Bone density, bone turnover and fracture susceptibility were evaluated in 1,132 randomly recruited women, all 75 years old. Seventy-four of the women had diabetes, while 1,058 women did not. Areal bone mineral density (aBMD) of the hip and lumbar spine was investigated by dual energy X-ray absorptiometry (DXA), and bone mass of the calcaneus was measured by ultrasound. Urinary deoxypyridinoline/creatinine (U-DPD/Crea) and serum C-terminal cross-linked telopeptide of type 1 collagen (S-CTX) were assessed as markers of bone resorption. Serum bone-specific alkaline phosphatase (S-bone ALP) and serum osteocalcin (S-OC) were assessed as markers of bone formation. Also, serum 25(OH) vitamin D and serum parathyroid hormone (S-PTH) were assessed. Fracture susceptibility was evaluated retrospectively and prospectively for up to 6.5 years. In diabetic women, the aBMD of the femoral neck was 11% higher (p<0.001), and BMD of the lumbar spine was 8% higher (p=0.002) than in non-diabetic women. There was no difference in bone mass by ultrasound of the calcaneus. Women with diabetes had higher BMD of the femoral neck (p<0.001) and lumbar spine (p=0.03) also after correction for differences in body weight. In diabetic women, U-DPD/Crea, S-CTX, and S-OC were decreased when compared with non-diabetic women (p=0.001 or less). After correction for covariance of body weight and plasma creatinine, S-CTX (p<0.001) and S-OC (p<0.001) were still lower in the diabetic women. Diabetic patients had hypovitaminosis D (p=0.008), a difference explained by differences in time spent outdoors and body weight. S-PTH did not differ between the groups. Women with diabetes had no more lifetime fractures (52%) than women without diabetic disease (57%), (p=0.31). This study shows that elderly women with diabetes and without severe renal insufficiency have high bone mass and low bone turnover. The high bone mass and low bone turnover is not likely to have a strong influence on fracture susceptibility.