Date of Award
2008
Degree Type
Thesis
Degree Name
Doctor of Philosophy
Program
Epidemiology and Biostatistics
Abstract
The purpose of this dissertation was to add to our understanding of falls, fall- related injuries, and physical activity (PA) in community-dwelling older adults. The specific aims were to determine the relationship between PA and falls, explore alternative analysis approaches for falls as an outcome measure, and assess the relationship between injurious falls and subsequent PA (both household and recreational) in older male veterans. In order to carry out this research, related aims of this project were to devise and evaluate a brief interview tool appropriate for the assessment of PA in fall prevention studies of community-dwelling older adults (the Phone-FITT), and to evaluate two additional single questions of PA.
The main data source for the present thesis arose from a trial evaluating the effectiveness of preventing falls through risk factor screening and modification. A variety of regression models for time to event were adopted to examine the relationship between PA and falls. Intraclass correlation coefficients, unweighted and weighted kappa statistics were used to measure test-retest reliability, and different correlation coefficients (e.g. Cramer’s v, phi, point-biserial, and Spearman’s rho) were used to evaluate criterion- related and construct validity of the PA measures. The precision of these coefficients was quantified using two-sided 95% confidence intervals. Two multiple linear regression analyses were conducted to assess the relationship between injurious falls and subsequent
household and recreational PA. The regression models for recurrent events showed that male veterans who
reported being less active than their peers were significantly more likely to fall than those who reported being about as or more active than their peers. Initial evidence for test-
iii
retest reliability as well as various forms of validity (concurrent, convergent, discriminant and known-groups) was found for the Phone-FITT, and for the single questions of PA. Injurious falls were found to be significantly associated with lower household PA and higher recreational PA in older male veterans. These results may have important implications specifically for the measurement of PA in large samples of older adults and the analysis of fall outcome data, and more generally fall prevention and the promotion of PA in community-dwelling older adults.
Recommended Citation
Gill, Dawn P., "FALLS, FALL-RELATED INJURIES AND PHYSICAL ACTIVITY IN COMMUNITY-DWELLING OLDER ADULTS" (2008). Digitized Theses. 4774.
https://ir.lib.uwo.ca/digitizedtheses/4774