Department of Medicine and Therapeutics, Prince of Wales Hospital, The Chinese University of Hong Kong, Shatin, Hong Kong, China.
Jockey Club Centre for Osteoporosis Care and Control, The Chinese University of Hong Kong, Shatin, Hong Kong, China.
Osteoporos Int. 2018 Feb;29(2):355-363. doi: 10.1007/s00198-017-4277-8. Epub 2017 Oct 24.
Falls are a major concern in terms of fracture risk. Although awareness rising for the absence of falls in the FRAX algorithm, our study only identified the independent predictive role of previous recurrent falls and their better conjunction use with FRAX for major osteoporotic fracture prediction in older Chinese men.
Although the association of falls with fracture has been widely explored, the impact of previous falls is not included in the FRAX algorithm currently. Our aim was to examine the FRAX-independent associations between falls in the previous year and subsequent fracture risk, as well as the conjunctive use of falls and the FRAX score for major osteoporotic fracture (MOF) prediction in older Chinese people.
Four thousand community older men and women aged 65 years or older were followed up for 9.9 ± 2.7 and 8.8 ± 1.5 years, respectively. The associations between falls in the previous 1 year and MOF risk by follow-up years were evaluated using the Fine and Gray model. New prediction scores were calculated by incorporating the falls and FRAX scores using the Fine and Gray model, or developed by adjusting the FRAX scores by 30% increased risk for each fall in the previous year. The predictive powers for MOF risk between the new scores and FRAX scores were evaluated by the area under the curve (AUC) and category-based net reclassification improvement index (NRI).
During the follow-up period, 139 (7.0%) men and 236 (11.8%) women had at least one incident MOF. One previous fall significantly predicted the first year incident MOF in men [hazard ratio (HR) (95%CI), 3.47 (1.02, 11.80)]. Previous recurrent falls significantly predicted a 10-year incident MOF in men [HR (95%CI), 2.42 (1.30, 4.51)]. In men, the fall-adjusting FRAX scores showed significant improvement on total net reclassification of fracture (3-6%). No improved predictive accuracy shown in women.
Falls in the previous year are likely to provide some predictive power to FRAX for MOF risk assessment in older Chinese men, but not women.
就骨折风险而言，跌倒仍是一个主要关注点。尽管 FRAX 算法中人们对无跌倒的认识有所提高，但我们的研究仅确定了既往复发性跌倒的独立预测作用，以及其与 FRAX 更好地结合，有助于预测中国老年男性的主要骨质疏松性骨折。
尽管跌倒与骨折的相关性已被广泛探讨，但 FRAX 算法目前并未包含既往跌倒的影响。我们的目的是检验既往 1 年内跌倒与随后骨折风险之间的 FRAX 独立相关性，以及在年龄较大的中国人中，将跌倒与 FRAX 评分相结合对主要骨质疏松性骨折（MOF）的预测作用。
对 4000 名年龄在 65 岁或以上的社区男女居民进行了 9.9±2.7 年和 8.8±1.5 年的随访。采用 Fine 和 Gray 模型评估了既往 1 年内的跌倒与随访年内 MOF 风险之间的关联。采用 Fine 和 Gray 模型将跌倒和 FRAX 评分结合起来，或通过将 FRAX 评分调整为每年增加 30%的跌倒风险，计算出新的预测评分。通过曲线下面积（AUC）和基于类别净重新分类改善指数（NRI）评估新评分与 FRAX 评分对 MOF 风险的预测能力。
在随访期间，139 名男性（7.0%）和 236 名女性（11.8%）至少发生了一次 MOF。既往 1 次跌倒显著预测了男性第 1 年的 MOF 事件[风险比（HR）（95%CI），3.47（1.02，11.80）]。既往复发性跌倒显著预测了男性 10 年内的 MOF 事件[HR（95%CI），2.42（1.30，4.51）]。在男性中，调整了跌倒的 FRAX 评分在 3%至 6%的骨折总重新分类方面有显著改善。但在女性中未显示出更高的预测准确性。
既往 1 年内的跌倒可能为 FRAX 评估中国老年男性 MOF 风险提供一定的预测能力，但对女性则不然。
