Having regard to the global burden of dementia, comprehensive research on possible instruments to properly determine cognitive decline has received a great amount of interest.
Diagnostic criteria
The mild cognitive impairment (MCI), which can be considered as a prodrome of dementia, has no universal diagnostic criteria but the following considerations are generally accepted: memory complaint; memory impairment for age and education; preserved general cognitive functions; intact activities of daily living. In contrast, in dementia everyday functionality is no longer maintained and significant cognitive impairment occurs in one or more areas [1-5].
Epidemiology
Dementia is a disease that afflicts a large proportion of the elderly. The WHO estimates that it affects around 50 million people worldwide and that there are around 10 million new cases a year [6].
The process of developing dementia may remain clinically unnoticed for a long time: the onset of molecular pathology can precede the onset of symptoms by up to 20 years. To intervene effectively and promptly, we need to recognize dementia before it fully develops clinically [7].
In light of this, it is increasingly crucial to have appropriate screening instruments available that are sufficiently accurate and have convenient merits in the early identification of cognitive deficit. These should preferably be screening tests that have outstanding sensitivity and specificity in the identification of cognitive impairment at the early stage including MCI and mild dementia.
Assessment
The most commonly used procedures in Hungary for the identification of AD are the Mini-Mental State Exam (MMSE) and the Early Mental Test developed by the University of Szeged [8,9]. There is no international consensus on which is the best test and with which cut-off points. In recent years, the Montreal Cognitive Assessment (MoCA) [10] and the Short Test of Mental Status [11] have been used for this purpose, the Hungarian version of which is not yet available [12].
The MMSE has been shown to be particularly useful in identifying dementia with a short filling time (around 5-10 mins). In a systematic review, the authors found that the sensitivity and specificity of MMSE were 0.81 and 0.89 respectively [13]. Janka et al. translated and validated it into Hungarian, the cut-off point they established was 24 [14]. However, it is unreliable in the early detection and not suitable for distinguishing healthy people from those with MCI [15].
For this purpose, Alzheimer's Disease Assessment Scale-Cognitive Subscale (ADAS-Cog) [16], which has more reliable properties but is more time-consuming (about 35 to 45 minutes depending on the number and the extent of cognitive areas affected) is used [17]. Although the potential for improvement of the test is raised by a large literature review, it is acknowledged that the ADAS-Cog is a useful measurement instrument in pre-dementia syndromes [18]. In its Hungarian validated form, outstandingly high values were found in the separation of healthy people from AD patients [19]. Both the sensitivity and the specificity are quite high at the mathematically optimal decision point of 12.15, respectively 96.97% and 91.49%.
We have selected a short, self-administered test developed by the English author Jeremy Brown, which is validated in different countries worldwide, also in Hungary (TYM-HUN) [20,21]. This is the Test Your Memory (TYM), which examines a fairly wide range of cognitive functions (orientation, attention, executive functions, language, memory, and visuospatial skills) in a short period of time (5-10 mins) and provides reliable information to the clinician when properly evaluated [22]. The TYM-HUN test, which was previously translated into Hungarian and validated, was found to be reliably used to screen for dementia using a 35/36 cut-off, with which both the sensitivity and specificity of the test were found to be 94% [21].
The goal of this study was to compare the diagnostic utility of the TYM-HUN with the MMSE and the ADAS-Cog tests for MCI in a Hungarian population. We determined a 'cut-off' point of TYM-HUN where optimal sensitivity and specificity values were obtained to screen MCI.