Rolland Yves, Lauwers-Cances Valerie, Cesari Matteo, Vellas Bruno, Pahor Marco, Grandjean Hélène
Service de Médecine Interne et de Gérontologie Clinique, Hôpital La Grave-Casselardit, 31300, Toulouse, France.
Eur J Epidemiol. 2006;21(2):113-22. doi: 10.1007/s10654-005-5458-x.
This prospective cohort study evaluate the predictive value of physical performance measures for mortality in older French women, in particular those with a high health status. The subjects were 7,250 community-dwelling non-disabled French women aged 75 years or older, enrolled in the Epidémiologie de l'ostéoporose (EPIDOS) study. The short physical performance battery (SPPB), including walking speed, repeated chair stands, and balance tests, was administered and handgrip strength was measured. Anthropometric measurements, physical function, cognitive performance, sensory status, smoking, medical history, medication use, subjective self-assessment of health status, and physical activity level were assessed at the baseline visit. During a mean follow-up of 3.8 years, 754 (10.4%) participants died. Complementary analysis was performed on the 2,157 non-disabled healthiest participants (no disease at baseline). The SPPB and handgrip strength distinguished a gradient of risk for mortality from a low to high functional spectrum. Risk of death was 2.04-fold higher in poor (SPPB 0-6) than in good (SPPB 10-12) performers and 1.56-fold higher in participants with lower tertile grip strength. Walking speed alone also distinguished a gradient of mortality risk. After adjustment for confounders, low SPPB, grip strength score and slow walking speed remained significantly associated with death. In the non-disabled healthiest women, no physical performance measure predicted death. In community-dwelling elderly French women, physical performance measures significantly and independently predicted mortality. Increased risk of death was partly explained by baseline health status and was absent in the healthiest elderly.
这项前瞻性队列研究评估了身体机能指标对法国老年女性，尤其是健康状况良好的老年女性死亡率的预测价值。研究对象为7250名年龄在75岁及以上、居住在社区且无残疾的法国女性，她们参与了骨质疏松症流行病学（EPIDOS）研究。研究人员实施了简短身体机能测试组合（SPPB），包括步行速度、重复起坐测试和平衡测试，并测量了握力。在基线访视时评估了人体测量指标、身体功能、认知能力、感觉状态、吸烟情况、病史、用药情况、健康状况主观自评以及身体活动水平。在平均3.8年的随访期内，754名（10.4%）参与者死亡。对2157名最健康的无残疾参与者（基线时无疾病）进行了补充分析。SPPB和握力区分了从低到高功能谱的死亡风险梯度。表现较差（SPPB为0 - 6）的参与者死亡风险比表现良好（SPPB为10 - 12）的参与者高2.04倍，握力处于三分位数较低水平的参与者死亡风险高1.56倍。仅步行速度也区分了死亡风险梯度。在对混杂因素进行调整后，低SPPB、握力得分和慢步行速度仍与死亡显著相关。在最健康的无残疾女性中，没有身体机能指标能够预测死亡。在居住在社区的法国老年女性中，身体机能指标能够显著且独立地预测死亡率。死亡风险增加部分可由基线健康状况解释，而在最健康的老年人中则不存在这种情况。
