Validation of a food frequency questionnaire for women with osteoporosis.

The importance of nutrient intakes in osteoporosis prevention in treatment is widely recognized. The objective of the present study was to develop and validate a FFQ for women with osteoporosis. The questionnaire was composed of 60 items, separated into 10 groups. The relative validation was accomplished through comparison of the 3-Day Food Record (3DR) with the FFQ. The 3DR was applied to 30 elderly women with confirmed osteoporosis, and after 45 days the FFQ was administrated. Statistical analysis comprised the Kolmogorov-Smirnov, Student T test and Pearson correlation coefficient. The agreement between two methods was evaluated by the frequency of similar classification into quartiles, and by the Bland-Altman method. No significant differences between methods were observed for the mean evaluated nutrients, except for carbohydrate and magnesium. Pearson correlation coefficients were positive and statistically significant for all nutrients. The overall proportion of subjects classified in the same quartile by the two methods was on average 50.01% and in the opposite quartile 0.47%. For calcium intake, only 3% of subjects were classified in opposite extreme quartiles by the two methods. The Bland-Altman analysis demonstrated that the differences obtained by the two methods in each subject were well distributed around the mean of the difference, and the disagreement increases as the mean intake increases. These results indicates that the FFQ for elderly women with osteoporosis presented here is highly acceptable and is an accurate method that can be used in large-scale or clinical studies for evaluation of nutrient intakes in a similar population.

Effects of simple hysterectomy on bone loss.

Lumbar spine and proximal femoral bone densities of Caucasian women, aged 35-45, were measured by dual photon densitometry model DPX. The measurement sites were assessed at the lumbar spine (vertebrae L2 to L4) and at the proximal femur (trochanter, femoral neck and Ward's triangle). After exclusion of women with climacteric symptoms, sterilized patients or those with menopausal concentrations of gonadotrophins, the study included 22 subjects: 11 menstruant (control group) and 11 hysterectomized. The hysterectomies were without oophorectomy and had been performed during the previous five years. The bone densities of the hysterectomized women were lower than those of the normal ones, but significantly lower at the Ward's triangle.