To investigate the influence of diabetes mellitus and the bone assessment index (a stiffness index measured by quantitative ultrasound) on mortality in 80-year-old Japanese subjects from the general population, we compared the following 4 groups: groups I (non-diabetic subjects with a high stiffness index, n=334), II (non-diabetic subjects with a low stiffness index, n=129), III (diabetic subjects with a high stiffness index, n=48), and IV (diabetic subjects with a low stiffness index, n=8). The subjects all lived in Fukuoka Prefecture and were followed-up for 12 years after the baseline examination in order to collect the date and cause of death. The cumulative survival rate during the 12-year follow-up period for all-cause mortality, pneumonia mortality, and cancer mortality in group IV was lower than that in group I. A multivariate Cox regression analysis revealed a high mortality risk in group IV compared with group I; the all-cause mortality, pneumonia mortality, and cardiovascular disease mortality during the 12-year follow-up in group IV was 3.39-fold, 6.14-fold, and 5.04-fold higher than that in group I, respectively. The presence of diabetes mellitus and a decreased stiffness index independently and additively increased the mortality risk, even in 80-year-old people.

We examined gender differences in the plasma glucose level during a 75-g oral glucose tolerance test (OGTT) in patients with an HbA1c of 6.0-6.4 % (n=375). Men had significantly higher plasma glucose levels than women at 4 time points during the OGTT (0, 30, 60 and 120 min after an oral glucose load [OGL]). Glucose tolerance was categorized according to the Japanese Society of Diabetes Mellitus criteria. Among patients with borderline type glucose tolerance, men had significantly higher plasma glucose levels at all time-points except 120 min post-OGL; women had significantly higher plasma glucose levels at 120 min post-OGL. The same trend, was observed in patients with normal type glucose tolerance; however, the result did not reach statistical significance. Among patients with diabetic type glucose tolerance, men had higher plasma glucose levels, than women at 120 min post-OGL-with a trend toward statistical significance (p=0.064). There were also no marked gender differences in the HbA1c values in any of the glucose tolerance conditions. The gender difference in the plasma glucose levels of patients with borderline type glucose tolerance disappeared in patients with diabetic type glucose tolerance.

The ASSET-K study was an observational study using a retrospective database. In the ASSET-K study, the target was to achieve HbA1c<7.0 % with sitagliptin. This study included 519 patients who had been taking sitagliptin for 60 months. The subjects were divided into groups that achieved HbA1c<7 % or HbA1c≥8 % at 60 months. The patients' clinical characteristics were then compared between the two groups. A logistic regression analysis was performed to identify factors contributing to achieving HbA1c<7 %. HbA1c decreased significantly from 7.7 %±0.9 % to 7.1 %±0.9 % at 60 months (p<0.001), but the weight was relatively unchanged. The HbA1c<7 % group had a significantly higher age at baseline, larger body weight reduction, lower HbA1c at baseline, shorter diabetic duration and fewer drugs with different mechanisms of action prescribed at 60 months than the HbA1c≥7 % group. In a logistic regression analysis, the baseline HbA1c, presence of SU and change in body weight were significantly related to achieving HbA1c<7 %. Weight management is important for maintaining good long-term disease control with sitagliptin.

In 230 diabetic patients, the 2-year score on the exercise self-efficacy scale for diabetes self-care (ESESD), which were measured during outpatient visits showed a negative relationship with the glycosylated hemoglobin (HbA1c) level measured 2 years after the initial visit (p=0.022). The change value obtained by subtracting the HbA1c level at the time of the initial visit from the level measured at 2 years showed a negative relationship with the change value obtained by subtracting the ESESD score obtained at the time of the initial visit from the 2-year score (p=0.049). In 93 patients with diabetes, both the self-efficacy scale for diabetes self-care (SESD) and ESESD scores obtained at a later date were significantly higher than the scores measured before educative hospitalization (p<0.001, p<0.001). The change value obtained by subtracting the HbA1c level at the time of hospitalization from the level measured one year later showed a negative correlation with the change value obtained by subtracting the ESESD score before hospitalization from the score after hospitalization (p=0.005). Increases in self-efficacy led to improvement of blood glucose control, and self-efficacy scores, as assessed by the SESD and ESESD. The information provided by these scores is useful for supporting self-care.