Objectives: To test the hypothesis that unintentional weight loss increases the rate of bone loss and risk of hip fracture more than intentional weight loss.
Design: Prospective cohort study.
Setting: Four communities within the United States.
Participants: Six thousand seven hundred eighty-five elderly white women with measurement of weight change and assessment of intention to lose weight.
Measurements: Weight change between baseline and fourth examinations (average 5.7 years between examinations) and assessment of intention to lose weight. Weight loss was defined as a decrease of 5% or more from baseline weight, stable weight was defined as less than a 5% change from baseline weight, and weight gain was defined as an increase of 5% or more from baseline weight. Rate of change in bone mineral density at the hip between fourth and sixth examinations (average 4.4 years between examinations) was measured using dual-energy x-ray absorptiometry. Incident hip fractures occurring after the fourth examination until June 1, 2001 (average follow-up 6.6 years) was confirmed using radiographic reports.
Results: The adjusted average rate of decline in total hipbone density steadily increased from −0.52% per year in women with weight gain to −0.68% per year in women with stable weight to −0.92% per year in women with weight loss (P-value for trend <.001). Higher rates of hip-bone loss were observed in women with weight loss irrespective of body mass index (BMI) or intention to lose weight. During follow-up of an average 6.6 years after the fourth examination, 400 (6%) of the cohort suffered a first hip fracture. Women with weight loss had 1.8 times the risk (95% confidence interval (CI)=1.43–2.24) of subsequent hip fracture as those with stable or increasing weight. The association between weight loss and increased risk of hip fracture was consistent across categories of BMI and intention to lose weight. Even voluntary weight loss in overweight women with a BMI of 25.9 kg/m2 (median) or greater increased the risk of hip fracture (multivariate hazard ratio=2.48, 95% CI=1.33–4.62).
Conclusion: Older women who experience weight loss in later years have increased rates of hip-bone loss and a two-fold greater risk of subsequent hip fracture, irrespective of current weight or intention to lose weight. These findings indicate that even voluntary weight loss in overweight elderly women increases hip fracture risk.