(Wiley, 2022-02-06) Rodríguez Blázquez, Carmen; Ayala García, Alba; Forjaz, Maria João; García Herranz, Sara; Venero Núñez, César; Rodríguez Fernández, Raquel; Díaz Mardomingo, María del Carmen
Aim
The Memory Failures of Everyday (MFE) is a widely used instrument for assessing memory failure. The aim of the study was to analyze the MFE items using the Rasch model in a sample of cognitively older adults in Spain.
Methods
A cross-sectional validation study in a sample of 214 healthy people aged ≥60 years who used centers for older people in Madrid (Spain). The MFE for the assessment of memory complaints was used. The following properties of the Rasch model were assessed: data fit, reliability, unidimensionality, local dependence and lack of differential item functioning by gender, age and marital status.
Results
The MFE showed a good fit to the Rasch model (χ2(140) = 160.2; P = 0.116) and high reliability (person separation index = 0.808). The questionnaire was unidimensional (6.54% t-test; IC binomial = 0.036–0.095). The items showed lack of local dependence between them and differential item functioning. The MFE scores were transformed into linear interval scores with a median of 44.31 and an observed range of 17.9–89.2 (theoretical range: 0–100).
Conclusions
The MFE is a unidimensional, reliable instrument to assess memory complaints in cognitively healthy older adults in Spain, with usefulness in clinical research and practice. The construct validity of the MFE linear score could not be fully confirmed and this deserves further investigation.
