|
|
 |
 |
| |
 Printable version |
TOP 12: how to use it to identify pathological aging? |
Revue de neuropsychologie. Volume 3, Number 4, 273-83, Décembre 2011, Article méthodologique
|
 Résumé
Texte intégral
|
Author(s) : Émilie Lacot, Emmanuel J. Barbeau, Catherine Thomas-Antérion, Sandrine Basaglia-Pappas, Jérémie Pariente, Michèle Puel, Stéphane Vautier |
Summary : <\;p>The TOP 12 is a test designed to assess semantic memory quickly and easily. It relies on eight questions about the life of 12 celebrities that are named (verbal presentation).<\;/p><\;p>To verify the correlation between the mean scores of different groups of subjects and severity of the pathology\; to determine a threshold so that the status of each subject (ill or not) can be predicted based on his performance.<\;/p><\;p>Hundred and forty-five subjects were recruited (91 control subjects, 32 patients with Alzheimer's disease, 21 with amnestic Mild Cognitive Impairment [aMCI] and one with semantic dementia). The diagnostic properties of the TOP 12 were assessed through the confrontation of two methods: centile and Receiver Operator Characteristic (ROC) curve.<\;/p><\;p>The order of the means and severity of the pathology were correlated. The threshold that optimizes sensibility (Se) and specificity (Sp) is provided by the ROC methodology (83 points, Se \= 0.83, Sp \= 0.70). The fifth percentile is not optimal since it increases omissions.<\;/p><\;p>This external validation of the TOP 12 demonstrates the interest of using the ROC methodology.<\;/p> |
Keywords : collective memory, TOP 12, diagnostic properties, centile method, ROC curve |
|
