Mini-Cog Scoring · 3-Minute Dementia Screen (0-5) and Cut Points

The Mini-Cog is a three-minute cognitive screen made of two tasks: remembering three unrelated words and drawing a clock. It is scored 0 to 5 and was developed by Soo Borson and colleagues at the University of Washington as a cognitive "vital signs" measure for older adults ^{[1] [3]}.

What does the Mini-Cog measure?

The Mini-Cog screens for dementia using two tasks ^[1]:

• Three-word recall tests short-term memory.
• Clock drawing tests planning, visuospatial skill and executive function. The clock is set to ten past eleven; the Mini-Cog website notes that the nearness of the 10 and the 11 can unmask subtler executive difficulties ^[5].

It was developed in a culturally, linguistically and educationally diverse community sample of older adults, and needs little language interpretation ^[1]. In US nursing homes, the Minimum Data Set uses a different short screen, the BIMS. For a fuller clock drawing score, see the clock drawing test; for longer cognitive screens, see the MoCA, MMSE or SLUMS.

Terms of use

The standardized form carries the notice "Mini-Cog © S. Borson. All rights reserved. Reprinted with permission of the author solely for clinical and educational purposes. May not be modified or used for commercial, marketing, or research purposes without permission of the author ([email protected])." ^[4]. The Mini-Cog website adds that it "should always be made available free of charge for clinical and educational use, and there is usually no charge for other uses", and that the copyright is kept to make sure the test is used according to instructions ^[5]. This page does not reproduce the word lists; use the form from mini-cog.com.

How do you give the Mini-Cog?

• Who gives it. Any trained member of the care team, including lay health care workers, medical assistants, nurses, social workers and physicians ^[5]. The standardized form needs brief training; a graphical pass-or-fail version is designed to be given without it ^[4].
• Time. About three minutes ^[1].
• Step 1, registration. Say three unrelated words from one of the published word lists and ask the person to repeat them, with up to three tries. If they cannot, move on to the clock ^[4].
• Step 2, clock drawing. On a preprinted circle, ask the person to put in all the numbers, then set the hands to ten past eleven. Repeat the instructions as needed; this is not a memory test. Move on if the clock is not finished within three minutes ^[4].
• Step 3, recall. Ask for the three words, even if the person could not repeat them in step 1 ^[4].
• Repeat testing. Use an alternative word list for repeated administrations, and record which list was used ^{[4] [5]}.

How do you score and interpret the Mini-Cog?

A normal clock has all numbers 1 to 12, each once, in the correct order and approximate position, with hands pointing to the 11 and the 2. Hand length is not scored ^{[3] [4]}. The clock has no score of 1 on purpose: it is a simple pass or fail that people without experience in cognitive assessment can apply ^[3].

The author's scoring guidance reads the total this way ^{[3] [4]}:

The form notes that a cut point below 3 has been validated for dementia screening, but many people with clinically meaningful impairment score higher; when greater sensitivity is wanted, a cut point below 4 is recommended ^[4]. A person with a normal score who reports concern about memory or thinking should always be followed up ^[3].

Psychometric properties

• Original study. In 249 older adults (129 with probable dementia, 120 without cognitive decline; 124 were non-English speakers), the Mini-Cog had the highest sensitivity (99%) and correctly classified the most people (96%), compared with the MMSE and the CASI. Its accuracy was not affected by education or language, and it took 3 minutes against 7 for the MMSE ^[1].
• Population sample. Applied after the fact to 1,119 older adults in a Pennsylvania population study, the Mini-Cog had 76% sensitivity and 89% specificity for dementia, similar to the MMSE at a cut point of 25 (79% and 88%) and to a full neuropsychological battery (75% and 90%) ^[2].
• Primary care. A 2021 Cochrane review found four studies (1,517 participants) with sensitivity from 0.76 to 1.00 and specificity from 0.27 to 0.85. The one low-risk-of-bias study reported 0.76 and 0.73. The reviewers judged the evidence insufficient to recommend it as a dementia screen in primary care ^[6].
• Community settings. A 2021 Cochrane review of three studies (1,620 participants) found sensitivities of 0.99, 0.76 and 0.99 and specificities of 0.93, 0.89 and 0.83, but too few studies to recommend for or against it ^[7].

Limitations

• Screening, not diagnosis. The Mini-Cog does not diagnose dementia or show its cause; impairment found on it can be temporary, from a medical illness or a medication ^{[3] [5]}.
• Misses milder impairment. It can be normal in people with very mild cognitive impairment or high prior ability ^[5], and many people with clinically meaningful impairment score above the validated cut point ^[4].
• Mixed evidence. Accuracy varied widely across studies, and Cochrane reviewers found most studies at high risk of bias ^{[6] [7]}.
• Narrow coverage. Two tasks cannot profile cognition. A MoCA, an informant report such as the IQCODE, or staging with FAST or the Global Deterioration Scale answer different questions.
• Other causes to check. The Mini-Cog website advises reviewing medical history and medications for contributing factors after a low score ^[5]; a depression screen such as the GDS is a separate question in older adults.