The impact of the CERAD-NP on diagnosis of cognitive deficiencies in late onset depression and Alzheimer's disease.

Objective: To identify the most appropriate test combination for distinguishing between late onset depression (LOD) and Alzheimer's disease (AD). To achieve this objective, the Consortium to Establish a Register for Alzheimer's Disease-Neuropsychological Battery (CERAD-NP) data of patients diagnosed with these two conditions were analyzed using multiple regression analysis.

Methods: In the first regression analysis, the following CERAD-NP subtests were included: verbal fluency, Boston naming test, word list learning, constructional praxis, word list recall, and constructional praxis recall. In a second regression analysis, only CERAD-NP memory parameters were included: word list learning, word list recall, word list intrusions, word list savings, word list recognition, word list false positive errors, constructional praxis recall, and constructional praxis savings.

Results: The combination of word list recall and constructional praxis recall best distinguished between LOD and AD, with a ROC of 0.91. In the stepwise regression of memory measures, word list recall, word list savings, and constructional praxis recall was the best combination, resulting in a ROC of 0.92.

Conclusions: The most efficacious combination of the CERAD-NP battery for discriminating between LOD and AD consisted of word list recall and constructional praxis recall. Of the CERAD-NP memory measures, word list recall, word list savings, and constructional praxis recall represented the best diagnostic combination.