Background:Fragility fractures (FF) can occur at various sites of the skeleton, and are associated with multiple risk factors [1]. The prevalence of FF markedly increases with age. As the longevity of the population increases, so will the incidence of FF, and that of associated co-morbidities and risk factors. There are few data on co-morbidities associated with fractures at each site.Objectives:Identify associations of co-morbidities with sites of FF, by applying cluster analysis.Methods:We reviewed 28868 patients presenting for BMD estimation at a district general hospital in North West England, 2004-2016. We identified patients who had sustained one or more FF at time of presentation. Site(s) of FF were recorded for each patient, including femur, forearm, humerus, pelvis, ribs, spine, tibia or fibula. The following co-morbidities or treatments were recorded: excess alcohol consumption (previous or current); bisphosphonates; coeliac disease; family history of FF; hormone replacement therapy; hyperparathyroidism; hyperthyroidism; inflammatory bowel disease; polymyalgia rheumatica; rheumatoid arthritis; smoking (previous or current); corticosteroids (previous or current). Cluster analysis was performed on fracture sites and co-morbidities, using Jaccard similarity coefficient, and plotted on a dendrogram. Results were divided into an optimal number of clusters, derived using the elbow and silhouette methods.Results:11003 of 28868 patients had sustained one or more FF at time of BMD estimation. Overall, 84.6% patients were female, mean age 67.5years, and median T-score -1.12 SD. Cluster analysis was performed for FF sites and co-morbidities, with Jaccard similarity coefficients calculated. 4 clusters were identified (Figure 1): FF of forearm (n=5054), tibia/fibula (n=2617), spine (n=2352), associated with family history of FF, smoking, corticosteroids, and bisphosphonate treatment; FF of pelvis (n=300) associated with hyperparathyroidism, PMR, coeliac disease, and HRT; FF of femur (n=1181) and humerus (n=1131) associated with IBD and RA; FF of ribs (n=1022) associated with alcohol and hyperthyroidism.Conclusion:Cluster analysis demonstrated 4 distinct subgroups of FF sites and associated co-morbidities. To our knowledge, this is the first study applying cluster analysis to evaluate co-morbidities associated with FF sites. Risk factors may influence trabecular more than cortical bone, accounting for the difference in clusters. Knowledge of risk factors associated with FF site subgroups will aid prophylaxis and management in at-risk patients.References:[1]Kanis JA. Diagnosis of osteoporosis and assessment of fracture risk. Lancet (London, England). 2002 Jun 1;359(9321):1929–36Disclosure of Interests:Mrinalini Dey: None declared, Marwan Bukhari Speakers bureau: Bristol-Myers Squib, UCB celltech, Roche/Chugai, Pfizer, Abbvie, Merck, Mennarini, Sanofi-aventis, Eli-Lilly, Janssen, Amgen and Novartis.