Reliability and validity of the American Orthopaedic Foot and Ankle Society clinical eating scales: A pilot study for the hallux and lesser toes

Background: The use of clinical outcomes instruments is essential for the effective interpretation of individual patient progress as well as the comparison of treatment groups. An outcomes instrument must be reliable and valid to obtain any meaningful data. The purpose of the present study was to examine the reliability and validity of the American Orthopaedic Foot and Ankle Society (AOFAS) clinical rating scale for the hallux metatarsophalangeal- interphalangeal and lesser toes metatarsoplialangeal-interphalangeal joints. Methods: Eleven patients (one man, 10 women) with an average age of 54 (range 40 to 72) years and with classic rheumatoid arthritis not currently treated for foot complaints were enrolled in the present study. The average duration of rheumatoid arthritis was 14 years. Each patient completed a set of two outcomes instruments and had & physical examination by a single clinician at the initial visit and returned at 1 week for completion of the same scales and examination. The outcomes scales used were the AOFAS clinical rating scale for the hallux, the AOFAS clinical rating scale for the lesser toes, and the previously validated Foot Function Index (FFI). Test-retest reliability was evaluated using intraclass correlation coefficients between week 1 and week 2 for the summary scores as well as for the sebscales of pain and activity. Consistency between the two instruments was evaluated with Pearson correlation coefficients. Results: The AOFAS clinical rating scale for the hallux and lesser toes is repeatable between 1-week trials (ICC 0.95; p < 0.05; ICC 0.80; p < 0.05, respectively). Moderately strong correlations were found between the mean values for the AOFAS hallux and FFI (r = -0.81;p < 0.05). Weaker correlations were seen between the mean vataes for the AOFAS lesser toes and FFI scales (r = -0.69;p < 0.05). Conclusions: The faallux sabscale for pain correlates strongly with the FFI sabscale for pain, suggesting high content validity (r = -0.94;p < 0.001). Ceiling effects were seen with the AOFAS lesser toe subscale for activity, limiting its usefulness in a general patient population. The AOFAS lesser toe sabscale for pain and the AOFAS hallux subscale for activity correlated weakly with the FFI values (r = -0.31;r = -0.37;p > 0.05, respectively). Conclusions: Although the AOFAS hallux and lesser toe scales were found to be reliable in a rheumatoid patient population, their validity remains in question. These findings must be confirmed with larger subject numbers, with the inclusion of symptomatic patients before recommended routine use of the hallux clinical rating and lesser toe clinical rating scales.