Send to

Choose Destination
See comment in PubMed Commons below
Foot Ankle Int. 2006 Jul;27(7):519-27.

Theoretical model and Rasch analysis to develop a revised Foot Function Index.

Author information

VA Hines Hospital HSR & D and Loyola University Stritch School of Medicine, 5th Ave and Roosevelt Rd, Hines, IL 60141, USA.



The Foot Function Index (FFI) is a widely used self-reported measure of health-related foot function. Several areas have been identified for potential improvement, and this study responds to such criticisms. The objectives of this study were to: (1) develop a theoretical model of foot functioning, (2) develop a revised FFI (FFI-R), and (3) field-test the FFI-R.


A literature review was conducted to develop the theoretical model. The FFI-R items were developed from the original 23 FFI items, and more items were added as a result of the literature review. A focus group discussion with clinicians and pilot interviews with patients resulted in a final draft of the FFI-R. This draft consisted of four subscales and comprised 68 items with a six-point response scale. The FFI-R was field tested on 92 patients in the podiatry clinic of a Veterans Administration Hospital in the Midwest. Psychometric analyses were conducted with modern item response theory (IRT) methods.


A theoretical model of foot functioning was developed. The FFI-R response scale was revised from six to five categories since confusion was found between categories 4 and 5. Rasch analyses indicated a person reliability of 0.96 and item reliability of 0.93. The subscale reliability of pain and stiffness, psychosocial, and disability were all >0.80; the exception was assistive devices (>0.50). Construct validity of FFI-R was supported based on the correlation of 50-ft walk time with an FFI-R total of 0.306, p = 0.018, N = 59. Rasch analyses indicated several items with poor fit statistics and a short form with 34 items was developed.


The FFI was revised, and new items were added to compose the FFI-R. The chief theoretical change was adding a psychosocial scale. Both long and short forms had very good psychometric properties.

[Indexed for MEDLINE]
PubMed Commons home

PubMed Commons


    Supplemental Content

    Full text links

    Icon for Atypon
    Loading ...
    Support Center