Send to

Choose Destination
See comment in PubMed Commons below
Int J Geriatr Psychiatry. 2008 Nov;23(11):1127-33. doi: 10.1002/gps.2038.

CAMcog as a screening tool for diagnosis of mild cognitive impairment and dementia in a Brazilian clinical sample of moderate to high education.

Author information

Psychogeriatric Clinic and Laboratory of Neuroscience LIM-27, Department and Institute of Psychiatry, Faculty of Medicine, University of São Paulo, Brazil.



The CAMCOG is a brief neuropsychological battery designed to assess global cognitive function and ascertain the impairments that are required for the diagnosis of dementia. To date, the cut-off scores for mild cognitive impairment (MCI) have not been determined. Given the need for an earlier diagnosis of mild dementia, new cut-off values are also necessary, taking into account cultural and educational effects.


One hundred and fifty-seven older adults (mean age: 69.6 +/- 7.4 years) with 8 or more years of formal education (mean years of schooling 14.2 +/- 3.8) attending a memory clinic at the Institute of Psychiatry University of Sao Paulo were included. Subjects were divided into three groups according to their cognitive status, established through clinical and neuropsychological assessment: normal controls, n = 62; MCI, n = 65; and mild or moderate dementia, n = 30. ROC curve analyses were performed for dementia vs controls, MCI vs controls and MCI vs dementia.


The cut-off values were: 92/93 for dementia vs controls (AUC = 0.99: sensitivity: 100%, specificity: 95%); 95/96 for MCI vs controls (AUC = 0.83, sensitivity: 64%, specificity: 88%), and 85/86 for MCI vs dementia (AUC = 0.91, sensitivity: 81%, specificity: 88%). The total CAMCOG score was more accurate than its subtests Mini-mental State Examination, Verbal Fluency Test and Clock Drawing Test when used separately.


The CAMCOG discriminated controls and MCI from demented patients, but was less accurate to discriminate MCI from controls. The best cut-off value to differentiate controls and demented was higher than suggested in the original publication, probably because only cases of mild to moderate dementia were included. This is important given the need for a diagnostic at earlier stages of Alzheimer's disease.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons

How to join PubMed Commons

    Supplemental Content

    Loading ...
    Support Center