Background Physical activity (PA) may be an important fall prevention strategy. Current PA guidelines emphasize total PA dose, but daily patterning of PA is underappreciated. With aging, PA bouts become less frequent and shorter in duration (i.e.: more fragmented). PA fragmentation may be an indicator of fall risk, but the relationship is not well understood. This study examined daily PA accumulation and patterns with fall risk in older adults. Methods Participants (n=685, 54.3% female, 61.5% aged 70-79 years) from the National Health and Aging Trends Study (NHATS) with wrist-worn accelerometry PA data from Round 11 (baseline) and sample person interviews with fall data from Round 12 (follow-up) were included. PA variables were categorized into tertiles and incident falls were defined as ≥1 self-reported fall in the year following the PA assessment between baseline and follow-up. A modified Poisson approach was used to estimate the relative risk of both PA accumulation and fragmentation with falls. Results Overall, 40.0% reported an incident fall. After adjustment for sociodemographic and health characteristics, those in the highest tertile of total PA accumulation had lower fall risk (aRR=0.74, 95% CI: 0.57, 0.95) and those in the highest tertile of PA fragmentation had increased fall risk (aRR=1.33, 95% CI: 1.03, 1.73). Models were attenuated after adjustment for physical functioning. Conclusions PA fragmentation may identify fall risk in older adults. Longitudinal studies are needed to disentangle the temporal sequencing of the complex relationship between PA and physical functioning across the life-course.

中文翻译：

---

老年人体力活动分散和跌倒：国家健康和老龄化趋势研究的结果

背景 体力活动（PA）可能是一种重要的跌倒预防策略。目前的 PA 指南强调 PA 总剂量，但 PA 的每日模式却未被充分认识。随着年龄的增长，PA 发作的频率越来越低，持续时间也越来越短（即：更加分散）。 PA 碎片可能是跌倒风险的一个指标，但其中的关系尚不清楚。这项研究检查了老年人每日 PA 积累和跌倒风险的模式。方法 国家健康和老龄化趋势研究 (NHATS) 的参与者（n=685，其中 54.3% 为女性，61.5% 年龄为 70-79 岁），具有第 11 轮（基线）的腕戴式加速度 PA 数据以及对跌倒数据进行的样本访谈第 12 轮（后续）的比赛也包括在内。 PA 变量被分为三分位数，跌倒事件定义为在基线和随访期间进行 PA 评估后一年内自我报告的跌倒次数≥1 次。使用改良的泊松方法来估计跌倒时 PA 积累和碎片的相对风险。结果 总体而言，40.0% 的人报告发生过跌倒事件。调整社会人口统计学和健康特征后，PA 总积累最高三分位的人跌倒风险较低（aRR=0.74，95% CI：0.57，0.95），PA 碎片最高三分位的人跌倒风险较高（aRR=0.74，95% CI：0.57，0.95）。 1.33，95% CI：1.03，1.73）。在调整身体机能后模型被削弱。结论 PA 碎片可以识别老年人跌倒的风险。需要进行纵向研究来理清整个生命过程中 PA 与身体机能之间复杂关系的时间顺序。