Bone mineral density as a dose-response predictor for osteoporosis: a propensity score analysis of longitudinal...

Osteoporosis is a global disease burden for aging society. The role of quantitative ultrasound (QUS) in the prediction for osteoporosis in a dose-response manner is hardly addressed.

We aimed to show the dose-response of QUS measurement in the prediction for osteoporosis by a community-based study.

A prospective cohort study.

Participants were recruited between 2000 and 2004. Demographic data and heel QUS measurement were collected at baseline. Diagnosis of osteoporosis was ascertained by the follow-up of this cohort over time. In order to reduce the imbalance of baseline characteristics in the observational study, we applied propensity score by using proportional odds regression analysis to match the quintiles of QUS T-score.

A total of 44 957 subjects composed of 17 678 men (39.3%) and 27 279 women (69.7%) were recruited. After adjustments for propensity score, an increase in one unit of QUB T-score led to 7% reduction in the risk for osteoporosis [adjusted odds ratio (OR) = 0.93, 95% confidence interval (CI): 0.89-0.96, P < 0.0001]. Higher quintile of QUS T-score yielded a lower risk of osteoporosis with a gradient relationship [OR: 0.82 (95%CI: 0.72-0.92); OR: 0.81 (95%CI: 0.71-0.91); OR: 0.77 (95%CI: 0.68-0.87) and OR: 0.76 (95%CI: 0.67-0.86)] from the second to highest quintile opposed to first quintile (P < 0.0001). The cumulative incidence of osteoporosis was higher in the lower quintile during follow-up (log-rank test, P < 0.001).

QUS is an independent predictor for osteoporosis in a dose-response manner using a large population-based cohort. Due to the lower cost and portability of QUS measurement, the pre-screening for osteoporosis by QUS can be considered in the area with limited resources can be a feasible and alternative method.