Covinsky K E, Kahana E, Kahana B, Kercher K, Schumacher J G, Justice A C
Division of Geriatrics and Department of Medicine, University of California, San Francisco, USA.
J Gerontol A Biol Sci Med Sci. 2001 Apr;56(4):M253-9. doi: 10.1093/gerona/56.4.m253.
Falls are common in community-dwelling elderly persons and are a frequent source of morbidity. Simple indices to prospectively stratify people into categories at different fall-risk would be useful to health care practitioners. Our goal was to develop a fall-risk index that discriminated between people at high and low risk of falling.
We evaluated the risk of falling over a one-year period in 557 elderly persons (mean age 81.6) living in a retirement community. On the baseline interview, we asked subjects if they had fallen in the previous year and evaluated risk factors in six additional conceptual categories. On the follow-up interview one year later, we again asked subjects if they had fallen in the prior year. We evaluated risk factors in the different conceptual categories and used logistic regression to determine the independent predictors of falling over a one-year period. We used these independent predictors to create a fall-risk index. We compared the ability of a prior falls history with other risk factors and with the combination of a falls history and other risk factors to discriminate fallers from nonfallers.
A fall in the previous year (OR = 2.42, 95% CI = 1.49-3.93), a symptom of either balance difficulty or dizziness (OR = 1.83, 95% CI = 1.16-2.89), or an abnormal mobility exam (OR = 2.64, 95% CI = 1.64-4.26) were independent predictors of falling over the subsequent year. These three risk factors together (c statistic =.71) discriminated fallers from nonfallers better than previous history of falls alone (c statistic =.61) or the symptomatic and exam risk factors alone (c statistic =.68). When combined into a risk index, the three independent risk factors stratify people into groups whose risk for falling over the subsequent year ranges from 10% to 51%.
A history of falling over the prior year, a risk factor that can be obtained from a clinical history (balance difficulty or dizziness), and a risk factor that can be obtained from a physical exam (mobility difficulty) stratify people into groups at low and high risk of falling over the subsequent year. This risk index may provide a simple method of assessing fall risk in community-dwelling elderly persons. However, it requires validation in other subjects before it can be recommended for widespread use.
跌倒在社区居住的老年人中很常见,并且是发病的常见原因。对于医疗保健从业者而言,能前瞻性地将人们分为不同跌倒风险类别的简单指标会很有用。我们的目标是开发一种能区分跌倒高风险和低风险人群的跌倒风险指数。
我们评估了居住在退休社区的557名老年人(平均年龄81.6岁)在一年时间内的跌倒风险。在基线访谈中,我们询问受试者前一年是否跌倒,并评估了另外六个概念类别中的风险因素。在一年后的随访访谈中,我们再次询问受试者前一年是否跌倒。我们评估了不同概念类别中的风险因素,并使用逻辑回归来确定一年期间跌倒的独立预测因素。我们使用这些独立预测因素创建了一个跌倒风险指数。我们比较了既往跌倒史与其他风险因素以及既往跌倒史与其他风险因素的组合区分跌倒者和非跌倒者的能力。
前一年跌倒(比值比[OR]=2.42,95%置信区间[CI]=1.49 - 3.93)、平衡困难或头晕症状(OR = 1.83,95%CI = 1.16 - 2.89)或异常的活动能力检查(OR = 2.64,95%CI = 1.64 - 4.26)是随后一年跌倒的独立预测因素。这三个风险因素共同作用(c统计量=0.71)区分跌倒者和非跌倒者的能力优于单独的既往跌倒史(c统计量=0.61)或单独的症状和检查风险因素(c统计量=0.68)。当组合成一个风险指数时,这三个独立风险因素将人们分为随后一年跌倒风险范围从10%到51%的不同组。
前一年的跌倒史、一个可从临床病史中获得的风险因素(平衡困难或头晕)以及一个可从体格检查中获得的风险因素(活动能力困难)将人们分为随后一年跌倒低风险和高风险组。这个风险指数可能提供了一种评估社区居住老年人跌倒风险的简单方法。然而,在推荐广泛使用之前,它需要在其他受试者中进行验证。