LAUSR

AIMS Determine the prevalence of glucocorticoid use in U.S. adults with diabetes and whether prevalence is associated with reduced skeletal health, as measured by fracture history and bisphosphonate use. METHODS Participants were age≥20years from the cross-sectional National Health and Nutrition Examination Survey (1999-2010; N=15,661). Diabetes was determined by self-report, fasting plasma glucose ≥126mg/dL (≥6.99mmol/L), or A1c ≥6.5% (≥47.5mmol/mol) (n=4539). Prevalences of fractures and bisphosphonate use were determined by diabetes status and glucocorticoid use. Logistic regression was stratified by sex and assessed the effect of glucocorticoid use and diabetes associated with fractures and bisphosphonates. RESULTS The age-standardized prevalence of glucocorticoid use was higher among persons with diabetes (3.2% vs. 2.0% without diabetes, p=0.001). Among adults with diabetes, the prevalence of fractures was significantly higher among those taking glucocorticoids vs. those not (38.3% vs. 26.1%, p=0.048). The prevalences of fractures and bisphosphonate use were generally similar in those with and without diabetes when stratified by glucocorticoid use. In logistic regression analysis among men, the combination of diabetes and glucocorticoid use (compared to those with neither) was highly associated with bisphosphonate use, while adjusting for demographic factors. Among women, having diabetes and glucocorticoid use increased the odds of fractures, while adjusting for demographic factors and menopause. CONCLUSIONS The prevalence of fractures was greater for those with diabetes taking glucocorticoids versus those not taking glucocorticoids. This study provides a national framework for further research on elucidating these associations.