MMSE Scores Explained: Ranges, Cutoffs, and What They Mean

Why an MMSE Score Is a Starting Point, Not a Verdict

The MMSE was introduced as a practical, brief way for clinicians to grade a person's cognitive state at the bedside or in the office (Folstein et al., 1975). Because it produces one tidy total out of 30, it can feel like a grade. In practice, that total is designed to flag whether thinking and memory deserve a closer look — not to describe everything about how your brain works.

That is why the same score can mean different things for different people. A score near the cutoff for someone tested after a poor night's sleep, in a second language, or with fewer years of formal schooling does not carry the same weight as the identical score for someone tested under ideal conditions. For the bigger picture on interpreting any result, start with our guide to understanding your cognitive test results.

Key Facts at a Glance

• The MMSE is scored on a 0–30 scale, with higher totals reflecting stronger performance across the sampled skills.
• The conventional cutoff is 24; scores below it suggest more evaluation may help, not that a condition is present (Cochrane, 2016).
• At that cutoff, pooled community accuracy was 85% sensitivity and 90% specificity — good, but not perfect.
• Scores vary with age and education: population medians range from 29 down to 22 depending on schooling (Crum et al., JAMA, 1993).
• Roughly 1.5 to 1.7 points is the anchor-based threshold for a clinically important within-person change (Borland et al., Neurology, 2022).
• The MMSE is a screening tool, not a diagnosis, and should not be used in isolation to confirm or rule out dementia.

How Is the MMSE Scored?

The exam takes roughly 5 to 10 minutes and adds up points across several short sections that sample different areas of thinking — for example, orientation to time and place, attention, short-term recall, and language. Each correctly completed task contributes points toward the maximum of 30. A trained administrator follows a standard procedure so the result is comparable across settings, and scoring the completed form takes about another five minutes (PAR, Inc.).

Three features of the scoring matter when you read a result. First, the total is a broad snapshot: it samples several skills rather than measuring any one of them precisely. Second, the sections are not weighted equally by difficulty, so two people can arrive at the same total having lost points in completely different areas — which is why clinicians look at where points were lost, not just the total. Third, lower totals indicate increasing difficulty on those tasks, but they do not point to a specific cause on their own (Cochrane, 2016). For a plain-language overview of how the exam itself is given, see the MMSE test.

MMSE Score Ranges and What They Suggest

In general terms:

• 24–30 falls within the conventional normal range.
• Below 24 suggests that further conversation and evaluation could be worthwhile.

Clinicians sometimes group lower totals into broad bands they describe as mild, moderate, or more significant difficulty. Those labels are guides for clinical judgment, not fixed diagnostic lines, and different settings use slightly different thresholds — worth remembering if two sources give different boundaries for the same score.

What the ranges genuinely tell you is direction and degree, not cause. The comprehensive review of the MMSE found high sensitivity for moderate-to-severe cognitive impairment and noticeably lower sensitivity for mild impairment, and concluded the exam fulfills its goal of quantifying severity and documenting change over time — while stating it should not, by itself, be used as a diagnostic tool (Tombaugh & McIntyre, JAGS, 1992). Cochrane lands in the same place: the MMSE contributes to a diagnosis of dementia but should not be used in isolation to confirm or exclude disease (Creavin et al., 2016). For how ranges are read across different tools, see our explainer on what cognitive test scores mean.

What Does an MMSE Score of 22 Out of 30 Mean?

Two points below the conventional cutoff, a 22 is the kind of result that prompts a clinician to keep looking rather than to conclude anything.

In a population study of 18,056 adults, the median MMSE score was 22 for people with zero to four years of schooling, 26 for those with five to eight years, and 29 for those with nine or more (Crum et al., JAMA, 1993). For someone with limited formal education, a 22 may be close to typical for their reference group. For a university-educated 60-year-old, the same 22 is further from expectation and would usually earn a fuller workup.

Everyday factors also move the number: acute illness, pain, poor sleep, medication side effects, uncorrected hearing or vision problems, and taking the exam in a second language. A single total should never be read on its own. If you have a low result in hand, we walk through what happens if you fail a cognitive test and how that conversation usually goes.

Why the Cutoff Isn't a Pass-or-Fail Line

The familiar "24" is a convention. The Cochrane review is explicit that while the cut point defining normal cognitive function is usually set at 24, it could theoretically fall anywhere from 1 to 30, and other cut points have been proposed (Creavin et al., 2016).

The review also shows what moving the line costs and buys. Pooling 15 community studies at a cut point of 24 gave a sensitivity of 0.85 and a specificity of 0.90. Raising the cut point to 25 across 10 studies raised sensitivity to 0.87 but dropped specificity to 0.82 — catching slightly more cases at the price of more false alarms. In seven studies that adjusted estimates for education, sensitivity rose to 0.97 while specificity fell to 0.70.

That trade-off is the argument against treating 24 as a verdict: every threshold sends some people for unnecessary follow-up and misses some who need it, and the right balance depends on the person in front of the clinician. If you are comparing tests, our look at how the MoCA and MMSE compare explains why different screens use different scales and cutoffs.

How Do Age and Education Change What a Score Means?

Substantially — enough that ignoring them is the most common way an MMSE result gets misread.

The same population study found an inverse relationship between MMSE scores and age, with medians falling from 29 among adults aged 18 to 24 to 25 among those 80 and older, alongside the education gradient described above (Crum et al., JAMA, 1993). The authors published percentile distributions specific to age and education precisely so clinicians could compare a patient against a relevant reference group rather than a single fixed number — and concluded that MMSE scores should be used to identify current cognitive difficulties, not to make formal diagnoses.

The comprehensive review of the MMSE reached a compatible conclusion, reporting that scores were affected by age, education, and cultural background, though not by gender, and that the exam is highly verbal — which matters for anyone tested outside their first language (Tombaugh & McIntyre, 1992).

Two related guides go deeper on this: how age-adjusted cognitive scores work and how cognitive test percentiles are read, which explains why a percentile often communicates more than a raw total.

How Much of a Change Between Two MMSE Scores Is Meaningful?

This is the question that matters most once you have two results, and two commonly quoted numbers answer different parts of it.

Using change on the Clinical Dementia Rating scale as an anchor, a study in the Swedish BioFINDER cohort estimated the minimal clinically important difference for the MMSE at about 1.5 points in cognitively unimpaired adults and 1.7 points in people with mild cognitive impairment (Borland et al., Neurology, 2022). Those are anchor-based thresholds for clinical importance in the cohorts studied — not a measurement of how much a score drifts on its own. Separately, in the population-based Mayo Clinic Study of Aging, the observed mean annualized MMSE change around the onset of mild cognitive impairment was about −1.0 point per year (95% CI −1.12 to −0.91) (Aakre et al., 2025) — a real rate of decline, which is a reason to take a small drop seriously rather than to wave it away.

So a difference of a point or two is neither proof of trouble nor something to dismiss. What it means depends on the interval between the two tests, the conditions on each day, who administered and scored the exam, the measurement error of the instrument, and the rest of the clinical picture. Judging it properly takes a statistical criterion rather than mental subtraction: the publisher of the second edition supplies reliable change scores at three levels of statistical significance for exactly that comparison, along with equivalent alternate forms, labeled Blue and Red, so the same person can be retested with reduced practice effects — because seeing the same material again can lift a score for reasons unrelated to brain health (PAR, Inc.). Bring both scores and both dates to your clinician and ask what the difference means in your case.

For intervals and what to do with a moving number, see how often you should retest cognitive function and what happens if results change over time.

What Does an MMSE Score Say About Severity?

Because the MMSE was designed partly to quantify severity, researchers have worked out how its totals line up with clinician-rated staging. A 2025 study developed and externally validated a linear mapping between the MMSE and the Clinical Dementia Rating Sum of Boxes, using 26,729 paired assessments from the US National Alzheimer's Coordinating Centers database; the fitted relationship was roughly a 0.68-point change in CDR-SB for each MMSE point (Brogaard et al., 2025).

So MMSE totals do carry real information about degree, which is why the scale is used to describe severity in research. But the same study found calibration was less accurate at the severe end, and its authors frame the mapping as a way to contextualize outcomes across studies — not as a tool for staging an individual.

Why a Normal MMSE Score Doesn't Always Settle the Question

A score of 24 or above is reassuring, but it is not proof that nothing is changing. The MMSE is least sensitive exactly where early concerns live: the comprehensive review reported high sensitivity for moderate-to-severe impairment and lower sensitivity for mild degrees of impairment, noting that some language items were relatively easy and lacked utility for identifying mild deficits (Tombaugh & McIntyre, 1992).

The evidence on prediction agrees. A Cochrane review of 11 studies following 1,569 people with mild cognitive impairment found that baseline MMSE scores predicted conversion to dementia with sensitivities ranging from 23% to 76% and specificities from 40% to 94%, and concluded there was no evidence supporting a substantial role for the MMSE as a stand-alone, single-administration test for identifying who will progress (Arevalo-Rodriguez et al., Cochrane, 2021).

The practical takeaway is not to distrust a good score — it is to keep describing what you actually notice. If symptoms are present despite a normal total, that combination is worth raising directly, and our guide to what mild cognitive impairment is covers what early change can look like.

What a Lower-Than-Expected Score Means

A score below the normal range does not, by itself, diagnose anything. It tells a clinician that a fuller look could help. Many contributors to a lower score are treatable — including thyroid problems, vitamin deficiencies such as low B12, sleep disorders, mood conditions like depression, and infections such as a urinary tract infection (MedlinePlus, National Library of Medicine).

No single test settles the question either. As the Alzheimer's Association notes, there is no one test that determines whether someone is living with Alzheimer's; clinicians combine cognitive assessments with medical history, physical and neurological exams, and sometimes imaging or lab work, and dementia-like symptoms can come from depression, untreated sleep apnea, delirium, medication side effects, thyroid problems, vitamin deficiencies, or excessive alcohol use (Alzheimer's Association). An MMSE score is one input among several.

What Happens Next

The National Institute on Aging is direct about the sequence: positive results from a brief assessment warrant further evaluation, and combining neuropsychological evaluation with self-reports and reports from someone in frequent contact with the person — a spouse or other care partner — is the best way to assess cognitive impairment more fully (National Institute on Aging). A primary care provider may run that evaluation or refer to a geriatrician, neurologist, geriatric psychiatrist, or neuropsychologist.

A negative assessment is useful too: it eases a specific worry and leaves you with a baseline for comparison later. Either way, three questions are worth asking before you leave: what the recorded score was, what would count as a meaningful change if it is repeated, and what the next step is meant to rule in or out. Writing the number and the date down makes the next comparison far easier — our guide to sharing cognitive test results with your doctor covers what to bring to that conversation.

A screening result starts a conversation rather than ending one.

Taking the Next Step

For a related look at how a single number becomes meaningful, read our guide on what a good MoCA score is.

To complement a clinician-administered screen with a structured check-in you can do from home and share with your doctor, learn how Orena's clinically validated at-home test works.