Construct and diagnostic validities of the Free and Cued Selective Reminding Test in the Alzheimer’s disease spectrum

ABSTRACT

Introduction: The Free and Cued Selective Reminding Test (FCSRT) is a memory test that controls attention and acquisition, by providing category cues in the learning process. Because it enables an assessment of memory not confounded by normal age-related changes in cognition and a high accuracy on Alzheimer’s disease (AD) evaluation, it has been suggested by the International Working Group on AD. Our aim was to assess the construct related validity of the FCSRT in the AD spectrum disorders. Method: Patients were classified in two groups according to standard criteria: amnestic mild cognitive impairment (n = 100) and AD (n = 70). A matched control group (n = 101) of cognitively healthy subjects was included. The factorial structure of two models and respective construct and diagnostic validities were analyzed. Results: Both models revealed adequate fit values. The appropriated convergent validity and the lack of discriminant validity support the two factors as measuring the same construct (memory ability). The recalls of the FCSRT enabled high classification accuracy and diagnostic validity for both pathological groups. Conclusions: This study represents a novel contribution regarding the adequacy of the FCSRT in terms of construct and diagnostic validities and shows the interest of including both immediate (learning) and delayed (retention) recalls. It gives also new possibilities regarding the use of the FCSRT in the memory assessment of AD spectrum disorders.

KEYWORDS:

• Factor analysis
• Validity
• Alzheimer’s disease
• Cognitive impairment
• Memory

Acknowledgements

The authors would like to acknowledge the International Neuropsychological Society’s Research Editing and Consulting Program, in the persons of Alexandra Pentaraki and Michael A. Sugarman, for assistance in coordinating English language editing and statistical consulting, respectively.

Disclosure statement

All authors of this study declare that there are no conflicts of interest.

Additional information

Funding

This work was supported by the Foundation for Science and Technology (FCT)–Portugal [grant number SFRH/BD/74070/2010], [grant number PIC/IC/83206/2007].