaluation of a mental test score for assessment of mental impairment in the elderly. Age Ageing. 1972;1(4):233–8.
- British Geriatrics Society. Abbreviated mental test score (bgs.org.uk, attachment of 5 July 2018), citing Hodkinson 1972 (accessed 9 October 2026).
- Jitapunkul S, Pillay I, Ebrahim S. The abbreviated mental test: its use and validity. Age Ageing. 1991;20(5):332–6.
- Wikipedia. Abbreviated Mental Test Score. en.wikipedia.org (accessed 9 October 2026).
Not medical advice. For healthcare professionals and education. Reference intervals vary by laboratory and assay — always use your own laboratory's. Never base a dose or a treatment decision on this page alone. Full disclaimer at calcengines.com/disclaimer/