These tests are indeed useful markers for testing cognitive impairment. There are other health issues that can mimic dementia, how could we differentiate these and identify dementia as the true cause. I think a more sensitive testing would be good but would it be too intrusive and worrying for the participant?

The MMSE and other short cognitive "bedside" tests are not very good at helping the clinician diagnose pre-dementia mild cognitive impairment. See http://www.ncbi.nlm.nih.gov/pubmed/18579155. This paper is updated for a new book here http://www.springer.com/medicine/neurology/book/978-1-4471-2451-1?changeHeader. Tests also struggle with early dementia. This is why a more comprehensive neuropsychological battery combined with clinical history and examination by a suitably qualified person is recommended.

Well, if you do like neuropsychology and wish to understand your patient's memory functioning, you probably might think of using theoretically grounded memory tests such as free and cued retrieval tests and tests with in depth analyses of responses such as the California Verbal Learning Test.

The CANTAB Paired Associates Learning test has been used frequently in relation to this

I too think that the MMSE is not sufficiently accurate. The Dementia Rating Scale by Mattis in my opinion is a good tool for early diagnosis.

problems with recall of recent learned memories and executive dysfuction seem to be the cogntive domains most effected in early and pre-clinical dementia. Try Belleville et al (07) DOI: 10.1037/0894-4105.21.4.458.

I recommend California Verbal Learning Test with additional analysis of the learning curve during List A free recall.

In a study we conducted in 4 european countries, we used MMSE and MoCA.

http://www.mocatest.org/

The MMSE is the most commonly used cognitive screen, but is not most accepted for its sensitivity. MMSE is best suited for distinguishing gross cognitive impairment, but not for detecting damage to abstract reasoning, executive functioning, and visual perception, which MoCA is. MoCA has been shown to have a lesser completion time (by 5 minutes) and 50% more sensitivity towards cognitive impairments. Conclusions are that it is well tolerated, accurate, and provides additional information over the MMSE. Additionally, MoCA is available in over 25 languages via free internet download, without restrictions, and can be administered by any member of the care team.

I had also contacted Dr. Z Nasreddine regarding translations into Finnish and he was very helpful :)

Indeed I remember of a paper in which the test that discrimitated most between the patients whose will convert to dementia and whose won't was the Stroop task..For my clinical experience memory tests are not sufficient to detect the early onset of dementia. I agree with the suggestion of using memory and executive function tests

I think it is important to administer a neuropsychological battery sensitive to processing speed, orientation, attention, verbal learning and memory, language, visuoperception, gnosis, praxis and executive functions (all functions in a reasonable amount of time, approx. 45 minutes). In the Diagnostic Unit of Fundació ACE we adminiter the NBACE (neuropsychological battery of Fundació ACE) (Alegret et al., 2012) to all patients and we have found that it is useful to detect early changes before the onset of dementia. It includes the following tests: Temporal, Spatial and Personal Orientation; Digit spans (forwards and backwards), Block Design (abbreviated so that items 6 to 9 were scored only for accuracy (1 point) without a time bonus) and Similarities (abbreviated to the first 10 items) subtests of Wechsler Adult Intelligence Scale-Third Edition (WAIS-III); The Word List Learning test from the Wechsler Memory Scale-Third Edition (WMS-III) (without using the interference list); Repetition (2 words and 2 sentences); Verbal comprehension (to correctly execute 2 simple, 2 semi-complex and 2 complex commands extracted from the ADAS-cog and the Barcelona test battery; the abbreviated 15-item Boston Naming Test; the Poppelreuter test; Luria’s Clock test ; The 15-Objects test; the Automatic Inhibition subtest of the Syndrom Kurtz Test (SKT); Phonetic Verbal Fluency (words beginning with ‘P’ in one minute); Semantic Verbal Fluency (‘animals’ in one minute); and the Spanish version of the Clock Test.

We found that the 15-Objects tests is able to detect deficits in subjects in Mild Cognitive Impairment (it does clearly distinguish between healthy elderly subjects, mild cognitive impairment and dementia and it is very easy and quick to administer) and the Word List Learning Test from the WMS-III is useful to detect impairment and the determine which pattern of impairment (retrieval or storage). We have published about this issue. If you need something else, do not hesitate to contact with me. Sincerely

If you are talking about Alzheimer's etiology, definitely CVLT, BNT, comparing semantic vs phonemic fluency (semantic should be performed better), as well as the DKEFS trails (especially 4). MMSE is NOT sensitive enough to detect early changes. If you are talking about vascular dementia, then anything that looks at speed of processing and multitasking, low phonemic fluency, and difficulty planning on visuo-construction tasks.

If you want ot choose a single test which is most sensitive in nondemented elderly : do choose a Digit Symbol (WAIS-R) / Digit Symbol-Coding (WAIS-III).

It is superior to other in reflecting mild organic changes in subjects with MMSE between 26-30.

There is no one 'best' cognitive test. The MMSE was developed to detect dementia, and is not very sensitive to earlier, or more subtle, cognitive changes, as it does not include any test of executive function. Additionally, as others have pointed out, it has a ceiling effect which is most evident in those with large cognitive reserve (highly educated and/or with pre-morbid high innate intelligence). In clinical practice the MOCA, again as others have stated, has been shown to be more sensitive to early cognitive changes, and has the advantage of being readily available in multiple languages through the internet. Comprehensive neuropsychological testing is the preferred methodology for detecting early/subtle cognitive change, but takes considerable time (2+ hrs) depending on the battery of tests used. Trails B, Stroop and CVLT are three of the more useful tests in our experience for detecting early changes.

This depends on how you define "best".

If you are interesting in a cognitive assessment for delayed verification of all cause dementia then you have a couple of options - choice will depend on the time / resources you have.

There is a published literature on test accuracy of various tools for dementia diagnosis, however this is spread across the geriatric / neurology / psychiatry / psychology literature.

I am working with the Cochrane Dementia and Cognitive Improvement Group on a series of systematic reviews and meta-analyses of cognitive screening tests - watch this space.

I recommend a thorough neuropsychological evaluation. Too many things can impact cognitive functioning that have nothing to do with MCI amnestic or non-amnestic type. I did attach a review of neuropsychological predictors for MCI. If it doesn't load the citation is: Neuropsychological Predictors of Transition From Healthy Cognitive Aging to Mild Cognitive Impairment: The PATH Through Life Study Nicolas Cherbuin, Ph.D., Perminder Sachdev, M.D., Ph.D., F.R.A.N.Z.C.P.,Kaarin J. Anstey, Ph.D Am J Geriatr Psychiatry 18:8, August 2010

Short version:

1) Don't diagnose dementia UNTIL other causes of abnormal test score are EXCLUDED first

2) decide WHICH SYNDROME of dementia are you screening for. Different psychometric tests detect different types diffirently

3) dementia is a FUNCTIONAL impairment rather than a cognitive impairment or a specific pathological process; the environmental factors and social norms can DRAMATICALLY interfer with the clinical picture

4) in my experience, to distinguish dementia from other differential diagnosis, and to select the most probable dementia syndrome (the only SURE diagnosis of dementia is made by AUTOPSY) we need someone who followed up previous patients and worked with them for some time (years)

So, if you want to screen for ALL dementias, I strongly suggest using more than just one test tocapture early stages of common dementias

: any time-limited tasks test + executive functions + memory

Modified MMSE or original Folstein's MMSE + MoCA / EXIT25 / Clock drawing

Notice that MMSE has many advantages and disadvantages as well! it is copy-righted and may cost you money, though not much.

The early status change occurs earlier in time, then it can be followed with diagnoses. That's why we measure the reflex time and the usage time of the work memory. The slowdown of the usage time of the work memory is an early sign of the decrease of brain functions, and in that case the process can be slowed down or halted with certain methods.

Neuropsychological evaluation is the sine qua non of diagnosis of conditions such as dementia and mild cognitive impairment (see National Institute of Neurological Communicative Disorders and Stroke-Association Internationale pour la Recherche’ et l’Enseignement en Nourosciences: NINCDS-AIREN, Roman et al., 1993, Neurology, 43, 250-260; NINCDS-Alzheimers Disease and Related Disorders Association, McKhann et al., 1984, Neurology, 34, 939-944; and R. C. Petersen et al., 2001, AAN Practice parameter: Early detection of dementia: Mild cognitive impairment Neurology 56, 1133-1142).

However, MCI criteria has been all over the place. Before you can measure a thing you must define it. Attached (perhaps) is the following article: Quantification of five neuropsychological approaches to defining mild cognitive impairment Amy J. Jak, PhD, Mark W. Bondi, PhD et al. Am J Geriatr Psychiatry. 2009 May ; 17(5): 368–375. doi:10.1097/JGP.0b013e31819431d5.

The MMSE, although widely used, is not sensitive or specific enough for a diagnosis of mild cognitive impairment; de Jager CA, Schrijnemaekers AC, Honey TE, et al: Detection of MCI in the clinic: evaluation of the sensitivity and specificity of a

computerised test battery, the Hopkins Verbal Learning Test and

the MMSE. Age Ageing 2009; 38:455–460

I think in addition to all the previous answers I would add that the most sensitive and specific tests depend on the type of dementia that is suspected. For Alzheimer Disease (AD), the Free and Cued Selective Reminding Test performed better than other memory paradigms in a study by Sarazin. But AD can also manifest with more frontal-lobe or parietal-lobe like symptoms. Logopenic aphasia has been associated with AD. For Frontotemporal dementia, the best test/s is/are behavioural inventories. The best approach is a broad neuropsychological battery, with tests of memory, language, visuo-spatial abilities, and executive function tasks, plus assessment of behaviour and mood. In a computerized test battery we are developing at UBC memory for object-scene associations and the location of objects in the scene is the test that most reliably detects MCI. But we have not launched this battery yet.

I suggest you read this article:

Semenza, S. Mondini, F. Borgo, M. Pasini & M. T. Sgaramella (2003): Proper Names in Patients With Early Alzheimer’s Disease, Neurocase: The Neural Basis of Cognition, 9:1, 63-69.

Then you need to build your own test, tailored on the environment the people you test live in. For each patient I would run about ten controls and then match the pt's performance to them according to the single case statistical methods proposed by Crawford. It's a painful procedure, but if you want to detect early Alzheimer you cannot use MMSE and the likes

In a paper we're trying to publish, we used the Modified Mini-Mental State (3MS) test (Teng & Chui, 1987), with 117 patients at a memory disorders clinic who presented with cognitive impairment or dementia, and found that delayed recall from the 3MS test (cutpoint = 4, score range = 0 to 9) was 87% sensitive and 47% specific in predicting Alzheimer disease about 5 years later. This was comparable to consensus clinical diagnosis done about the same time, which was 82% sensitive and 45% specific. For a subset of patients with initial Mini-Mental State Examination (MMSE) >= 19 (the median), scores = 20 at initial visit, delayed recall was 78% sensitive and 83% specific (cutpoint

I can speak as the spouse of a man with fronto-temporal dementia. He does wonderfully on the Mini Mental, but cannot figure out the proper way to wear his shoes. It was a Godsend when we finally did the battery of psycho metric tests that gave us a diagnosis. His written perseveration and inability to reproduce tile pictures plus other tests got him a diagnosis. Keep on searching you may never know how much your work means to families.

I agree with the above comment, especially in using Modified MMSE or original Folstein's MMSE + MoCA / EXIT25 / Clock drawing as Tamer mentioned above. More importantly any of these tests should be used as "baseline" evaluations; the changes in results at 6 months and one year are more important than the initial score........

You can also use Dementia Rating Scale by Mattis as well as verbal fluency tests (FAS and animal naming) in combination with Clock-drawing test. Also, some sort of delayed-memory test should be used as well.

We have developed a cognitive instrument called the Audio Recorded Cognitive Screen (ARCS) that is administered by an audio device (eg MP3 player or portable CD player) to unsupervised patients who write their responses in a sopecially designed book for later scoring. It comprises a 12 word list learning task with three learning trials, delayed recall and recognition trials (modelled on the Hopkins Verbal Learning Test revised), verbal fluency measures, 10 pictured items for naming, a clock drawing task, and a complex attention/executive functioning test. The test lasts 34 minutes but it takes someone just a couple of minutes to score. Scoring is entered into an excel spreadsheet and produces demographically adjusted scores. (We have normed across the adult age range). Further details can be found at www.cognitionhealth.com if you are interested. We use it all the time in our Neuropsychiatry outpatients to triage need for further assessments. The list of publications relating to the ARCS is on the website.

There are multiple subtypes of dementia and there are also multiple reasons for failing neuropsychological screening tests. For those who have the later-onset form of Alzheimer's disease (affecting the medial temporal lobes in the first instance), tests of memory and learning are very sensitive. However, people can fail tests of memory and learning for many reasons ranging from anxiety and depression to brain tumours, strokes and Alzheimer's. Even a degenerative (progressive insidious) loss of memory is not necessarily a harbinger of future decline. Around half of those with impaired memory on cognitive screening will not progress to show Alzheimer's (possibly especially if they care for themselves in terms of nutrition and exercise).

As a practitioner Neuropsychologist in a specialist clinic for Young onset dementia I am no longer surprised to see people who "break the mould" and present to the clinic with unusual patterns of degenerative impairment - language, memory, thinking, action organisation, planning, spatial perception... so the chances of finding a "gold standard" cognitive/behavioural criterion for diagnosing dementia is possibly remote in those under 65/70.

In the older age group, tests of memory and evidence of difficulty with performing activities of daily living, taken together, can suggest there is a pathological process -- but only suggest. Cognitive/behavioural signs and symptoms should ideally be followed up by neurological (or neuropsychiatric) screening to see if there is a reversible cause for the problem and a comprehensive neuropsychological evaluation can help to clarify the position -- especially if the pattern of impairment does not follow the "classical" amnestic profile.

Thank you all for the useful suggestions, comments and references. I understand the difficulty of finding a "gold standard" test for the early signs of dementia, but I do hope to find a more calibrated set of tests which could provide a clearer picture than the standard MMSE or 3MS. As Christine highlights, multiple types of tests are also necessary!

I forget to point out in my previous answer that odor identification tests could be useful for the eraly signs of Alzheimer dementia. If you need some more information you can find it in our article regarding this topic in my RG profile: https://www.researchgate.net/publication/225759894_Odor_identification_and_cognitive_abilities_in_Alzheimers_disease.

Article Odor identification and cognitive abilities in Alzheimer’s disease

What about Trails A and B, or word generation exercises to help detect executive dysfunction?

We have been using MoCA (1) , an easy to use, short (about 10-15 minutes) test with good sensitivity and specificity for MCI.

Freely available for clincal and non-pharma research use. Also available in multiple languages:

http://www.mocatest.org/

1. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, Cummings JL, Chertkow H. The Montreal Cognitive Assessment, MoCA: a brief screening tool for mild cognitive impairment.J Am Geriatr Soc. 2005 Apr;53(4):695-9.

The most important is to detect what is not dementia by elimination of pseudomentia depression.

It is a probable diagnosis if the patient can read upside down and performe calculation like 103-4=99.

We could detect improvement and decline in nondemented adults using Trails B-A and the Digit Recall test (both forward and backward; Chan et al., 2010; J Nutrition Health Aging 14:224-230). We never used Trails in our clinical studies with AD or MCI, but, conversely, our MCI population showed ceiling effects with Clox-1 and DRS (while AD participants did not). Seems we should have included Trails for MCI/earliest cognitive decline... I would strongly suggest using Trails, as Ramesh suggested, and the Digit Recall. They take only minutes each, and can be presented in a very non-intimidating, almost game-like manner, so there is little to lose by incorporating them.

A few reflections from a researcher/clinician.

Thomas Shea makes some very good points. I agree that Trails A&B can be sensitive -- but they do have practice effects and I do not know of suitable parallel forms (something you need if there is going to be follow-up testing). Tasks such as Coloured Trails are somewhat easier than the originals and, in my hands somewhat less sensitive.

We tend to use the DKEF's number-letter alternation task as a complementary form to Trails A and B (as part of our ~4 hour cognitive assessments of people suspected of dementia) but the DKEFS task uses a much bigger scanning area than the original and also places subtly different demands on the participants. The Symbol-Digit Modalities test is another option to Trails (it does have multiple parallel forms for serial monitoring). SDMT makes similar demands to trails B in terms of spatial cognition, attentional processing and set shifting under time pressure.

Digits forwards and back can also be good - and I use it routinely -but it is often passed by people with temporal lobe pathology (and in a medico-legal context are often failed by people who want to persuade the examiner that they have a memory deficit). Just what counts as "failure" is difficult to define (and also probably needs to take account of premorbid competencies).

The MOCA (as helpfully recommended by Adrian Preda) is also very useful, (especially for spatial and dysexecutive deficits), and it also has some parallel forms but it can be over sensitive in people with low intellectual level and limited education. It is possibly over sensitive to the mild attentional deficits seen with small vessel damage. To maximise sensitivity I'd recommend adding verbal fluency (letter and category) and use the "optional" recognition test with the memory task in order to distinguish failures of recall (often executive) from amnestic forgetting (often AD).

However, as mentioned in my previous posting, all of these tasks are susceptible to failure as a consequence of non-degenerative problems (e.g., the effects of depression, chronic pain, hypothyroidism and high anxiety). There is inevitably going to be noise in the data and your "at risk" group of those who score poorly on your tests will still probably include 50% who happily recover and do not decline for many more years.

Dear Colleagues, I also recommend the MoCA - Montreal Cognitive Assessment. As indicated previously there is a website that discusses all the work that has been undertaken using the MoCA - http://www.mocatest.org/

I recommend measurement cognitive time. This is early, measurably demonstrates the problem, and you can track changes.

Do not forget to use the Lawton & Brody inventory of IADLs, since patients with MCI will still be able to take their medications and balance their checkbooks, whereas patients with early dementia will need their caregivers to help them with these tasks.

Bothh the VA-St louis University Mental Status(SLUMS) examination and the MOCA pick up early MCI and have been validated. The SLUMS has no copyright ie its free owned by the USA government and is very user friedly and has been highly adopted by clinicians. The SLUMS is available in a number of languages. it is predictive of long term outcomes.

I wish i knew all of the great things written here when we were doing our last study!

II would highlight the opportunity of re-conceptualizing the relationships between cognition & function in Mild Cognitive Impairment. Performance in activities of daily living should be "essentially intact" according to Petersen's criteria. Yet impairment in complex ADL seems to be already present in MCI [Perneczky & Coll. Age and Ageing 2006; 35: 240-5].

Bill paying, tracking current events & transportation are the items of greatest diagnostic utility to discriminate MCI from early Alzheimer's Disease, by informant-rated "Functional Activities Questionnaire" [Teng & Coll. Alzheimer Dis Assoc Disord. available in PMC 2011 December 1]. The fore-mentioned abilities are more impaired in AD patients, yet mild but measurable deficits on IADL are detectable in MCI subjects.

Using the ecologic Performance-Based Skill Assessment [UPSA (University of California, San Diego)] yielded a large effect size distinguishing cognitively healthy from MCI subjects (d=0.86). UPSA showed strong relationships with cognitive performances in speed (R2 = 0.37), episodic memory (R2 = 0.1), and semantic processing (R2 = 0.03).

Best regards, Mauro Colombo

It must be underlined that Mild Cognitive Impairment is the desciption of a consequence of an underlying mechanism and that Alzheimer Disease is a cause of dementia, but not the only one.

The important questions are: does MCI always lead to dementia, then is the cause of this possible dementia possibly curable?

My understanding from attending a conference here in Sydney earlier this year, was that as many as 80% of people with MCI develop progressive cognitive decline or dementia. Studies indicated that there was a substantial loss of the hippocampus in individuals diagnosed with MCI. There was some conversation that one of the main causes was due to hypertension.

In work that I have been doing for University in the past couple of weeks, I have read that its is possible for the hippocampus to be regenerated (brain plasticity). Again, I guess the question is, can this work if Tau and Beta Amyloid proteins are actively degrading brain tissue?

I would recommend the Addenbroke´s Cognitive Examination Revised, which has been used as screening test to detect early cognitive impairment and dementia. Reference paper: Mioshi et al. Int J Geriatr Psychiatry 2006; 21: 1078–1085.

The test has been recently modified by the authors. More information may be found in:http://www.neura.edu.au/frontier/research/test-downloads/

in my opinion questionaries like the mini mental state or iadl´s are less sensitiv. i would prefer test batteries underlining processing speed or working memory. i think one of the most sensitiv tests for cognitiv decline is the operation span test (unsworth, 2005). it can be downloaded for free http://www.millisecond.com/download/library/OSPAN/

I would also recommend the Addenbrooke's Cognitive Examination Scale - it assesses a range variety of cognitive functions; is short and easy to proceed

Kind Dorina Cadar,

in a similar blog, Roberto Gallassi (University of Bologna) asked how to improve the diagnosis of MCI.

Please here let me introduce the answer by Stephen Hurt (Cornell University):

"Your question seems to be primarily about standardizing the assessment / criteria for the diagnosis of MCI. There are two working diagnosis sets beyond the Petersen criteria. One was published in the last year by the National Institute of Aging / Alzheimer's Association and are usually referred to as the NIA-AA criteria. The second set was published by a European consortium. You might find these articles interesting.

AD New Criteria_Diagnositic Recommendations MCI due to Alz proof_Apr2011.pdf

Here's the 2nd ariticle.

Portet_J Neurol Neurosurg Psychiat 2010_MCI criteria by EACD.pdf "

Sorry, here the links are probably mute, but I may suggest to follow that blog, wherein I just wrote as follows: Stephen Hurt highlights 2 inputs. Both propose a step by step appoach, stemming from primary care to second level / research contexts. Both sets imply some difficulties while performing complex activities (errors, time, etc.). NIA-AA criteria underscore the value of 1 to 1,5 standard deviations below the mean for matched peers / normative as guidelines, not cutoff scores.

The link kindly offered by Stephen Hurt reminds to a 2006 JNNP paper: I wasn't able to find in the web Portet's paper dating 2010: may be it's my fault.

Pink snowed mountains at sundown while homeworking ...

I would consider looking at the Memory Orientation Screening Test (MOST). It is now available on the iPAD, in 2 forms, one of which meets CMS criteria for a computerized based neuropsychological test reimbursable under the CPT code of 96120.

It is more sensitive for the detection of MCI, and does a better job than any of the other instruments presently available in distinguishing normal from not normal, even in patients who do not meet the 2011 NIA/AA criteria for MCI.

I would be happy to forward the validation articles to you. In the interest of full disclosure, I am one of the developers of the MOST.

This test can be administered by a medical assistant or anyone with a 12th grade education. It takes five minutes even with severely demented patients. And, when disease severity changes, in either direction, improvement or decline, it is sensitive to the degree of change, as any good neuropsych test should be.

"The Memory Orientation Screening Test (MOST) is a test demonstrated to be sensitive to disease severity and correlates highly with more extensive and elaborate standardized neuropsychological test batteries." (Clionsky, Mitchell; Clionsky, Emilymarie. Development and Validation of the Memory Orientation Screening Test (MOST™): A Better Screening Test for Dementia. doi: 10.1177/1533317510386216 AM J ALZHEIMERS DIS OTHER DEMEN December 4, 2010 vol. 25 no. 8 650-656.)

It is actually now on the approved list of a major insurance company. It has been deployed worldwide on the iPAD.

Feel free to contact me for assistance. Good luck.

Paired-Associate Learning test is quite sensitive in detecting amnestic Mild Cognitive Impairment. It is available for iPads, as a bundle. It is marketed as CANTABmobile.

Blackwell AD, Sahakian BJ, Vesey R, Semple JM, Robbins TW, Hodges JR. (2004). Detecting dementia: novel neuropsychological markers of preclinical Alzheimer's disease. Dement Geriatr Cogn Disord. 17(1-2):42-8

Diagnosis of MCI is not ideally assessed in a brief neurobehavioral exam. For example, the MMSE, although widely used, has poor sensitivity and specificity for cognitive impairment in this population (P. B. Rosenberg, et al. 2006 A Clinical Approach to Mild Cognitive Impairment Am J Psychiatry 163: 1884-1890). Patients with high educational levels may show normal cognitive function scores on tests such as the MMSE. (G.W. Small, et. al. 1997 Diagnosis and treatment of Alzheimer disease and related disorders: consensus statement of the American Association for Geriatric Psychiatry, the Alzheimer's Association, and the American Geriatrics Society. JAMA, 278, no. 16: 1363). It is recommended that “Quantitative neuropsychological testing, ideally performed by someone trained in neuropsychology, (aka not someone with a 12th grade education) should be considered in patients with prodromal, mild, or moderate stage of dementia” (G. Waldemar et al. 2007 Recommendations for the diagnosis and management of Alzheimer's disease and other disorders associated with dementia: EFNS guideline. European Journal of Neurology 14, e1-e26).

The accurate diagnosis of MCI requires neuropsychological evaluation to objectively assess the degree of cognitive impairment and intraindividual change with evidence that that cognitive performance is impaired beyond what would have been expected for that individual relative to appropriate normative data. (M.S. Albert et al. 2011 Alzheimer’s & Dementia, 1–10; P. B. Rosenberg, et al. 2006). As other cognitive domains aside from memory can be impaired among individuals with MCI, it is important to utilize validated clinical neuropsychological measures to assess cognitive domains including executive functions, language, visuospatial skills, and attentional control. A typical battery of neuropsychological instruments assessing these cognitive domains is considered more sensitive than routine office tests or single domain assessment and provide a more thorough profile of deficits to allow the distinction between amnestic and nonamnestic, and single- and multiple-domain MCI (M.S. Albert et al. 2011; Jak et al. 2009 Quantification of five neuropsychological approaches to defining mild cognitive impairment. American Journal of Geriatric Psychiatry, 17: 368-375; N. Cherbuin, P. Sachdev, K. J. Anstey 2010 Neuropsychological Predictors of Transition From Healthy Cognitive Aging to Mild Cognitive Impairment: The PATH Through Life Study American Journal of Geriatric Psychiatry 18, 723-733).

This has significant implications for intervention planning as individuals with amnestic MCI are more likely to convert to Alzheimer’s Dementia. Neuropsychological testing has been shown to predict the likelihood of an AD diagnosis within 5 years and as more individuals are identified at earlier stages of cognitive impairment disease-modifying compounds may be found to be more valuable (B.C. Dickerson et al. 2007 Clinical Prediction of Alzheimer Disease Dementia Across the Spectrum of Mild Cognitive Impairment. Archives of General 64:12, 1443–1450). A report by the Alzheimer’s Association on the economic implications of the impending epidemic of AD, as the “baby boomer” generation ages, suggests that more than 13.5 million individuals just in the United States will manifest AD dementia by the year 2050. A hypothetical intervention that delayed the onset of AD dementia by 5 years would result in a 57% reduction in the number of patients with AD dementia, and reduce the projected Medicare costs of AD from $627 to $344 billion dollars.

Hi Dorina, this is a very interesting test for MCI: http://www.mocatest.org/

Dorina, we also have found that the Montreal Cognitive Assessment Test (MoCA) provides decent sensitivity and specificity for identifying changes in cognitive status that may signal MCI or be the prelude to dementia. The MoCA is domain oriented with more demanding criteria than the MMSE so ceiling effects are not problematic. Of course, the MoCA is intended to be a screening test, and scores below 25-26 should trigger a more comprehensive neuropsychological examination.

I would highly recommend the 18 point version of the Clock Drawing Test.

Dear All,

we also proficiently used MOCA [see Frank Webbe input] in

"Forloni G., Polito L., Davin A., Abbondanza S. ,Vaccaro R., Valle E., Guaita A., Colombo M., Vitali S., Ferretti V.V., Villani S. Cognitive stimulation and APOE genotype in non-demented elderly subjects: a randomized controller study (RCT). JHNA (The Journal of Nutrition, Health & Aging) 16 (9): 841-2, 2012"

Let me allege the conclusions from the abstract:

"These results provide further evidence that CS benefits cognition in elderly non-demented subjects. Moreover, few sessions of CS are potentially more beneficial for subjects without cognitive deficits rather than MCI participants. Follow-up evaluations are necessary to assess the long-term effect of this kind of intervention and potentially to highlight a benefit also for MCI subjects in terms of a slower cognitive decline over time. APOE genotype does not influence the outcome of CS compared to SE. Finally, our data show that APOE genotype impacts on the gain in MOCA score registered by administering the test after a short period of time."

Dear all, I created a intensive methodology to understand this period that may preceed dementia: subjective cognitive impairment and mild cognitive impairment. This methodology is based in 3 months of workshops 2 times a week where older people learn and improve in digital literacy. We use MOCA, Mini Mental, and CDR, we video tape and write down all they say . So we have a quantitative and qualitative view of their evolution. I really think that this is a way to have a deep understand of what is happening and the multidomain effect ofa CST.

dear André, an "ecological" intensive observation indeed: maybe technology could help in sparing time and intrusiveness [see papers by Philippe Robert, Nice (France) University]; good work, Mauro

This is not testing but I wonder if the kirtan chant - useful for MCI would be of HELP as potential treatment modality

See: http://www.anti-aging-articles.com/Kirtan-Kriya-Brain.html

Hi all. In agreement with most of the recommendations above, I would also suggest looking into collecting data from multiple domains. However, in addition to that, there is also some tentative evidence that failure to delay recall primacy words in the AVLT (i.e., early list words) might be an indicator of poor consolidation. See attached paper.

Another potentially interesting reference to chew on (also to do with failure to consolidate information):

Dewar, M., Pesallaccia, M., Cowan, N., Provinciali, L. & Della Sala, S. (2012). Insights into spared memory capacity in amnestic MCI and Alzheimer's Disease via minimal interference. Brain and Cognition, 78, 189-199.

Dorina Hi,

I think you have been given a fairly comprehensive outline of the problem of using screening instruments for diagnosing very early dementia. MCI is now regarded by most as a preclinical stae of dementia. Often the diagnosis is not sufficient on screening instruments such as MMSE and the addition of MoCa or SLUMS does give some guidance. Often more advanced neuropsychological testing is required.

On the whole the use of a combination of various instruments strengthens the clinicians ability to diagnose the problem Obviously special demetnia such as Fronto-temporal Dementia are more difficult to diagnose with screening instruments and need maybe more detailed examination in specialized clinics.

I suggest a very simple test: reverse calculation.

If the patient can reply 103-4=99, it is improbable that he (or she) is begining dementia.

It is especially useful when other signs suggest dementia.

If reverse calculation is good, it is probably a depressive, curable, pseudo dementia.

Hi everyone, I've saw that the MoCa test is the main screening test for who recommend better specificity and sensitive markers than those that one can expect with the MMS, however, the MoCa has a high ceiling effect for people with low cultural or education background, tending to mark as MCI the old healthy persons with poor educational level. So, which is really the best cognitive test to detect dementia on its initial state, taking in mind the variable of the education??

http://download.journals.elsevierhealth.com/pdfs/journals/0003-9993/PIIS0003999309002858.pdf

This links relate to an article where it shows a summary of the prons and cons of the most used screening test, hope it serves for you Dorina, bye.

I don't know which test can be the best, but maybe other things can be looked at apart from tests. Changes in speech, changes in writing, changes in mood...

We are doing some work in this direction:

1) "On the Selection of Non-Invasive Methods Based on Speech Analysis Oriented to Automatic Alzheimer Disease Diagnosis", Sensors. 05/2013; 5(13):6730-6745. DOI:10.3390/s130506730

2) "Biometric Applications Related to Human Beings: There Is Life beyond Security", Cognitive Computation 01/2013

3) "Diagnosis of Alzheimer's disease from EEG by means of synchrony measures in optimized frequency bands", Annual International Conference of the IEEE Engineering in Medicine and Biology Society. IEEE Engineering in Medicine and Biology Society. Conference 08/2012; 2012:4266-70.

Hope it helps!

The battery of tests mentioned by contributors is excellent and I will add interviews with close relatives and caregivers as yet another tool. Ms. Meulemans' comments supports this.

I'm working with colleagues in Sweden with data from two 20-year longitudinal studies of aging where we are using psychiatric examinations, interviews with relatives, and "cause of death" as rough measures of whether or not the panelists had Alzheimer's. N = 3,700. We are going to do whole genome genotyping for everyone in the studies in order to explore some epigenetic questions associated with the disease. Earlier we had found some evidence that social networks contributed to healthy aging, so now we want to take another look at SN plus other social and medical factors as they interact with ApoE-4 and whatever other genotypes might be involved with Alz.

We should be able to analyze the differences between 4 groups of panelists: 1. People with Alz. and the genotype(s); 2. People with the genotype(s) without Alz; 3. People with Alz. without the genotype(s); and 4. People with neither Alz. not the genotype.

If panelists lived for the 20-year study period, they should have 5 frozen blood samples on file (medical/social inventories were done every 4 years in most cases). We will prepare the blood for gene sequencing by noting dates of the samples, among other things. The preparation alone will take around 6 months to complete.

But, the question of how one diagnoses Alz. is a significant problem for our study.

Dear Dorina Cadar,

Blessed probably is the best because of the more detailed rating score. MMSE is used in virtually all evaluations but have a lot of limitations in sensitivity and specificity; it's inaccurate in many middle/low income countries worsening the ceiling or floor effects with confounding problems "between the celling and the floor". Both are not enough sensitive in many regions because of illiteracy or very poor schooling. So a complementary clinical tests or neuropsychological tests and retest are necessary to improve MCI diagnostic. The MOCA works good in some places; but the original test present animals as a figure that are not seen in many places, as Brazil for example; so cross-cultural barriers may compromise e evaluation even if a validation test is available. This is a major challenge around the World! I suggest you to examine the "Alzheimer’s Disease Neuroimaging Initiative": http://www.adni-info.org/

Our Institution is working on that; researching the differences with Normal cognition versus MCI (and it variations) versus "late-onset minor/subsyndormal/mood disorders". In the same way we are trying to identify pré-prodromal and predictors (neuropsychological and clinical evaluations plus biomarkers and advanced neuroimaging).

Best

Eduardo

Dear Dorina and colleagues, Please try or just take a look in the ACE-R.

In the clinical examination we are using The ACE-R, because we believe it's perform better (more discriminative) than MMSE or MOCA alone (or combined). You should to see this tool...

See you,

Eduardo

The paper test has any problems. Or is very simpler, or difficult. We made measurement, the hand tremor and cognitive time. They very earlier do very good answer, and we can see the change in the time of cure.

I would suggest to look at sensorimotor tests too. Recent studies demonstrated that people prevalent to cognitive alterations presented concomittant changes in sensorimotor markers.

Dear colleague,

I also agree that the MMSE is not a valid test for evaluate the early stage diagnosis of dementia. I've had experience with the ACE-R and I think it's much more reliable especially if the patient has no clear clinical signs. In the case in which the clinical symptoms discolse for a type rather than another I believe that nothing is more adequate of comprehensive assessment.

I've been intrigued by Jean-Jacques's hint: motion abilities could be a valuable key-point, not subsisting obviously an equation "bad locomotion = bad cognition" [and vice versa]. That fits also with my personal clinical experience [without claims of being a "narrative medicine" reputable author], Mauro

here follows an attachment just missed in my previous message, Mauro

I believe that David Simoni showed us a good answer. A comprehensive assessment is very important. One or other structured evaluation may be included in the clinical evaluation, but not used as a isolated examination.

Eduardo

There is no doubt that there are many cognitive tests currently used to diagnose dementia, but it lacks sensitivity for the detection of early onset dementia. I came across this new measure (http://www.cantabmobile.com/) which is currently being tested in Cambridge clinics for clients who come in with memory problems. It is administered by doctors and takes about 10 minutes. Scientific research from this group (which helped develop the CANTAB assessments) says that this test is sensitive to detecting early changes in cognition/memory and is attempting to promote the use of this product. Perhaps this is something we can look forward to in the future.

Quite a bit of recent literature out about gait speed as an early indicator of cognitive decline, supporting what a few other colleagues have mentioned in this discussion. Two papers attached here.

And the second.

I would strongly consider evaluating the MOST and the MOST 96120, the latter of which is reimbursable by Medicare in the U.S.

The MOST accurately classifies individuals as normal, or having MCI, or dementia with a high degree of sensitivity and specificity. The test is also stable over time ( test-retest r>= .90 over 9 months); has a high degree of inter-rater scoring reliability ; (r>=.9). The MOST takes 5 minutes to administer and does not require a skilled staff to do so. It is now available on the iPAD/iPAD mini. The validation of the test characteristics on the computer is in process of submission with the same excellent performance on the computer as well administered on paper.

Full disclosure: I am a co-developer of the test which is now being used across the U.S. and in over 33 other countries. And, yes, it assesses executive function which is something that the MMSE is not able to able to do.

Sructural imaging provides information about the position, shape, and volume of brain tissue. These techniques include MRI's and CT's. Structural imaging studies have shown that the brains of people with Alzheimer's shrink significantly as the disease progresses..

Functional imaging reveals how cells in certain brain regions are working by showing how actively the cells use oxygen and sugar

This core issue of detecting insidious changes in cognition prior to the onset of dementia takes on new meaning with the game changing concept that the the "disease" of dementia may be present for decades before the appearance of the clinical deficits. In years past, we were concerned only about the temporally immediate precursors to frank dementia. Now, we have to be concerned about subtle changes that may begin many years prior to traditional ages of expected onset. In that sense, Sandra's comments about the utility of imaging approaches (and other biomarkers) may give us the earliest hints of the presence of the disease. Reports from informants and the patients themselves of subtle changes in behavior may be equally revelatory. Which of our current cognitive tests may be most sensitive in this early detection has not been empirically demonstrated. Clearly, none of the screens such as the MMSE or MoCA would be players, since by the time reliable changes in those scores are noted, either MCI or early dementia has likely begun. Although some of the computerized tests such as CANTAB, as Joyce mentions, may turn out to be good metrics, we do not know whether they will be sufficiently sensitive to parse out the effects of normal aging on functions such as speed of mental processing or complex reaction time. Paul notes that changes in gait speed may be good early indicators, and the studies he cites are clearly good beginnings in laying out good empirical support.This whole area will be an exciting one for young researchers to enter. In identifying the best contenders for early identification, the stage will be set for measuring the impact of new interventions (pharmaceutical and others) for prevention.

i work in a liaison psychiatry service, the ACE-R is a comprehensive and well evaluated tool if the respondant is willing, we also use the MOCA which quite focal in pinpointing specific deficts.

Unless you have previous measures on the same person and you can control for confounding effects (or measuremente error), it will be very difficult to identify change from previous levels. All you can test is decline from previous SUSPECTED level. But, you can test several times and see if that person benefits from test-retest practice effects. I studied a woman diagnosed with non-amnestic mild cognitive impairment who performed within the average range on the Free and Cued Selective Reminding Test in three successive assessments. However, her preformance was significantly lower than what would be expected based on practice effects, which indicates that objective memory impairments were present. Thus, memory impairments were identified early in the course of the disease and the intervention was significantly useful for her.

I hope this helped

Best

a keen observation indeed, tackling clinic and epidemiology

Happy new year to everybody, Mauro

The Road Map Test, originally developed by Money 1976 for children, has been very sensitive to cerebrovascular and Alzheimer's dementias. This is a simple test requiring monotonic rotation of an object in mental space. In our study, it showed decline in patients characterized with early dementia. I can provide you with references, and normative data for adults should you wish to use it and need control values.

Dear Carol Amstrong, I would love to have those references and normative data. And, if possible, original material to study if there are differences between english- and non-english speakers. Thank you so much!

All the references are in the attached paper. Tell me about how you would like to use the normative data. The normative data is all in English speakers. You might need to pilot a small sample of English and native language speakers on the test, but there is little in the procedures and directions that would appear to be culturally sensitive, and no language needed to perform the test except to state "left" and "right".

Thank you for the paper. I will collect some data based on your paper and will contact you again so we can analyze them together (if you agree to). I will start soon a research with AD, so i will include it in the battery both with patients and healthy controls.

Thanks again, and happy new year!

Javier, my email: [email protected]

Predictors of cognitive decline in Alzheimer's disease and mild cognitive impairment using the CAMCOG: a five-year follow-up. International Psychogeriatrics, 2012; 24(6): 948-958

Deficits in orientation and memory learning are a good indicator. CAMCOG test

Article Predictors of cognitive decline in Alzheimer's disease and m...

I suggest that you at least check on www.memtrax.com

This test is free, easy, and fun. But the main point is that is can be taken as often as you like over a long period of time and identify subtle memory changes from baseline or a declining trajectory. There is no other test with close to the power of memtrax for early detection of neurocognitive changes and user-friendliness. This site is also under long-term development, based on the latest scientific studies of memory mechanisms, so it is worth considering the direction memtrax is taking for the future.

Thanks for the link, Mr. Ashford. Are there test-retest data for that task?

Yes, we have test-retest data, not yet published - we have data from over 20,000 web users. Second takes of the test are highly similar. This matches my experience from giving the test to a wide variety of patients in my office over the last 12 years.

We have worked very much with the Barrow Neurologigal Institute Screen (BNIS) by GP Prigatano et al. We have experienced this screening test so good that we have translated and validated it to Swedish circumstances. It is very godd I think to use as a "first" screening" and if you need yoyu can continue with a broader neuropsycological test battey. I have ofen felt this unnessery. The BNIS can also be used as a bed-side test.. It is easy to administer. There is a prescreen test to first assure that the patient is testable. The screen then includes 7 subscales where the affect subscale and the self-awareness subscale vs performance are unique. When we have used this test besides Minimental test (MMSE) on cardiac arrest patients where we found the ceiling effects of the MMSE.

The BNIS is translated to several languages.

I include a poster and an abstract presented at the 5th World Congress of Brain Injur 2003 to show the ceiling effect we found

You can contact dr GP Prigatano at the Barrow Neurological Institue in Phoenix, Arizona, USA directly if you wish .I wish you the very best!

.

Abstract title: Comparison between two screening instruments for cognitive functions in survivors of cardiac arrest considering ADL function.

BACKGROUND: Assessment of cognitive functions needs simple and convenient screening instruments easily administered to many patients also when physical or mental condition is reduced. A brief, reliable and sensitive screening test used as a detector for cognitive impairment in the early phase after brain damage can identify the need for further evaluation and rehabilitation efforts and might also predict outcome.

A cardiac arrest can cause brain injury due to reduced blood flow, causing impaired oxygen and nutrition delivery to the brain. Anoxic brain damage is known to produce neuropsychological impairment of different severity.

OBJECTIVES: To evaluate the predictive value of two screening instruments of cognitive function with respect to independence in ADL function in a group of survivors of cardiac arrest. This prospective study was performed in a community based setting.

METHODS: 50 consecutive survivors of cardiac arrest, during 1996-1999, aged 18 – 75 years were included. 28 patients were assessed (mortality caused missing data). Evaluations were made at 14, 45, 90 days and at 1 year after the arrest. The BNI Screen for Higher Cerebral Functions (BNIS) and The Mini Mental State Examination (MMSE) were used for assessing cognitive functions and the Functional Independence Measure (FIM) for assessment of ADL independence.

RESULTS: A clear ceiling effect was shown for the MMSE but not for the BNIS, where the results were more evenly distributed.

Significant correlations were found between the BNIS and the FIM as well as between the MMSE and the FIM at the different follow up occasions. Early measurement (90 days) with the BNIS and the MMSE showed significant correlation with late measurement with FIM (1 year). The MMSE and the BNIS were significantly intercorrelated.

The profile of BNIS subscales indicated that the cognitive areas most affected were attention/concentration, memory and awareness vs performance but also visuospatial and visual problem solving.

CONCLUSION: Both the BNIS and the MMSE seem to have predictive value vis-à-vis independence in ADL function. Compared with MMSE, the BNIS does not have a ceiling effect and it has a more homogenous distribution of values. A profile of BNIS subscales shows that memory is an area susceptible for impairment in patients with this diagnosis but that deficits in attention, self awareness and visuospatial problem solving also are common and should be recognised in rehabilitation programmes.

Referenser:

BNI:

Prigatano G P; Amin K, Rosenstein L : Manual for the BNI Screen for Higher Cerebral Functions, 1991

Prigatano G P: BNI Screen for Higher Cerebral Functions: Rationale and initial validation. BNI Quarterly 7: 2-9, 1991

Rosenstein L , Prigatano G P, Amin K: Reliability Studies for the BNI Screen for Higher Cerebral Functions, BNI Quarterly, 1992;8: 24-28

Prigatano G P, Amin K, Rosenstein L D: Validity Studies on the BNI Screen for Higher Cerebral Functions, BNI Quarterly, 1993;9:2-9

Prigatano G, Amin K, Rosenstein L: Administration and scoring manual for the BNI Screen for Higher Cerebral Functions. Phoenix, AZ: Barrow Neurological Institute, 1995

Denvall V, Elmståhl S, Prigatano G: Replication and construct validation of the Barrow Neurological Institute Screen of Higher Cerebral Function with a Swedish population.

Jr Rehabilitation Med 2002, 34; 153 – 157.

MMSE:

Folstein M F, Folstein S E, McHugh P R: “Mini-Mental State”. A Practical Method for Grading the Cognitive State of patients for the clinician.

Jr of Psychiatric Res 1975, Vol. 12, pp. 189 – 198.

Tombaugh T N,McIntyre N J: The Mini – Mental State Examination: A Comprehensive review. JAGS 40: 922-935,1992

Caisa Hofgren Eva Esbjörnsson Katharina Stibrant-Sunnerhagen

Clin. Psychologist Clin. Psychologist, Ph D. MD, Ph D.

Adress: Skriv som på visitkort!

Sorry my answer includes so many spelling errors. It was also impossible for me to include the poster I wanted to show you.

Good luck

Eva Esbjörnsson

Maybe, you may find useful the screening of executive functions with FAB.

Dear Dorina

A person has dementia when the capacities of the brain deteriorate enough and cease to be as before. Most important warning signs of dementia are forgetfulness ,problems with language, no longer knowing guide, bizarre behavior, difficulties in daily tasks, difficulties in planning and organizations, lack of energy. These warning signs does not automatically imply a form of dementia . However , a medical examination if these situations affect the day-to-day it is necessary

The MMSE is a pretty blunt instrument and not able to detect early changes reliably. There is some research on more complex behaviour such as language, which draws from a large neural network and can change gradually. For instance, check "Connected speech as a marker of disease progression in autopsy-proven Alzheimer's disease" by Ahmed et al. There is still much work to be done here, though.

One big problem is that different types of dementia will result in different patterns of decline, so it is questionable that looking at one aspect behaviour will suffice. Also, there is always large heterogeneity in healthy people, which makes it a challenge to determine what the normal range is and when a result falls outside it.

Mr. Steding, The only publication relating to MemTrax is at:

http://www.medafile.com/JWA-JAD-CRT-2011.pdf

We have data on over 20,000 individuals who took the test on-line. We have found little difference by sex, but education has an effect, as it did in the audience paper. Unpublished data shows the recognition time increasing with age. Then there is a very difficult question about whether to use age norms or absolute cut-offs. I find that normal individuals usually have a reaction time less than 1 second and performance levels better than 92% correct. Between 1.2 to 1.4 second recognition times and 80% to 88% are mildly impaired and slower than 1,4 second and less than 76% correct is definitely impaired - but the result has to be interpreted, so an individual with poor vision may be inaccurate and slow, but not because of cognitive problems. Once registered (very easy to do, a legitimate email address is not required), tests are never repeated and the same image will not reappear until the test is taken 80 times. More targeted tests are under development: www.memtrax.org

- Thank you for your interest.