1 / 1

Comparative Quality of Aging and Psychological Predictors on Fall-Related Outcomes

Comparative Quality of Aging and Psychological Predictors on Fall-Related Outcomes. Michael L. Stellefson, Ph.D. 1 ; John F. Yannessa, Ph.D. 2 , & Greg F. Martel, Ph.D. 2. Department of Health Education and Behavior, University of Florida, Gainesville, FL 1

garren
Télécharger la présentation

Comparative Quality of Aging and Psychological Predictors on Fall-Related Outcomes

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Comparative Quality of Aging and Psychological Predictors on Fall-Related Outcomes Michael L. Stellefson, Ph.D.1; John F. Yannessa, Ph.D.2, & Greg F. Martel, Ph.D.2 Department of Health Education and Behavior, University of Florida, Gainesville, FL1 College of Science, Coastal Carolina University, Conway, SC2 Data Analysis Unique & Common Components of Predictive Ability of Each Independent Variable on LBFUNCT I Abstract Purpose: The effects of important variables measuring the psychobiological aspects of falls among older adults were considered to determine their utility in predicting balance functioning among older adults. To partition the effects of aging and falls efficacy on balance and leg strength simultaneously, canonical commonality analysis (CCA) was used. Methods: CCA is a multivariate technique which decomposes squared semi-partial correlation effect sizes into constituent, non-overlapping segments that describe unique and common explanatory powers of predictor variables. Data from a study conducted to examine the psychobiological and aging influences on unintended falls among physically active older adults were analyzed. Findings: CCA showed balance confidence as measured by the Activities-Specific Balance Confidence Scale (ABC) and age to be noteworthy predictors of balance; yet, age was determined to be more important than balance confidence when predicting balance and leg strength (i.e., balance functioning) simultaneously. In addition, results suggested that data obtained from the ABC better predicted balance functioning among active older adults as compared to the Tinetti Falls Efficacy Scale (FES), a traditional measure used to assess the construct. Conclusion: The ABC stands as a viable alternative to consider when assessing falls efficacy among dynamic older adults. Future research would benefit from using CCA to understand how various psychobiological constructs predict fall-related outcomes. • The distribution of scores on all scales was examined, and the skewness and kurtosis statistics were examined for normality • A canonical correlation analysis determined the correlation between the pair of composite variables (i.e., the FES/ABC/AGE function with BBT/LGPRS): Λ = .49, F(6, 160) = 11.46, p < .001 • Each individual participant score on the LBFUNCT I (Rc2 = .509) was then computed by multiplying participants’ z-scores on the two criterion variables (i.e., BBT/LGPRS) and summed together to get product variate scores, which were saved under a unique name that represented the concurrent composition of each dependent variable within the variate. • Once LBFUNCT I z-scores were computed, simple and multiple regressions predicted these scores with all possible combinations of the predictor variable set. There were seven possible combinations using at least one of the 3 independent measures (i.e., FES, ABC, and AGE) to predict the LBFUNCT I scores, which reflected latent multivariate measures (rc2/Rc2). Discussion • The majority of the unique explained variance on the LBFUNCT I was predicted by AGE, followed by the ABC measure of falls efficacy, followed by the FES. Moreover, age uniquely explained more variance in balance and leg strength, and was a superior predictor to falls efficacy. • Most notably, over 90% (15.4% / 16.8% = 91.7%) of the rc2 attributable to variance in FES scores was commonly shared among the ABC and AGE variables, albeit mostly with the ABC measure of falls efficacy (7.1% / 16.8% = 42.2%), suggesting one could reasonably drop FES from the predictive model without losing much predictive power. • Results from this study, and others involving moderate to high functioning older adults (e.g., Myers, et al., 1996), suggest that the ABC does a better job predicting lower body functioning among active older adults. Formulas for determining unique and common components of shared variance Summary Statistics (n = 85) For more: Stellefson, M. L., Yannessa, J. F., & Martel, G. F. (2011). Using univariate and canonical commonality analysis to examine the predictive quality of psychological variables and aging on physical performance among older adults. Evaluation and the Health Professions. (in press)

More Related