Osteoarthritis (OA) is the most common form of arthritis in the United States and is the leading cause of physical disability in older individuals. Currently, there are few efficacious treatments for OA. While many studies have evaluated clinical and genetic risk factors for OA susceptibility, little is known about the clinical risk factors and genetic mechanisms underlying OA progression. This dissertation addresses this gap through a set of studies utilizing data from two well-characterized multi-center longitudinal cohorts, the Genetics of Generalized Osteoarthritis (GOGO) and the Osteoarthritis Initiative (OAI). In the first study, we used data from the OAI to identify clinical risk prediction models for OA progression and validated these models in GOGO, an independent cohort. We found that bulge sign positive knee joint effusion, BMI, gender, diabetes, prescription analgesics usage, and KOOS pain score could be used to identify individuals at increased risk for total knee joint replacement. In the second study, we identified evidence for a major gene influencing knee OA progression in the GOGO that we localized to chromosome 18q21-22. Subsequent fine-mapping revealed genetic associations to variants within and near CCBE1, FECH, and SERPINB10. In the third study, we evaluated whether OA at multiple large joints increases the risk of OA progression. We found that OA at more joints increased the risk of structural OA progression at the knee and hip, but not the spine. We further observed that serum hyaluronic acid (sHA), a marker reflecting synovial inflammation, to be associated with knee and hip OA progression and prevalent OA at the hands, knees, and hips. In summary, we show that while clinical predictors have limited utility in predicting OA progression, biomarkers that are more closely related to the metabolic changes underlying OA progression may be more useful. Furthermore, several findings from these studies support a role for inflammatory mechanisms in structural OA progression.