The Tampa Scale of Kinesiophobia (TSK) has been used for a decade and is a valuable tool in researching pain-related fear. A variety of different factor models exist, however, and there are inconsistencies as to which model to use. The purpose of the study was twofold: 1) to thoroughly review existing factor models and 2) to empirically evaluate the previously proposed factor models in a large sample with persistent musculoskeletal pain. Subjects included 578 of 711 (81%) consecutive patients (aged 18-65 years) with persistent musculoskeletal pain from three different orthopedic outpatient clinics. We reviewed all existing factor models and performed confirmatory factor analyses on the existing models. Our review identified 11 factor models of the TSK. The identified models were tested on a large Swedish sample. All models were rejected because of unacceptable goodness-of-fit statistics in that specific sample. This study supports the fact that TSK is a multidimensional construct. Rather than searching for new factor solutions, future research should be devoted to forming a consensus for the conceptual and operational definitions of the construct kinesiophobia and the application of the Tampa Scale for Kinesiophobia. Physiotherapists are encouraged to take part in building new theories.