1 Scientific RepoRts | 7: 3781 | DOI:10.1038/s41598-017-03748-z www.nature.com/scientificreports Insulin use and excess Fracture Risk in Patients with Type 2 Diabetes: A propensity-Matched cohort analysis eladio Losada-Grande 1,2 , Samuel Hawley 3 , Berta Soldevila 4,5 , Daniel Martinez-Laguna 6,8 , Xavier Nogues 7,8 , Adolfo Diez-Perez 7,8 , Manel Puig-Domingo 1,4,5 , Dídac Mauricio 4,5 & Daniel Prieto-Alhambra 3,6,8 Despite normal to high bone mineral density, patients with type 2 diabetes (T2DM) have an increased fracture risk. T2DM medications could partially account for this excess risk. The aim of this study was to assess the association between insulin use and bone fracture risk in T2DM patients. A population-based matched cohort study based on a primary care records database validated for research use (Catalonia, Spain) was performed. Propensity score (PS) for insulin use was calculated using logistic regression including predefned predictors of fractures. A total of 2,979 insulin users and 14,895 non-users were observed for a median of 1.42 and 4.58 years respectively. Major fracture rates were 11.2/1,000 person-years for insulin users, compared with 8.3/1,000 among non-users. Matched models confrmed a signifcant association, with an adjusted subhazard ratio (adj SHR) of 1.38 [95% CI 1.06 to 1.80] for major fractures. No diferences between types of insulin or diferent regimens were found. Estimated number needed to harm (fracture) was 82 (95% CI 32 to 416). Insulin use appears to be associated with a 38% excess fracture risk among T2DM patients in the early stages of the disease. Fracture risk should be included among the considerations to initiate insulin treatment. Patients with type 2 diabetes mellitus (T2DM) have an increased risk of bone fractures 1–13 despite a normal to high bone mineral density (BMD) compared with non-diabetic subjects 10, 14–17 . Although the mechanisms underlying this observed excess fracture risk among T2DM patients remain unclear, some evidence indicates that fragility fractures in T2DM patients may be explained by the presence of impaired structural properties that compromise bone quality and ultimately lead to bone fragility 18, 19 . Risk factors associated with fracture risk in T2DM include duration of disease 7, 9, 14 , diabetic complications [impaired vision 1, 14 , peripheral neuropathy 20 , orthostatic hypotension 21 , etc.], episodes of hypoglycaemia 22 , increased risk of falling 20, 22–24 , inadequate glycaemic control 25 , and some antidiabetic medications that appear to afect bone metabolism, such as glitazones 26, 27 . Conversely, data on the impact of insulin treatment on fractures in T2DM are scarce and remain controversial, with some 4, 6–9, 13 but not all 1, 28 studies showing an increased risk of fracture. Most of the previous cohorts had no information on date of diagnosis or of insulin therapy initiation, making it difcult for researchers to accurately estimate attributable excess risk 1, 4, 6–9, 13 . Te increased risk observed in most observational studies has also been attributed to disease severity as most of the previously analysed cohorts lacked information on glycated haemoglobin (HbA1c), a key parameter to 1 Department of Medicine, Autonomous University of Barcelona, Barcelona, Spain. 2 endocrinology Section, internal Medicine Department, Hospital can Misses, ibiza, Spain. 3 Musculoskeletal Pharmaco- and Device epidemiology, centre for Statistics in Medicine, nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences, University of Oxford, Oxford, UK. 4 Department of endocrinology and nutrition, University Hospital & Health Sciences Research institute “Germans trias i Pujol”, Badalona, Spain. 5 ciBeR of Diabetes and Associated Metabolic Diseases (ciBeRDeM), instituto de Salud carlos iii, Majadahonda, Spain. 6 GReMPAL Research Group, iDiAP Jordi Gol Primary care Research institute, Autonomous University of Barcelona, Barcelona, Spain. 7 internal Medicine Department, iMiM (Hospital del Mar Research institute), Autonomous University of Barcelona, Barcelona, Spain. 8 ciBeR of Healthy Ageing and frailty Research ( ciBeRfes), instituto de Salud carlos iii, Majadahonda, Spain. correspondence and requests for materials should be addressed to D.M. (email: didacmauricio@gmail.com) Received: 2 September 2016 Accepted: 5 May 2017 Published: xx xx xxxx opeN