The avoidance model in patients with knee and hip osteoarthritis (OA) hypothesizes that pain and psychological distress lead to avoidance of activities, and thereby to muscle weakness and activity limitations. This paper systematically reviews the scientific evidence for the validity of this avoidance model. A qualitative data synthesis was used to identify levels of evidence. Sixty studies were included. In knee OA, strong evidence was found that avoidance of activities is associated with activity limitations via muscle weakness (mediation by muscle weakness), strong evidence was found for an association between muscle weakness and activity limitations, and weak evidence was found that pain and psychological distress are associated with muscle weakness via avoidance of activities (mediation by avoidance). In hip OA, weak evidence was found for mediation by muscle weakness; and strong evidence was found for an association between muscle weakness and activity limitations. More research is needed on the consecutive associations between pain or psychological distress, avoidance of activities and muscle weakness, and to confirm causal relationships.