Biochemical evidence of hyperthyroidism may be associated with low bone mass, particularly in older postmenopausal women, but no prospective studies of thyroid function and subsequent fracture risk have been done.

To examine the association between low levels of thyroid-stimulating hormone (TSH) and fracture in older women.

Prospective cohort study with case-cohort sampling.

Four clinical centers in the United States.

686 women older than 65 years of age from a cohort of 9704 women recruited from population-based listings between 1986 and 1988.

Baseline assessment of calcaneal bone mass, spine radiography, and history of thyroid disease. Spine radiography was repeated after a mean follow-up of 3.7 years; nonspine fractures were centrally adjudicated. Thyroid-stimulating hormone was measured in sera obtained at baseline from 148 women with new hip fractures, 149 women with new vertebral fractures, and a subsample of 398 women randomly selected from the cohort.

After adjustment for age, history of previous hyperthyroidism, self-rated health, and use of estrogen and thyroid hormone, women with a low TSH level (≤ 0.1 mU/L) had a threefold increased risk for hip fracture (relative hazard, 3.6 [95% CI, 1.0 to 12.9]) and a fourfold increased risk for vertebral fracture (odds ratio, 4.5 [CI, 1.3 to 15.6]) compared with women who had normal TSH levels (0.5 to 5.5 mU/L). After adjustment for TSH level, a history of hyperthyroidism was associated with a twofold increase in hip fracture (relative hazard, 2.2 [CI, 1.0 to 4.4]), but use of thyroid hormone itself was not associated with increased risk for hip fracture (relative hazard, 0.5 [CI, 0.2 to 1.3]).

Women older than 65 years of age who have low serum TSH levels, indicating physiologic hyperthyroidism, are at increased risk for new hip and vertebral fractures. Use of thyroid hormone itself does not increase risk for fracture if TSH levels are normal.