The Mini-Cog is a brief cognitive screen for older adults combining two tasks: registration and delayed recall of three words, and the clock drawing test. It takes about three minutes and was designed as a "cognitive vital sign" usable in primary care and multilingual populations, with minimal education and language bias compared with longer instruments (Borson et al., 2000, Int J Geriatr Psychiatry).
It is administered by a clinician rather than self-completed, so no interactive version is offered here. Developed by Soo Borson and colleagues (2000).
02 - Origin & purpose
Where it comes from.
Soo Borson and colleagues at the University of Washington published the Mini-Cog in the International Journal of Geriatric Psychiatry in 2000. The problem they set out to solve was practical: the Mini-Mental State Examination took too long for routine primary care and was biased by education and language, which mattered in the ethnically diverse, often non-English-speaking older population they served. They combined the two elements of a longer battery that carried most of the discriminating power - three-word recall and clock drawing - into a test that takes about three minutes. In the original community sample of 249 older adults, half of them non-English speakers, the Mini-Cog classified 96% correctly and, unlike the MMSE, was not influenced by education or language.
The original scoring used a decision algorithm combining recall and the clock; Scanlan and Borson described its receiver operating characteristics in 2001, and the algorithm was later replaced by an equivalent 0-5 numerical score. Borson's group validated the test in a population-based sample of 1,119 participants in 2003 and compared it with the MMSE in a multiethnic sample of 371 in 2005, where it matched the MMSE's accuracy without the MMSE's education effect. The standardised form now in use (version 01.19.16) is distributed through mini-cog.com. The instrument remains the copyright of Dr Borson: free for clinical and educational use, with permission required for commercial, research or modified use.
03 - Scoring & cutoffs
How scoring works.
Scores range 0-5: one point per word recalled (0-3) and 2 points for a normal clock (0 or 2). A total of 0-2 is a positive screen for possible dementia; 3-5 is a negative screen. When higher sensitivity is wanted, some guidance uses a cutoff below 4. A positive screen is not a diagnosis - it indicates the need for fuller cognitive assessment.
04 - Validation evidence
How well it performs.
The Mini-Cog has been evaluated in community, primary-care and hospital samples, with accuracy that varies by setting and reference standard. In the 2003 population-based validation (n=1,119) it had 76% sensitivity and 89% specificity for dementia, essentially the same as the MMSE at a cutoff of 25 (79% and 88%). A 2024 systematic review and meta-analysis of 14 studies using the standard cutoff of 2 or less found pooled sensitivity of 76% (95% CI 73-78%) and specificity of 83% (82-84%) for dementia across 4,772 participants, with an area under the curve of 0.85, but heterogeneity was high (I² above 90%).
The Cochrane review of primary-care studies (Seitz et al., 2018, updated 2021) was more cautious: across four studies of 1,517 participants sensitivity ranged from 0.76 to 1.00 and specificity from 0.27 to 0.85, and in the only study at low risk of bias sensitivity was 0.76 and specificity 0.73. The authors concluded the evidence was insufficient to recommend routine use in general practice. In the 2000 and 2005 studies the Mini-Cog's accuracy was not affected by years of education, whereas the MMSE's was. Inter-rater agreement for clock scoring between untrained and expert raters exceeded 98% for normal and clearly abnormal clocks but fell to about 60% for mildly impaired clocks (Scanlan & Borson, 2001).
76%
SENSITIVITY (≤2)
Dementia, population-based validation (2003)
89%
SPECIFICITY (≤2)
Dementia, population-based validation (2003)
76%
POOLED SENSITIVITY
14-study meta-analysis, 2024 (specificity 83%)
76% / 73%
PRIMARY CARE, LOW BIAS
Sensitivity / specificity, Cochrane 2018
05 - How it compares
How it compares to the alternatives.
Instrument
Items
Time
When to reach for it
Mini-Cog
5 pts / 2 tasks
~3 min
Clinician-administered. Free for clinical and educational use; commercial, research or modified use needs permission.
Free for clinical use. Informant-rated, so unaffected by the patient's literacy, vision or motor ability.
06 - When to use it
Right tool, wrong tool.
The Mini-Cog is a first-stage screen for dementia in older adults. Its place is the annual health check, the hospital admission, the pre-operative assessment or the psychiatric intake where a clinician wants to know within three minutes whether cognitive impairment is likely enough to warrant a fuller assessment. A positive screen (0-2) should be followed by a longer cognitive test such as the MoCA or MMSE, a functional history and medical work-up for reversible causes; it is not a diagnosis. Because the clock task is drawn rather than spoken, the test can be used with people whose first language is not the clinician's, and alternative word lists allow repeat administration.
It is the wrong tool for grading severity or tracking change: a five-point range and a pass/fail clock cannot register gradual decline. It is also not recommended as a screen for mild cognitive impairment, where sensitivity is inconsistent and the evidence base thin. In people with visual, motor or literacy limitations that prevent clock drawing, a verbal or informant-based tool such as the AD8 is more appropriate.
Reach for it when
-A three-minute first-stage dementia screen in primary care, hospital admission or psychiatric intake.
-Older adults with low literacy or whose first language differs from the clinician's - accuracy is not education-dependent.
-Deciding whether a fuller assessment (MoCA, MMSE, neuropsychology) is warranted.
-Settings where staff without cognitive-testing training must administer and score.
Reach for something else when
-Screening for mild cognitive impairment - the developers say the evidence does not support it.
-Grading severity or tracking change over time; the 0-5 range is too coarse.
-Visual, motor or literacy limitations that prevent clock drawing - use an informant tool such as the AD8.
-Adults under 65 or non-dementia cognitive complaints; validation is in older adults.
-Any diagnostic conclusion - a positive screen needs full assessment.
07 - Confidence & precision
Reading the score with care.
A Mini-Cog result is a probability statement, not a measurement of severity. At the recommended cutoff of 2 or less, roughly one in four people with dementia will screen negative and, depending on setting, one in six to one in four people without dementia will screen positive; in the best-quality primary-care study about 24% of cases were missed and 27% of non-cases flagged. Because the total has only six possible values and the clock contributes 0 or 2, one borderline judgement on the clock changes the outcome, and untrained raters disagree with experts on around 40% of mildly impaired clocks. A score of 3 to 5 lowers but does not exclude the likelihood of impairment, particularly in people with high premorbid ability, and a low score can be transient in acute illness or with sedating medication. Repeat testing should use an alternative word list and the standardised, not the graphical, form.
08 - Limitations
What it cannot tell you.
The Mini-Cog is a dementia screen only. It does not diagnose, stage or monitor cognitive disorder, and mini-cog.com states there is insufficient evidence to recommend it for mild cognitive impairment. Accuracy estimates vary widely between studies, and the Cochrane reviews of primary and secondary care found the evidence too heterogeneous and too often at risk of bias to recommend routine use.
Clock drawing requires vision, hand function, literacy and familiarity with analogue clocks; people who lack any of these can fail the task while cognitively intact. The narrow scoring range and pass/fail clock make the result sensitive to single scoring decisions and unsuitable for detecting gradual change. It has not been validated in adults under 65. Because it is copyrighted it may not be modified, embedded in commercial software or used in research without the author's permission - which is why no interactive version appears on this page.
09 - Licensing, explained
How licensing works.
The Mini-Cog is copyrighted by S. Borson. It is free for use by health professionals, hospitals, clinics and other non-commercial providers without needing permission. Commercial use, modification or use in research requires prior written permission and a licensing agreement (mini-cog.com). That is why no interactive version, word lists or stimulus materials are reproduced here. Official forms and administration instructions are available from mini-cog.com.
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.