Although I recognize the flaws that have been established with respect to MMSE scores, many brain banks or centers without a standard battery rely on MMSE as the more commonly described test during neurologic examination of patient's cognition. If you are revieweing a grant or paper, do you find yourself biased to the proposal or results if cognitive impairment is reported using MMSE?
You can expect the MMSE to have a very high false negative rate. That is, if the score is low, the patient really is demented... but if not, the patient may or may not be demented... it's just too crude a screen. Any serious publication reporting on cognition should be utilizing a better measure, for sure.
MMSE scores are known to vary by age and education within the population. That said, these factors have to be taken into account when interpreting the MMSE score.
See also http://faculty.pepperdine.edu/shimels/Courses/Files/MMSE.pdf
The MMSE represents a standard staging and assessment tool in AD. However, there are important drawbacks limiting the utility of MMSE. First of all, several items are relatively easy. Thus, patients with mild AD are not sensitively evaluated due to ceiling effects. On the other hand, some floor effects limit the application to patients with severe AD. Second, MMSE neglects important domains of AD assessment, such as function and behavior. Finally, factors such as age, education or cultural background may impair both sensitivity and validity of the assessment of disease progression. At least, only the "Neuropsychological Test Battery" is able to emphasize the assessment of memory and executive function in patients with mild to moderate AD, combining nine different cognitive tests and provides an index of global cognitive function in subjects with mild AD.
If reviewing grants and/or papers concerning multi-faceted conditions, of which cognition is one, then the MMSE is a rapid assay for detecting cognitive impariment. Typically to suggest that a group or disease in question has relatively normal cognition. or to identify outliers within a group that might be excluded due to cognitive issues.
As a definitive answer and classification of cognitive impairment comparing this score to some severity of a specific disease that is expected to have cognitive impairment, then i would be biased against such a publication or grant. If ti is used as a red flag that leads to further testing i am not biased against it.
The MMSE was intended to be a bedside quick exam to detect difficulties that may be teased out later by fuller cognitive assessment. In this capacity it functions very well, but has been quickly perverted to become a quick aand fast mass assay for research. Again all well and good if used to detect difficulties that will be teased later by fuller cognitive assessment.
Another rapid, and easy to mass assay scale that is considered better by a growing number of researchers that has been validated for a number of populations is the MOCA or Montreal Cognitive Assessment.
Single page capturing a number of abilities that are more extensive than the MMSE.
http://www.mocatest.org/
When I see a new "AD" patient, I first use the Lawton ADL screen to be sure that the pt does not have MCI. I also ask about education, since someone with a 5th grade education could score 5 points lower on the basis of poor education, not poor cognition). The MMSE should not be used to make the "yes-no" dementia decision. Then I use the Neuropsychiatric Inventory to be sure that the pt is not depressed (untreated depressed pts can score 5 points lower than their usual baseline on the MMSE). Yes, I use the MMSE to assess severity of cognitive impairment in AD, but only after assessing function, behavior and education.
All the reasons that the participants above had exposed (poor sensitive for high education background, low puntage tendency for the most elder) make me think that de MMSE is a tool that may be soon will be on retire. The Cognitive Psychology, and with it, the Neuropsychology, have both avanced on a way where the matter its not to shoot on the dark whit poor sensitive tools, instead, they had advanced in develop tests that may reflect with better specificity marks, not just if a cognitive domain (suchs memory on the AD) is in deficit, but wich of the cognitive process explain that defcit (is on the encoding, storage, or evocation phase). An example of this kind of test is the MIS (memory impairments screen), that had showed better ROC curves on sensitivity and specifcity measures, compared with MMSE.
Review the current literature on the MOCA. Recent studies have found it to be more sensitive than the MMSE, e.g. http://www.ncbi.nlm.nih.gov/pubmed/19933974. The MOCA is useful for detecting mild impairment, which is often missed by the MMSE.
Very much depends on setting and study population.
More objective measures easily instituted in health surveys might turn out to be a better option. For example, changes in timed walk seem to be a quite sensitive predictor of cognitive decline (see http://www.alzheimers.org.uk/site/scripts/press_article.php?pressReleaseID=814 and www.mayoclinic.org/news2012-rst/6989.html + others).
For self-report, would suggest looking at the 10/66 or WHO SAGE approach to assessing this.
- Badges
- Science topic
- Similar topics
- Psychology