Effects of exercise on bone density and physical performance in postmenopausal women: A systematic review and meta-analysis.

The exercise programmes were all different in type, dose and duration, so the review cannot say which exercise works best or how much is needed. Only 4 of the 19 trials described how they hid which group a woman was assigned to, and about 80% could not blind participants or staff, so some of the improvements could reflect expectation rather than exercise. Study quality was middling: PEDro scores ranged from 4 to 8 with a median of 5 out of 10, and most trials did not analyse everyone as originally assigned. No included trial reported fractures and only one reported the number of falls, so this review cannot say whether adding exercise prevents fractures or falls. Every outcome was graded low or very low certainty, mostly because of risk of bias, inconsistency between studies, imprecise results, and signs of publication bias. Only a limited range of medicines and supplements was used, so the review could not examine how exercise interacts with different drugs.

The authors declare no conflict of interest. The work was funded by the National Science and Technology Council.

Do not read this as proof that exercise protects bone. Bone density was unchanged at every site measured, and no included trial reported fractures. The balance and strength improvements each rest on only three or four small trials, and the review team graded every single outcome as low or very low certainty, so treat those results as promising rather than established.