Comparisons of Laboratory Tests and Simple Clinical Tests for Identifying Fallers among Healthy Older Persons

Abstract

Reduced postural control and related falls are a major cause of injury and death in older individuals. PURPOSE: To examine the capacity of a number of field and laboratory tests to identify fallers in a sample of older independently-living persons. METHODS: 66 healthy, independently-living older volunteers, ages 60 and older made three laboratory visits during which field tests (Timed Up and Go [TUG], One-Leg Stand [OLS], Functional Reach [FR], and Tinetti Performance Oriented Mobility Assessment [POMA]) and laboratory-based tests (force plate center of pressure [COP] and time-to-boundary [TTB], and dynamic posturography [Proprio 5000]) were compared to ascertain the capacities of these tests to distinguish fallers from non-fallers. Each participant was classified as a "faller" or "non-faller" based on whether he or she recalled experiencing a fall within the past year. RESULTS: Receiver-operated characteristics (ROC) curve analyses revealed the OLS and TUG had the capacity to distinguish fallers from non-fallers (P=0.002 and 0.008, respectively) as did selected Proprio 5000 and TTB variables (P<.015 and P<.042, respectively). One-way ANOVA results reflected ROC analyses with significant between-group differences for the OLS (P=0.002), TUG (P= 0.008), POMA (P<0.044), Proprio 5000 (P<.013) and TTB (P<.047). For the field tests, the OLS and TUG showed the largest effect sizes (0.93 and 0.79, respectively), while for laboratory tests, the Proprio 5000 revealed the largest effect sizes ranging from.73 to.88 depending on the plane of motion evaluated. CONCLUSION: Selected field and laboratory tests have the capacity to identify fallers in a healthy, independent-living sample of older persons. In addition, the laboratory tests revealed balance decrements in specific planes of motion that can provide more detailed information concerning directional falls risk. When examining the directional planes specifically, the rotational rather than translational variables were indicative of fallers. Unlike results previously reported for younger subjects with better balance than their contemporaries, high rotational scores were produced by non-fallers perhaps indicating increased degrees of freedom, as they employed multi-joint and multi-directional balance recovery strategies.