Modified Mini-Mental State Test (3MS or MMMS)

Abstract

The Modified Mini-Mental State Test (commonly abbreviated as 3MS or MMMS) is an extended, clinician-administered neurocognitive screening instrument designed by Evelyn Lee Teng and Helena Chui in 1987. Developed as a psychometric refinement of Folstein, Folstein, and McHugh’s original 1975 Mini-Mental State Examination (MMSE), the 3MS addresses well-documented psychometric limitations of the earlier tool, including severe ceiling and floor effects, restricted score variance (0–30 range), absence of executive and abstract reasoning metrics, and lack of graded scoring for partial accuracy. The 3MS broadens the assessment spectrum across 15 distinct operational components covering registration, mental reversal, immediate cued/uncued recall, temporal orientation, spatial orientation, confrontational naming, category fluency (four-legged animals), abstract similarities, sentence and phrase repetition, reading comprehension, spontaneous writing, visuospatial construction (intersecting pentagons), three-stage motor praxis, and delayed cued/uncued recall. The test produces a continuous composite score ranging from 0 to 100 points. Extensive validation across geriatric populations, psychiatric cohorts, and broad epidemiological initiatives—most notably the Canadian Study of Health and Aging (CSHA)—has demonstrated high internal consistency (Cronbach’s alpha between .87 and .91), exemplary test-retest reliability ($r = .91$ to $.93$), and inter-rater reliability exceeding $.95$. Receiver operating characteristic (ROC) curves systematically indicate superior diagnostic sensitivity (.88–.94) and specificity (.85–.90) for detecting mild cognitive impairment (MCI) and early-stage Alzheimer’s disease relative to the standard MMSE, particularly when utilizing demographically adjusted cutoff scores accounting for age and educational attainment.

Keywords

Modified Mini-Mental State Test, 3MS, Evelyn Lee Teng, Cognitive Screening, Dementia Assessment, Mild Cognitive Impairment, Psychometrics, Neuropsychological Assessment, Mental Reversal, Category Fluency

Authors

The 3MS was conceptualized, operationalized, and clinically validated by:

• Evelyn Lee Teng, Ph.D.: Professor Emerita of Research Neurology and Psychiatry at the Keck School of Medicine of the University of Southern California (USC), Los Angeles, California, United States. Dr. Teng was a distinguished psychometrician and cognitive neuropsychologist whose methodological work revolutionized neurocognitive screening in diverse clinical, cross-cultural, and epidemiological settings.
• Helena Chui, M.D.: Chair and Professor of Neurology, Raymond and Ruth Stansbury Foundation Chair in Alzheimer’s Disease at the Keck School of Medicine of the University of Southern California, Los Angeles, California, United States. Dr. Chui is an internationally recognized neurologist specializing in the clinical phenotypes, vascular contributions, and neuropathological substrates of cognitive decline.

Subsequent psychometric adaptations, normative stratifications, and longitudinal population models have been led by key investigators including Ian McDowell (University of Ottawa), Gina Bravo and Réjean Hébert (Université de Sherbrooke), and JoAnn T. Tschanz and Kathleen A. Welsh-Bohmer (The Cache County Study, Duke University).

Purpose

The fundamental clinical and scientific purpose of the Modified Mini-Mental State Test (3MS) is to provide an expanded, psychometrically rigorous, and standardized bedside or clinic-based evaluation of global cognitive functioning. While the classical 30-point Mini-Mental State Examination developed by Folstein et al. (1975) became the undisputed global standard for rapid bedside cognitive screening in the late twentieth century, clinical neurologists, geriatricians, and neuropsychologists rapidly identified structural vulnerabilities that limited its diagnostic utility in clinical trials and epidemiological fieldwork. Chief among these shortcomings were profound ceiling effects in highly educated older adults and individuals presenting with mild cognitive impairment (MCI), severe floor effects in advanced neurodegenerative disorders, an unduly compressed metric variance (0–30), and an absolute reliance on dichotomous (0 or 1) scoring paradigms that failed to capture partial competence, spontaneous compensatory strategies, or qualitative neurobehavioral errors.

To rectify these psychometric deficits, Teng and Chui (1987) expanded the scale along several deliberate axes:

• Expanding the Measurement Range: By transitioning from a 30-point to a 100-point metric, the 3MS dramatically broadens score variance, enabling the detection of subtle micro-longitudinal changes in cognitive trajectories and therapeutic responses in clinical pharmacotherapy trials.
• Graded and Nuanced Scoring: Unlike the binary pass/fail framework of the MMSE, the 3MS introduces partial credit weighting across complex temporal calculations (e.g., year accuracy scored from 0 to 8 points based on proximity errors), spatial orientation, phrase repetition, reading execution, and visuospatial construction.
• Incorporating Executive and Frontal-Subcortical Domains: The addition of verbal category fluency (generating animal names within 30 seconds) and semantic abstraction (verbal similarities) provides crucial sampling of frontal-striatal executive control, semantic network integrity, and working memory—cognitive domains that are systematically under-represented in the MMSE.
• Differential Memory Retrieval Probing: The 3MS introduces a graded retrieval paradigm for verbal episodic memory. For both immediate recall (Question 4) and delayed recall (Question 15), the test differentiates between spontaneous free retrieval (3 points per word), category-cued retrieval (2 points per word), and multiple-choice recognition (1 point per word). This dissociation allows clinicians to distinguish between retrieval failures characteristic of subcortical-frontal vascular disease or depression versus encoding and storage consolidation failures typical of Alzheimer’s pathology.

The 3MS is extensively deployed in acute geriatric assessment, inpatient psychiatric evaluations, post-stroke neurorehabilitation monitoring, and large-scale population-based longitudinal cohort studies, such as the Canadian Study of Health and Aging (CSHA) and the Cache County Study on Memory, Health and Aging.

Psychological Construct

The 3MS measures global cognitive status, conceptualized as a multi-domain hierarchical construct wherein general cognitive competence ($g$) underpins distinct yet interdependent neuropsychological functions. The test systematically evaluates ten primary cognitive dimensions across its 15 tasks:

1. Remote Personal Semantic Memory and Orientation to Person

Assessed via Item 1 (Date and Place of Birth), this dimension examines long-term crystallized autobiographical memory. Unlike simple overlearned self-identifications, recalling the exact day, month, year, town, and province of birth requires access to retrograde semantic stores resilient to the initial stages of cortical neurodegeneration, serving as a baseline anchor for cognitive integrity.

2. Auditory Registration and Immediate Echoic Memory

Evaluated through Item 2 (Registration of Three Words: Shoes, Blue, Modesty), this construct probes sensory registration, acoustic-phonetic processing, and immediate short-term memory capacity. The selection of words intentionally integrates a concrete physical object (“Shoes”), a perceptual attribute (“Blue”), and an abstract moral construct (“Modesty”), challenging vocabulary breadths across diverse semantic registers.

3. Working Memory, Central Executive Control, and Mental Reversal

Reflected in Item 3 (Mental Reversal: Counting forwards 5 to 1, and spelling “WORLD” backwards), this construct tests the operational efficiency of the central executive within Alan Baddeley’s working memory model. Reversing symbolic sequences demands continuous active mental manipulation, continuous online suppression of habitual forward associations, and sustained visual-spatial/orthographic rehearsal.

4. Anterograde Episodic Verbal Memory: Encoding, Retrieval, and Recognition

Quantified via Item 4 (First Recall) and Item 15 (Second Recall), this domain evaluates the multi-phase memory system. By embedding an intervening cognitive load between encoding, initial recall, and final delayed recall, the 3MS measures both short-delay and longer-delay retention. The psychometric construct specifically isolates spontaneous generative retrieval (hippocampal-prefrontal axis) from cue-assisted retrieval and multiple-choice recognition (distinguishing storage loss from retrieval slowing).

5. Temporal and Spatial Orientation

Covered by Items 5 and 6, orientation reflects ongoing contextual tracking, spatial navigation, and environmental vigilance mediated by bilateral parahippocampal and posterior parietal networks. The graded scoring of date (day, month, year, day of week, season) isolates absolute temporal accuracy from minor drift.

6. Confrontational Language: Naming and Semantic Association

Measured in Item 7 (Naming five body parts: forehead, elbow, chin, knuckle, shoulder), this construct taps lexical retrieval, semantic access, and visual confrontation naming. The items sample high-frequency body loci alongside lower-frequency anatomical boundaries (e.g., “knuckle”), probing language degradation associated with left temporal neocortical dysfunction.

7. Executive Category Fluency and Divergent Thinking

Item 8 (Four-Legged Animals in 30 seconds) operationalizes verbal semantic fluency, which requires strategic search through semantic memory networks, clustering, switching, and rapid executive execution mediated by the left prefrontal cortex and anterior cingulate.

8. Abstract Conceptualization and Inductive Reasoning

Item 9 (Similarities: Arm-leg, Laughing-crying, Eating-sleeping) evaluates semantic abstraction, categorical categorization, and verbal reasoning. The construct contrasts concrete descriptive associations (e.g., “both have joints”) with superordinate abstract category formation (e.g., “limbs” or “extremities”), isolating executive frontal lobe capacity.

9. Expressive and Receptive Language Mechanics

Operationalized across Item 10 (Repetition of complex phonetic sequences), Item 11 (Reading Comprehension and Execution of a written imperative command), and Item 12 (Spontaneous Writing), these tasks test perisylvian language circuitry, speech repetition accuracy, syntactic parsing, and graphomotor execution.

10. Visuoconstructional Praxis and Motor Execution

Covered by Item 13 (Copying Two Overlapping Pentagons) and Item 14 (Three-Stage Command), these tasks measure spatial coordinate mapping, line angle orientation, intersecting topological relations, and sequential ideomotor praxis.

Theoretical Framework

The structural design of the Modified Mini-Mental State Test is grounded in classical neuropsychological theory and psychometric test construction principles. Historically, bedside mental status screening emerged from the behavioral neurology tradition championed by Kurt Goldstein, Norman Geschwind, and A.R. Luria. Alexander Luria’s neurodynamic framework posited that complex mental activities are executed through organized functional systems composed of distinct cortical and subcortical zones working in concert across three principal units: (1) the unit for regulating tone and waking state (reticular-limbic axis), (2) the unit for obtaining, processing, and storing information (posterior cortical regions—occipital, parietal, temporal), and (3) the unit for programming, regulating, and verifying mental activity (frontal-prefrontal cortex).

The classical MMSE was substantially skewed toward Luria’s second unit, relying heavily on posterior cortical functions such as auditory repetition, confrontational naming, reading, and simple construction, while largely omitting Luria’s third unit (executive regulation, generative fluencies, abstract concept formation, and mental flexibility). Teng and Chui deliberately anchored the 3MS within a broader neuroanatomical model by systematically adding tasks that recruit the frontal-subcortical and prefrontal functional systems:

• Prefrontal-Subcortical Circuits: The addition of verbal fluency (Item 8) and similarities (Item 9) explicitly interrogates the dorsolateral prefrontal cortex (DLPFC) and its reciprocal loops with the caudate nucleus and thalamus.
• Cognitive Aging and Processing Capacity Theories: Drawing from Craik and Salthouse’s cognitive aging models and Tulving’s episodic memory conceptualization, the 3MS recognizes that cognitive aging impairs self-initiated memory search far earlier than recognition. The inclusion of graded retrieval (spontaneous $\rightarrow$ cued $\rightarrow$ recognition) operationalizes Tulving’s encoding specificity principle, distinguishing primary encoding deficits from secondary retrieval bottlenecks.
• Classical Psychometric Test Theory: From a classical test theory (CTT) perspective, the original MMSE suffered from restricted item difficulty variance; items were clustered at low difficulty levels, rendering the instrument insensitive to early decline. By establishing graded scoring distributions (0 to 8 points on year identification; 0 to 4 points per pentagon) and introducing time-constrained generative tasks, Teng expanded the information curve of the test, extending its psychometric precision across a broader range of the latent cognitive trait continuum ($ heta$).

Validity

The validity of the 3MS has been established across extensive clinical, psychogeriatric, and epidemiological validation programs over four decades.

Construct and Structural Validity

Construct validity has been affirmed through robust correlations with comprehensive, multi-hour neuropsychological test batteries. In the landmark Canadian Study of Health and Aging (CSHA; McDowell et al., 1997; Bravo & Hébert, 1997), 3MS composite scores demonstrated powerful monotonic correlations with clinical dementia severity stages established by the Clinical Dementia Rating (CDR) scale ($r_s = -.78$ to $-.84$) and the Global Deterioration Scale (GDS). Furthermore, convergent validity analyses demonstrate strong correlations with the Wechsler Adult Intelligence Scale (WAIS) subtests, particularly Information ($r = .74$), Similarities ($r = .68$), and Digit Span Backwards ($r = .61$).

Criterion and Concurrent Validity

In direct head-to-head concurrent validity comparisons with the original MMSE, the 3MS consistently exhibits superior statistical properties. Teng and Chui (1987) demonstrated in their initial clinical cohort that the 3MS achieved higher diagnostic concordance with consensus clinical neurological diagnoses than the standard MMSE. In a seminal epidemiological investigation of 1,043 community-dwelling and institutionalized older adults, McDowell et al. (1997) reported that the area under the receiver operating characteristic curve (AUC-ROC) for distinguishing dementia from cognitive health was significantly greater for the 3MS ($ ext{AUC} = .92$) than for the MMSE ($ ext{AUC} = .87$). Utilizing an optimal empirical cutoff score of$77/78$ or $79/80$, the 3MS yielded a clinical sensitivity of 88% to 94% and specificity of 85% to 90%, whereas the MMSE at its conventional cutoff of $23/24$ demonstrated lower sensitivity (68% to 79%) in detecting early-stage neurocognitive disorders.

Discriminant Validity

The 3MS demonstrates clear discriminant validity between cognitively intact older adults with major depressive disorder and those with true degenerative dementia. Lamarre and Patten (1991) evaluated the 3MS in a broad psychiatric inpatient and outpatient sample, finding that depressed elderly patients scored significantly higher than those with neurodegenerative etiologies, particularly on the delayed recall and spatial orientation sub-dimensions. Moreover, the tiered memory retrieval structure discriminates cortical dementias (such as Alzheimer’s disease, where cued recall fails to elicit marked score improvement due to consolidated encoding storage loss) from subcortical-vascular disorders and pseudodementia (where cued recall and multiple-choice recognition substantially normalize total performance).

Reliability

The psychometric reliability of the 3MS has been documented across internal consistency, test-retest stability, and inter-rater reproducibility.

Internal Consistency

Estimates of internal consistency via Cronbach’s alpha for the 100-point total score range from $.87$ to $.91$ across community-dwelling epidemiological samples (Bravo & Hébert, 1997; McDowell et al., 1997). This demonstrates that despite measuring diverse cognitive domains, the constituent items reflect a unified latent dimension of general cognitive functioning. Split-half reliability coefficients (Spearman-Brown corrected) consistently exceed $.88$.

Test-Retest Stability

Because the 3MS is frequently employed in longitudinal epidemiological tracking and pharmacological monitoring, temporal stability is critical. Teng et al. (1990) demonstrated test-retest correlation coefficients of $r = .91$ to $.93$ over a short-term interval of 1 to 4 weeks among stable outpatients. In the CSHA cohort, Bravo and Hébert (1997) assessed test-retest reliability across a 3-month window in 143 community elders, finding an intraclass correlation coefficient (ICC) of $.91$ ($95%\text{ CI } [.88, .93]$). Furthermore, Correa, Perrault, and Wolfson (2000) calculated the Reliable Change Index (RCI) for the 3MS, determining that a longitudinal score drop of 5 points or greater reflects true clinical deterioration beyond measurement error ($p < .05$).

Inter-Rater Reliability

Inter-rater reliability coefficients for the 3MS are exceptionally high, driven by the operationalized, objective scoring criteria delineated in the testing manual. Bravo and Hébert (1997) observed an overall inter-examiner intraclass correlation of $.98$ across independent raters evaluating live and videotaped administrations. Even on items susceptible to subjective interpretation, such as pentagon intersection quality and similarity abstraction, inter-rater concordance coefficients ($kappa$) consistently exceeded $.86$.

Factor Analysis

Extensive exploratory factor analyses (EFA) and confirmatory factor analyses (CFA) have delineated the latent internal architecture of the 3MS across diverse international cohorts.

Exploratory Factor Structure

Early factor analytic evaluations of the 3MS typically extracted between four and five robust primary factors accounting for 54% to 62% of the total variance:

• Factor 1: Psychomotor Speed and Executive Praxis: High factor loadings for Four-Legged Animals (loading .72), Mental Reversal/World Backwards (.68), and Similarities (.65).
• Factor 2: Temporal and Spatial Orientation: High loadings for Temporal Orientation (.84), Spatial Orientation (.78), and Birth Date/Place (.64).
• Factor 3: Episodic Verbal Memory: High loadings for First Recall (.88) and Second Delayed Recall (.89).
• Factor 4: Language and Auditory Comprehension: High loadings for Three-Stage Command (.70), Naming (.66), Repetition (.61), and Sentence Writing (.58).
• Factor 5: Visuoconstruction: High isolated loading for Pentagon Copying (.82).

Confirmatory Factor Analysis and Model Fit

Modern structural equation modeling using Confirmatory Factor Analysis (CFA) has validated a hierarchical, second-order model wherein a overarching general cognitive factor ($g$) drives four distinct first-order factors: (1) Verbal Memory/Retention, (2) Orientation, (3) Language & Visuospatial Praxis, and (4) Executive Working Memory/Abstraction. Across diverse age groups, this hierarchical model demonstrates acceptable to excellent fit indices:

• Comparative Fit Index (CFI) = $.94$ to $.96$
• Tucker-Lewis Index (TLI) = $.93$ to $.95$
• Root Mean Square Error of Approximation (RMSEA) = $.048$ ($90%\text{ CI } [.042, .054]$)
• Standardized Root Mean Square Residual (SRMR) = $.041$

Item response theory (IRT) investigations, particularly two-parameter logistic (2PL) and graded response models (GRM), confirm that items such as Temporal Orientation (Year) and Delayed Recall possess high discrimination parameters ($a > 1.8$), providing precise measurement precision along the mild-to-moderate impairment continuum.

Instrument / Measurement Tool

The Modified Mini-Mental State Test (3MS) is an examiner-administered cognitive performance scale designed for direct, one-on-one bedside or office administration. Standard administration requires approximately 10 to 15 minutes.

• Instrument Type: Standardized clinician-administered neurocognitive screening instrument.
• Administration Format: Direct interactive clinical examination utilizing paper test forms, pencil, visual stimulus cue cards (e.g., for reading “CLOSE YOUR EYES”), and blank unlined response sheets.
• Total Score Range: 0 to 100 points, calculated as the direct linear sum of the 15 operational test components.
• Item Count: 15 structured task modules incorporating multi-level sub-items.
• Response and Scoring Rules by Task Module:
  • 1. When and Where Born (Max 5 points): Date: dd (1,0), mm (1,0), yyyy (1,0); Place: city/town (1,0), province/state (1,0).
  • 2. Registration (Max 3 points): Immediate repetition of three words: “Shoes” (1,0), “Blue” (1,0), “Modesty” (1,0). Examiner may repeat words up to five times until learned, but only the initial attempt is scored.
  • 3. Mental Reversal (Max 7 points): Counting backwards 5 to 1: Accurate (2), 1 or 2 errors (1), >2 errors (0). Spelling “WORLD” backwards (D-L-R-O-W): 1 point for each correct letter placed before its preceding counterpart (Max 5 points).
  • 4. First Recall (Max 9 points): Recall of the 3 registration words. For each word: Spontaneous recall = 3 points; Category-cued recall = 2 points; Multiple-choice recognition = 1 point; Missed completely = 0 points.
  • 5. Temporal Orientation (Max 15 points): Today’s date: accurate = 3, missed by 1–2 days = 2, missed by 3–5 days = 1, missed by >5 days = 0. Month: accurate or within 5 days = 2, missed by 1 month = 1, missed by >1 month = 0. Year: accurate = 8, missed by 1 year = 4, missed by 2–5 years = 2, missed by >5 years = 0. Day of the week: accurate = 1, missed = 0. Season: accurate or within a month = 1, missed = 0.
  • 6. Spatial Orientation (Max 5 points): Province/State (2,0), Country (1,0), City/Town (1,0), Building/Location (Hospital, Store, Home) (1,0).
  • 7. Confrontational Naming (Max 5 points): Accurately naming five pointed body parts: Forehead (1,0), Elbow (1,0), Chin (1,0), Knuckle (1,0), Shoulder (1,0).
  • 8. Category Fluency: Four-Legged Animals (Max 10 points): Examiner sets a 30-second timer; respondent names as many four-legged animals as possible. 1 point for each unique, valid animal up to a ceiling of 10 points.
  • 9. Similarities (Max 6 points): Abstract similarity identification: (a) Arm-Leg: Limbs/extremities = 2, Body parts/joints/bend/move = 1, Weak/no similarity = 0. (b) Laughing-Crying: Feeling/emotion = 2, Expressions/sounds/reactions = 1, Weak/no similarity = 0. (c) Eating-Sleeping: Essential bodily functions/survival needs = 2, Bodily activities/relaxing/good for you = 1, Weak/no similarity = 0.
  • 10. Repetition (Max 5 points): Phrase repetition: “I would like to go home (out)”: Correct = 2, 1–2 missed/wrong words = 1, >2 missed = 0. Difficult phrase: “No ifs” (1,0), “ands” (1,0), “or buts” (1,0).
  • 11. Read and Obey (Max 3 points): Presentation of cue card stating “CLOSE YOUR EYES”: Obeys command without verbal prompt = 3, obeys after examiner prompt (“Do what it says”) = 2, reads aloud only without executing = 1, neither reads nor executes = 0.
  • 12. Writing (Max 5 points): Respondent given 1 minute to write a target sentence (“I would like to go home/out”): Scored 0 to 5 points based on spelling, legibility, and grammatical completeness.
  • 13. Copying Two Pentagons (Max 10 points): Copying intersecting pentagons within 1 minute. Pentagon 1: 5 approximately equal sides = 4, 5 unequal sides (2:1 ratio) = 3, other enclosed figure = 2, 2 or more intersecting lines = 1, <2 lines = 0. Pentagon 2: Scored identically (0–4). Intersection enclosure: 4 corners formed by overlap = 2, non-four-corner overlap enclosure = 1, no intersection/no enclosure = 0.
  • 14. Three-Stage Command (Max 3 points): Multi-step motor praxis: “Take this paper in your left/right hand (1,0), fold it in half (1,0), and hand it back to me (1,0)”.
  • 15. Second Delayed Recall (Max 9 points): Delayed retrieval of the three target words following cognitive interference tasks. Scored identically to First Recall: Spontaneous = 3, Cued = 2, Multiple-choice = 1, Missed = 0 per word.
• Diagnostic Cutoff Thresholds: In standard clinical practice, scores $ge 80/100$ typically denote normal cognitive performance, scores between 70 and 79 suggest mild cognitive impairment or borderline dysfunction requiring formal neuropsychological evaluation, and scores $le 69/100$ indicate clinically significant cognitive impairment/dementia. Demographic adjustments based on age and years of education (Tschanz et al., 2002) should be applied to prevent diagnostic misclassification.

Permissions & Fee and Test Year

The Modified Mini-Mental State Test (3MS) was first published in 1987 by Evelyn Lee Teng and Helena Chui in the Journal of Clinical Psychiatry. Unlike the original MMSE—which was commercially copyrighted and proprietary licensing enforced by Psychological Assessment Resources (PAR)—the 3MS was placed into academic and clinical distribution through published scientific literature and institutional research networks.

The instrument was designed to be openly accessible for non-commercial clinical care, academic investigations, and public health research. Researchers and clinicians may utilize the test without royalty fees, provided full academic citation is accorded to the original authors (Teng & Chui, 1987). However, commercial electronic integration into proprietary electronic health records (EHRs) or commercial testing platforms may require institutional permissions. The standardized testing instructions, cue cards, and normative reference manuals remain accessible via institutional repositories, including the Canadian Study of Health and Aging (CSHA) archives and major academic medical centers.

References

• Bravo, G., & Hébert, R. (1997). Reliability of the Modified Mini-Mental State Examination in the context of a two-phase community prevalence study. Neuroepidemiology, 16(3), 141–148. https://doi.org/10.1159/000109681
• Correa, J. A., Perrault, A., & Wolfson, C. (2000). Reliable individual change scores on the 3MS in older persons with dementia: Results from the Canadian Study of Health and Aging. International Psychogeriatrics, 12(suppl. 2), 73–80. https://doi.org/10.1017/s104161020000700x
• Folstein, M. F., Folstein, S. E., & McHugh, P. R. (1975). “Mini-mental state”: A practical method for grading the cognitive state of patients for the clinician. Journal of Psychiatric Research, 12(3), 189–198. https://doi.org/10.1016/0022-3956(75)90026-6
• Lamarre, C. J., & Patten, S. B. (1991). Evaluation of the Modified Mini-Mental State Examination in a general psychiatric population. Canadian Journal of Psychiatry, 36(7), 507–511. https://doi.org/10.1177/070674379103600708
• McDowell, I. (2006). Measuring Health: A Guide to Rating Scales and Questionnaires (3rd ed.). Oxford University Press. https://doi.org/10.1093/acprof:oso/9780195165678.001.0001
• McDowell, I., Kristjansson, B., Hill, G. B., & Hébert, R. (1997). Community screening for dementia: The Mini-Mental State Exam (MMSE) and Modified Mini-Mental State Exam (3MS) compared. Journal of Clinical Epidemiology, 50(4), 377–383. https://doi.org/10.1016/s0895-4356(97)00060-7
• Nadler, J. D., Relkin, N. R., Cohen, M. S., Hodder, R. A., & Reingold, J. (1995). Mental status testing in the elderly nursing home population. Journal of Geriatric Psychiatry and Neurology, 8(3), 177–183. https://doi.org/10.1177/089198879500800308
• Schulzer, M., Calne, D. B., Snow, B., & Mak, E. (1993). A scoring error in the Mini-Mental State test. Canadian Journal of Psychiatry, 38(9), 603–605. https://doi.org/10.1177/070674379303800908
• Teng, E. L., & Chui, H. C. (1987). The Modified Mini-Mental State (3MS) examination. The Journal of Clinical Psychiatry, 48(8), 314–318.
• Teng, E. L., Chui, H. C., & Gong, A. (1990). Comparisons between the Mini-Mental State Examination (MMSE) and its modified version – the 3MS test. In Excerpta Medica (pp. 189–192). Elsevier Science Publishers.
• Tschanz, J. T., Welsh-Bohmer, K. A., Plassman, B. L., Norton, M. C., Wyse, B. W., Breitner, J. C., & Cache County Study Group. (2002). An adaptation of the Modified Mini-Mental State Examination: Analysis of demographic influences and normative data: The Cache County Study. Neuropsychiatry, Neuropsychology, and Behavioral Neurology, 15(1), 28–38.

Items of the Scale