Exercise: Screening for dementia

This website is for students following the M.Sc. in Evidence Based Practice at the University of York.

The following is the abstract of a paper:

Introduction. Screening tests for dementia are a fundamental tool in specialist consultation and primary care. The instruments currently used are time-consuming and the diagnostic performance they offer is rather poor The original version of the Buschke Memory Impairment Screen (MIS) is a quick simple test with high discriminatory power.

Aims. Our aim was to validate a Spanish version of the MIS in specialist consultation.

Patients and methods. We conducted a prospective study in 91 subjects aged over 60 who visited the outpatient department for cognitive evaluation. All the patients received a similar diagnostic evaluation and the MIS was administered as a blind test. This Spanish version of the MIS was produced by means of a cross-cultural adaptation.

Results. The area below the ROC curve for dementia on the MIS was 0.92, with a similar result for cognitive deterioration (CD). A cut-off score of 4 or less offered a sensitivity rate of 91.9% (CI 95%: 83.4-96.4%) and a specificity level of 81.0% (CI 95%; 70.3-88.6%) for the detection of CD. Likewise, a cut-off score of 3 or below presented a sensitivity rate of 96.1% (CI 95%: 85.7-99.3%) and a degree of specificity of 72.6% (CI 95%: 63.2-80.3%)for the detection of dementia. Interobserver and test-retest reliability (0.85 and 0.81, respectively) were adequate.

Conclusions. This version of the Buschke MIS offers a high level of discriminatory power both for dementia and for CD, as well as an adequate degree of inter and intraobserver reliability within the context of a specialist consultation. Because it is quick (less than 4 minutes) and simple to administer, MIS represents a reasonable alternative to other screening methods.

(Source: Perez-Martinez DA, Baztan JJ, Gonzalez-Becerra M, Socorro A. Evaluation of the diagnostic value of a Spanish adaptation of the Buschke Memory Impairment screen in the detection of dementia and cognitive impairment. Revista de Neurologia 2005; 40: 644-648.)

Questions about this abstract:

What is a ROC curve?
Check suggested answer.
What does the area below the ROC curve tell us? What can we conclude from area = 0.92?
Check suggested answer.
For the detection of dementia, a cut-off score of 4 or less had sensitivity = 91.9% and specificity = 81.0%, and a cut-off score of 3 or less had sensitivity = 96.1% and specificity = 72.6%. Why did lowering the cutoff increase sensitivity and reduce specificity?
Check suggested answer.

Back to Measurement in Health and Disease index.

To Martin Bland's home page.

This page maintained by Martin Bland.
Last updated: 1 March, 2007.