This article is only available in the PDF format. Download the PDF to view the article, as well as its associated figures and tables.
—Drs Kannus and Sievänen question the statistical significance of the findings of gain of bone mass in the third decade of life in women because of the wide range of the 95% CIs. In addition, they asked for a parametric statistical analysis of the results because of their skepticism about the validity of median values and nonparametric tests. The mean slopes (percentage per decade ±SD) were 4.82%±21.17% and 2.96%±18.32% for forearm bone mineral content and bone mineral density, respectively, 5.54%± 15.70% and 6.06%±15.07% for spinal bone mineral content and bone mineral density, respectively, and 12.62%±14.17% for whole body bone mineral content.In fact, we performed parametric tests (t test) of the mean slopes and found them all to show significant increases in bone mass during the third decade. However, the tests of normality showed that the data were not normally distributed. While the t test is fairly robust,
Robert R. Recker, K. Michael Davies, Sharilyn M. Hinders, Robert P. Heaney, Mary Ruth Stegman, Donald B. Kimmel. Variability in Bone Mass Measurement-Reply. JAMA. 1993;269(11):1386. doi:10.1001/jama.1993.03500110054029