 |
 |
|
|
|
Mild cognitive impairment (MCI) is a common condition that has been recognized as a prodrome to dementia (1). MCI is characterized by a subjective decline in memory without a change in functional ability (2–4). People with MCI typically complain of memory loss but have relatively normal general cognitive function. They maintain independence in instrumental activities of daily living (IADL), for example, cooking, finances, and driving, and some can still function in their occupational activities. Typically, history of memory loss for persons with MCI is corroborated by family members. Followed prospectively, between 6% and 14% of people with MCI convert to dementia each year (5–8). If people with MCI are followed for 8 years, nearly 100% convert to dementia (1). If MCI can be reliably diagnosed, it may be possible to start interventions that could prevent progression and conversion to dementia. Short and simple clinical screening tools to differentiate MCI from normal cognition and (or) early dementia would facilitate diagnosis. Diagnosing MCI is a challenge. The Mini-Mental State Examination (MMSE) (9), a widely used screening test for cognitive impairment, has been used primarily to screen patients with cognitive impairment to quantify cognitive deficits, identify dementia, and follow cognitive progression over time. The Standardized Mini-Mental State Examination (SMMSE) (10–12) has explicit guidelines for administration and scoring and improved interrater reliability, compared with the traditional MMSE. Although it is sufficiently specific for dementia, the MMSE is less sensitive in distinguishing among normal cognition, MCI, and dementia. MMSE scores are affected by education, which limits the MMSE’s usefulness as a screening tool. People with a median age of 70 years and over 7 years' education have a median MMSE score of 29; those with a median age of 70 years and only 3 years of education have a median score of 23 (13). The MMSE takes about 12 minutes to administer (10), which also limits its utility for busy clinicians who are often pressed for time. Given the influence of age and education on MMSE scores, and depending on the need to either rule in or rule out dementia, the lower limit of normal MMSE scores ranges from 24 to 26 (14–16). This means that the MMSE has at least a 5-point range within which people with normal cognition or patients with MCI might score (26–30). Thus people who are well-educated, have normal cognition, complain of memory loss, or have diagnosed MCI will usually score above 26. The MMSE lacks sensitivity and has a low ceiling; that is, many people with subjective complaints of memory loss and MCI may still score in the normal range on the MMSE, especially if they are well-educated and native English speakers. As a screening tool for persons with memory loss, the MMSE is useful in screening for dementia, but it lacks sensitivity in differentiating between people with normal cognition and those with MCI. Screening instruments are required by clinicians to reliably diagnose MCI and differentiate between normal cognition, MCI, and dementia. The current instruments available to diagnose MCI, for example, the Alzheimer’s Disease Assessment Scale-cognitive subscale (ADAS-cog, 17–19) and the Clinical Dementia Rating (CDR) scale (20), are not feasible for use by family doctors or other clinicians in the clinical setting because they take too long to administer. New instruments are required that have a higher ceiling and that are not as dependent on education. The AB Cognitive Screen (ABCS) (21) is a short screening cognitive assessment tool, specifically designed to differentiate normal cognition from MCI and dementia. We designed the ABCS to measure early changes in cognition and to differentiate between people with normal cognition and patients with MCI. At the same time, we developed the test to have a low enough ceiling to follow people with dementia and to measure changes in cognition, even in the later and more severe stages. The test has sections on orientation, registration, clock drawing, delayed recall, and verbal fluency. We designed the ABCS to be less affected by literacy; thus spelling, reading, and writing are not included. The test requires about 3 minutes to administer and score. The purpose of this study was to compare the sensitivity of the SMMSE with that of the ABCS in differentiating people with normal cognition from those with MCI. We hypothesized that the ABCS would be superior to the SMMSE in differentiating between people with MCI and those with normal cognition. Our secondary hypothesis was that the ABCS would be less affected by age or education than would the MMSE. MethodsABCS Description The ABCS evaluates 5 domains: orientation, registration, clock drawing, delayed recall, and word fluency. Orientation is assessed with 5 parameters, 4 relating to time (that is, year, month, date, and day) and 1 to place (that is, country). The subject is asked such questions as, “What year is it?” A correct response to each item scores 5 points. Near misses (for example, responding “1999” when it is 2004) score 2 points. The maximum score is 25. In word registration, the rater says 5 words to the subject, and the subject is asked to repeat all 5 words at once. One point is given for each word repeated. If all 5 are not repeated, the rater says all 5 words again, and the subject is asked to repeat them; 1 point is given for each word repeated. If the subject repeats all 5 before the fifth trial, full points are assumed for the remaining trials. Registration stops when the subject repeats all 5 words or reaches a maximum of 5 trials. The subject is then asked to remember the words for the delayed recall task later in the test. This version uses 5 words, mixes concrete and abstract terms, and involves several learning trials to avoid the floor and ceiling effects; the maximum score is 25. The clock drawing task tests executive ability and has been shown to be sensitive to early cognitive impairment (22). The subject is given a blank paper and a pencil and is asked to draw a clock with all the numbers on it and to set the time to 10 past 11. One minute is given to complete this task. A transparent scoring template is provided that divides the clock into quadrants and shows the correct placement of the hands within a range. The rater puts the transparency over the subject’s clock and adjusts it in relation to the hands and numbers to achieve the highest possible score. Two points are given for each number in the correct position and 1 point for each number out of place. One point is deducted for each number repeated. Three points are given for each hand correctly placed within the range lines, 2 points for a hand crossing the lines, and 1 point for a hand drawn outside the range. One point is deducted if the 2 hands do not join at a pivot point; the maximum score is 30. The scoring scheme with the template measures a wide range of scores with high reliability. In the delayed recall task, each subject is asked to recall the 5 words learned in the word registration task in 10 seconds. Five points are given for each word correctly recalled. The maximum score on this task is 25. Delayed recall is sensitive to early changes in cognition, particularly short-term memory. Word fluency is sensitive to early changes in cognitive function. In the word-fluency task, the subject is asked to name as many animals as possible in 1 minute. One point per animal name is given; the maximum score is 30. The maximum score for the entire ABCS is 135. Subjects We included patients and their caregivers who presented to 4 different geriatric specialty memory clinics, located in Hamilton, Paris, Niagara Falls, and Grand Bend, Ontario, between October 10, 2002, and August 14, 2003. Patients presented for assessment and treatment of cognitive impairment, and the diagnosis of dementia was made according to DSM-IV criteria. A geriatrician made the diagnosis of MCI after a comprehensive geriatric assessment. MCI was diagnosed in patients who had the following criteria: subjective memory loss with corroboration by a family member or friend, no deficits on IADL function, and no dementia. Caregivers and family members with normal cognitive function were recruited to the study to act as control subjects. Each control subject was asked if he or she had a problem with memory. We did not include any control subject in the study who admitted to a memory problem. Participants were included if they were aged 55 years or older and were able to communicate verbally in English. We excluded patients with depression (that is, Geriatric Depression Scale score of 7 or more, 23). All participants gave consent. Assessments Participants provided demographics including age, sex, years of education, first language, and history of memory loss, if present. Raters were blind to the results of the clinical assessment and to each patient’s diagnosis. A trained rater in the clinic randomly scored each participant’s SMMSE and ABCS. Statistical Analysis We entered data into SPSS 11.0 (24) and analyzed the information, using frequency counts, Student’s t test, logistic regression, and receiver operating characteristics (ROC) curves. We conducted the analyses, holding age and education constant. ResultsTable 1 shows subjects’ demographics. This study included 235 participants, 124 diagnosed with MCI, and 111 with normal cognitive function. Subjects’ mean age was 76 years, and 58% were women. The mean number of years of education was 12.2 and the mean duration of memory loss for patients with MCI was 21.4 months.
Table 2 compares the scores on the SMMSE and ABCS for people with normal cognition and MCI for different age groups and educational levels. Overall, normal subjects scored 0.6 points higher than MCI subjects on the SMMSE (P = 0.040). Normal participants scored about 7 points higher than MCI subjects on the ABCS (ABCS score 111.7 vs 104.6, P < 0.001). For those aged over 75 years, the difference in SMMSE scores between normal and MCI subjects was not significant (P = 0.299), while the difference on the ABCS (that is, 108.6 and 102.7) was statistically significant (P = 0.002). Among subjects aged between 55 and 75 years, the difference in SMMSE scores between normal subjects and those with MCI (SMMSE score 28.3 and 27.8) was not statistically significant (P = 0.205), while the difference in scores between normal and MCI subjects on the ABCS (that is, 114.1 and 107.1) was significant (P = 0.001).
Subjects were divided according to their level of education. The difference in SMMSE scores between normal subjects and those with MCI was statistically significant for persons with 12 or more years of education (SMMSE score 28.5 and 27.7) (P = 0.019) but was not significant for those with less than 12 years of education (SMMSE score 27.0 and 26.4) (P = 0.252). On the ABCS, the difference between normal subjects and MCI subjects in those with 12 or more years of education (ABCS score 114.6 and 106.1) and less than 12 years of education (ABCS score 107.8 and 102.5) were both statistically significant (P = 0.000 and P = 0.016, respectively). Figure 1 shows the ROC curves. Table 3 shows the area under the curve of the SMMSE and the ABCS. The ROC used different scores on the 2 tests to expresses the sensitivity and specificity of an MCI diagnosis. For the ABCS, the area under the curve was 0.7 (95%CI, 0.62 to 0.74). For the SMMSE, the area under the curve was 0.6 (95%CI, 0.53 to 0.67). Both were significant (P = 0.05). Figure 1 Sensitivity and specificity of the ABCS and SMMSE 
Table 4 displays the influence of education and age on the ABCS and SMMSE scores. The combined scores of normal and MCI subjects show that for both SMMSE and ABCS instruments, subjects aged 55 to 75 years scored higher than those aged 76 to 98 years. The combined scores also show that for both SMMSE and ABCS instruments, subjects with 12 to 17 years of education scored higher than those with 0 to 11 years of education. All differences were statistically significant. Within each group of normal and MCI subjects, SMMSE and ABCS scores were still higher for subjects aged 55 to 75 years than for those aged 76 to 98 years, and scores were higher for subjects with 12 to 17 years of education than for those with 0 to 11 years of education, with one exception: education did not significantly affect the ABCS scores of MCI subjects (ABCS score 106.1, vs 102.5; P = 0.59).
DiscussionThis study compared the ability of 2 different instruments to discriminate between people with MCI and people with normal cognition. These results suggest that the ABCS is as effective as the SMMSE and possibly more so. The ABCS is shorter and easier to administer and score, with clear and explicit guidelines for administration and scoring. The ABCS takes less than half the time to administer and score than the SMMSE. The differences in scores between those with MCI and normal subjects on the SMMSE ranged from 0.5 to 0.8 points. Although most differences were statistically significant, the difference was less than 1 point, which is not clinically significant. Attempts to increase the sensitivity of the MMSE and to provide a greater range of intermediate data points have resulted in instruments like the Modified Mini-Mental State Examination (3MS, 25), scored out of 100. The administration and scoring time is increased with this process. For the ABCS and SMMSE, younger and more educated subjects scored higher than older and less educated subjects. According to these data, the ABCS shows promise as a shorter cognitive screen with a higher ceiling that may diagnose MCI. This study has some weaknesses: we compared the ABCS with the SMMSE, and the SMMSE is not a gold standard for MCI. The diagnosis of MCI was based on established clinical criteria (2–4) performed by an experienced clinician after a comprehensive geriatric assessment that included measures of ADL and mood and behaviour, but no in-depth neuropsychological testing was performed. The MMSE does not discriminate well between MCI and normal cognition because the ceiling is too low on this test. The ABCS will need to be compared with gold standards such as the CDR (20) or the Alzheimer’s Disease Assessement Scale/MCI version (ADAS-cog/MCI, 18). As sensitive as any single test might be in the diagnosis of MCI, a more accurate diagnosis may result from serial tests measuring change over time and not from a single administration of any test. Serial assessments over time may pick up early cognitive changes. The strengths of this study are the large sample size, the use of subjects from multiple centres, ratings performed by several different raters, random order of test administration, and raters who were blinded to the diagnosis. Also, the diagnosis of MCI was made after a comprehensive examination by a specialist geriatrician and included a physical exam; assessments of function, behaviour, and mood; and a comprehensive history. While the ABCS is better than the SMMSE in differentiating normal cognition from MCI, these results show that age and education affected both the ABCS and the SMMSE. Age influenced both tests in normal subjects and in MCI subjects. Education significantly influenced ABCS and SMMSE scores, except for the ABCS in MCI subjects. Those with higher education had higher scores, but this difference was not statistically significant. We need a larger sample to establish the means and ranges for the following 4 separate groups: those aged over 75 years with less than 12 years of education, those aged over 75 years with 12 or more years of education, those aged 75 years or under with less than 12 years of education and those aged 75 years or under with 12 or more years of education. This study shows that the ABCS is a short test that can differentiate between normal cognition and MCI. It also shows how age and education affect the scores and confound the differentiation of normal cognition and MCI. Our sample population was not large enough to demonstrate conclusively the appropriate cut-off points or the range of scores for normal cognition and mild cognitive impairment. Further studies need to be done that take into account age and education level, and possibly even the rate of cognitive decline over time, to fine tune the diagnosis of MCI and to differentiate it reliably from normal cognition. FundingThis work was supported by an award from The Scottish Rite Charitable Foundation of Canada. References1. Morris JC, Storandt M, Miller JP, McKeel DW Jr, Price JL, Rubin EH, and others. Mild cognitive impairment represents early-stage Alzheimer’s disease. Arch Neurol 2001;58:397–405. 2. Ivnik RJ, Malec JF, Smith GE, Tangalos EG, Petersen RC, Kokmen E, and others. Mayo’s older Americans normative studies: WAIS-R, WMS-R, and AVLT norms for ages 56 to 97. Clin Neuropsychol 1992;6(Suppl):1–104. 3. Smith GE, Petersen RC, Parisi JE, Ivnik RJ. Definition, course and outcome of mild cognitive impairment. Aging, Neuropsychology and Cognition 1996;3:141–7. 4. Petersen RC. Mild cognitive impairment: transition from aging to Alzheimer’s disease. In: Iqbal K, editor. Alzheimer’s. Diseases: advances in etiology, pathogenesis and therapeutics. West Sussex (UK): J Wiley & Sons; 2001. p 141-51. 5. Tierney MC, Szalai JP, Snow WG, RH Fisher, A Nores, G Nadon, and others. Prediction of probable Alzheimer’s disease in memory-impaired patients: a prospective longitudinal study. Neurology 1996;46:661–5. 6. Tierney MC, Szalai JP, Snow WG, RH Fisher, T Tsuda, H Chi, and others. A prospective study of the clinical utility of ApoE genotype in the prediction of outcome in patients with memory impairment. Neurology 1996;46:149–54. 7. Bowen J, Teri L, Kukull W, McCurry SM, Larson EB. Progression to dementia in patients with isolated memory loss. Lancet 1997;349:763–5. 8. Daly E, Zaitchik D, Copeland M, Schmahmann J, Gunther J, Albert M. Predicting conversion to Alzheimer’s disease using standardized clinical information. Arch Neurol 2000;57:675–80. 9. Folstein M, Folstein S, McHugh P. Mini-mental state. A practical method for grading the cognitive status of patients for the clinician. J Psychiatric Res 1975;12:189–98. 10. Molloy DW, Alemayehu E, Roberts R. Reliability of a standardized Mini-Mental State Examination compared with the traditional Mini-Mental State Examination. Am J Psychiatry 1991;148:102–5. 11. Molloy DW, Standish TIM. A guide to the Standardized Mini-Mental State Examination. Int Psychogeriatr 1997;9(Suppl 1):87–94. 12. Vertesi A, Lever JA, Molloy DW, Sanderson B, Tuttle I, Principi E, and others. Standardized Mini-Mental State Examination: use and interpretation. Can Fam Physician 2001;47:2018–23. 13. Crum RM, Anthony JC, Bassett SS, Folstein MF. Population-based norms for the Mini-Mental State Examination by age and educational level. JAMA 1993;269:2386–91. 14. van der Cammen TJ, van Harskamp F, Stronks DL, Passchier J, Schudel WJ. Value of the Mini-Mental State Examination and informants’ data for the detection of dementia in geriatric outpatients. Psychol Rep 1992;71:1003–9. 15. Kay DWK, Henderson AS, Scott R, Wilson J, Richwood D, Grayson DA. Dementia and depression among the elderly living in the Hobart community: the effect of the diagnostic criteria on the prevalence rates. Psychol Med 1985;15:771–88. 16. O’Connor DW, Pollitt PA, Hyde JB, Fellowes JL, Miller ND, Brook CPB, and others. The reliability and validity of the Mini-Mental State in a British community survey. Psychiatry Res 1989;23:87–96. 17. Rosen WG, Mohs RE, Davis KL. A new rating scale for Alzheimer’s disease. Am J Psychiatry 1984;141:1356–64. 18. Mohs RE, Knopman D, Petersen RC, Ferris S, Ernesto C, Grundman N, and others. Development of cognitive instruments for use in clinical trials of antidementia drugs: additions to the Alzheimer’s Disease Assessment Scale that broaden its scope. Alzheimer Dis Assoc Disord 1997;11:S33–S39. 19. Standish T, Molloy DW, Bédard M, Layne EC, Murray EA, Strang DG. Improved reliability of the standardized Alzheimer’s Disease Assessment Scale (SADAS) compared with the Alzheimer’s Disease Assessment Scale (ADAS). J Am Geriatr Soc 1996;44:712–16. 20. Hughes CP, Berg L, Danzinger WL, Cohen LA, Martin RL. A new clinical scale for the staging of Dementia. Br J Psychiatry 1982;140:566–72. 21. Molloy W. abcdes Administration & Scoring Guideline: a practical guide to the abcde’s short comprehensive screen. Dundas: Newgrange Press Ltd; 2003. p 11–8. 22. Juby A, Tench S, Baker V. The value of clock drawing in identifying executive cognitive dysfunction in people with a normal Mini-Mental State Examination score. CMAJ 2002;167:859–64. 23. Yesavage JA. Geriatric Depression Scale. Psychopharmacol Bull 1988;24:709–11. 24. SPSS Inc. SPSS for Windows 11.5. Chicago (IL): SPSS Inc; 2002. 25. Teng EL, Chui HC. The Modified Mini-Mental State (3MS) examination. J Clin Psychiatry 1987;48:314–8. Author(s)1. Professor of Medicine, St Peter’s McMaster Chair for Studies in Aging, St Peter’s Hospital, Hamilton, Ontario. 2. Research Coordinator, St Peter’s Centre for Studies in Aging, St Peter’s Hospital, Hamilton, Ontario. 3. Assistant Clinical Professor, McMaster University, Hamilton, Ontario; Director, Research and Evaluation Regional Geriatric Program Central, St Peter’s Centre for Studies in Aging, St Peter’s Hospital, Hamilton, Ontario. Address for correspondence: Dr DW Molloy, St Peter’s Centre for Studies in Aging, St Peter’s Hospital, 88 Maplewood Avenue, Hamilton, ON L8M 1W9 e-mail: wmolloy@stpetes.ca
1 | 2
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
