Serum 25OH vitamin D level, femur length, and risk of type 2 diabetes among adults.

The objective of this study was to examine the relationship between serum 25-hydroxyvitamin-D (25(OH)D) and femur length (FL) and their risk in association with diabetes in adults. A total of 3983 adults aged 20+ years from the National Health and Nutrition Examination Survey (2001-2002, 2003-2004) who had no missing measurements of FL or 25(OH)D were used in this analysis. Diabetes was defined as either being diagnosed with diabetes or an overnight fasting plasma glucose of (FPG) > 125 mg·dL(-1). Prediabetes was defined as having an FPG of 100 mg·dL(-1) < FPG < 126 mg·dL(-1). In addition, each individual was grouped into a sex-specified quintile of FL (Q1-Q5). Overall, approximately 11% of males and 9% of females had diabetes and 33% of males and 20% of females had prediabetes. Although the Pearson correlation between 25(OH)D and FL was weak, since FL changed from Q1 to Q5, the adjusted means of 25(OH)D in both sexes increased, while the adjusted means of FPG, insulin, glycohemoglobin, and the adjusted prevalence of diabetes were all decreased (p for trends <0.05). Logistic regression results indicate that the levels of 25(OH)D and FL were independently and negatively associated with the prevalence of diabetes, unaffected by other known risk factors. Among those without diabetes we observed the same trends. An increase in the means of 25(OH)D and a decreased prevalence of prediabetes were observed with FL changing from Q1 to Q5. Both 25(OH)D and FL are independently associated with diabetes risk in adults. However, the novel hypothesis of this association needs further research.