OBJECTIVE: To determine the extent to which health items identified from the perspective of patients with knee osteoarthritis can be linked with the ICF; and to critically evaluate the ICF Comprehensive and Brief Core Sets for osteoarthritis. METHODS: Items identified from a focus group study were linked independently by two researchers based on the 10 a priori linking rules. Both percentage agreement and kappa statistics were calculated to measure inter-observer agreement. Any disagreements were resolved by reaching a consensus among the researchers. The categories linked with all items were compared with the Comprehensive Core Set, while the categories linked with those items reported as important by over 30% of subjects within each of 3 local ethnic groups (Chinese, Malay, and Indian) were compared with the Brief Core Set. Both comparisons were made only at the second level of the ICF. RESULTS: Totally 74 items were linked with 44 different ICF categories through 105 linkages with generally very good inter-observer agreement. The 69 items were linked with the ICF at the third or fourth levels. Both commonalities and disparities were found through comparison between the categories linked with these items and both Core Sets for osteoarthritis. CONCLUSIONS: All items could be successfully linked with the ICF. The Comprehensive Core Set demonstrated general conceptual validity, while the Brief Core Set needs to be supported by more empirical evidence in various socio-cultural contexts. This study specifically complemented the development and refinement of both Core Sets from the perspective of patients with knee osteoarthritis.