Related Experiment Video
Updated: Apr 19, 2026
A Machine Learning Approach to Design an Efficient Selective Screening of Mild Cognitive Impairment
Published on: January 11, 2020
[Validation of a brief screening test to assess functional capacity in Chilean older people]
Background:
The Pfeffer Functional Activities Questionnaire (PFAQ) is a validated instrument used in Chile to diagnose dependency in older people.
Aim:
To suggest a shorter and easier version of the PFAQ, with appropriate levels of sensitivity and specificity, which may be applied to the general older population.
Material And Methods:
Three Chilean surveys applied to older people were used, namely the Survey of Life Events of the Elderly, (n = 1,371) and the 2010 and 2013 National Surveys of Quality of Life of Older People (n = 1,954 and 2,644 respectively). To evaluate the instrument, we performed correlation analysis between items and principal components analysis. We calculated internal consistency using Cronbach alpha, we computed sensitivity and specificity indicators, and analyzed convergent and predictive validity.
Results:
After constructing different scales, the one that gave more information and had the best balance between sensitivity and specificity was chosen. Items with less weight factor and greater gender bias were eliminated, keeping seven questions from the original PFAQ. Reliability was verified using Cronbach alpha's coefficient and from correlations between the new instrument and key variables.
Conclusions:
According to our goals, the proposed scale is solely conformed by items that do not have gender bias. It discriminates in the general population adequately, allowing the design of a shorter test, of easier implementation.
