Abstract
OBJECTIVE: To assess the clinimetric value of the Dutch version of the Foot Function Index (FFI) in comparison with the original FFI using verbal rating scales (FFI-5pt) rather than visual analog scales (VAS). METHODS: A prospective study was performed on 206 patients with nontraumatic forefoot complaints. Scoring, internal consistency, and construct validity of the FFI-5pt were compared with those of the original FFI, which rates all items on VAS. We also studied agreement between the scores at baseline and after one and 8 weeks and the scale scores with regard to sex, age, presence of osteoarthritis, limitation of mobility, bodily pain, and poor physical functioning (using SF-36). RESULTS: Some items were not applicable; removal of these items left 2 scales (Pain and Disability) with high internal consistency (alpha = 0.88 to 0.94) and good agreement between both versions (intraclass correlation coefficient 0.64 to 0.79). Principal component analysis with varimax rotation using a forced 2 factor model fitted well (65% explained variance). Test-retest reliability was high (ICC 0.70 to 0.83), while the stability over 8 weeks was lower, but still good (ICC 0.63 to 0.71). Responsiveness to change was low to moderate. However, a small number of patients reported an overall change (19%). Scores of patients with limited mobility and poor physical health (SF-36) were higher than those of patients with fewer physical problems, indicating good concurrent validity. CONCLUSION: The FFI-5pt is a suitable generic measure. Its clinimetric properties are comparable with those of the original FFI. Its administration and data entry are less time consuming. However, responsiveness has to be more exactly assessed in an intervention study.