New approach to risk determination: development of risk profile fornew falls among community-dwelling older people by use of a GeneticAlgorithm Neural Network (GANN)

Open Access
Publisher Website
Google Scholar
Abstract
BACKGROUND: Falls risk in older people is multifactorial and complex.There is uncertainty about the importance of specific risk factors. Geneticalgorithm neural networks (GANNs) can examine all available data and selectthe best nonlinear combination of variables for predicting falls. The aimof this work was to develop a risk profile for operationally defined newfalls in a random sample of older people by use of a GANN approach.METHODS: A random sample of 1042 community- dwelling people aged 65 andolder, living in Nottingham, England, were interviewed at baseline (1985)and survivors reinterviewed at a 4-year follow-up (n = 690). The at-riskgroup (n = 435) was defined as those survivors who had not fallen in theyear before the baseline interview. A GANN was used to examine allavailable attributes and, from these, to select the best nonlinearcombination of variables that predicted those people who fell 4 yearslater. RESULTS: The GANN selected a combination of 16 from a potential 253variables and correctly predicted 35/114 new fallers (sensitivity = 31%;positive predictive value = 57%) and 295/321 nonfallers (specificity = 92%;negative predictive value = 79%); total correct = 76%. The variablesselected by the GANN related to personal health, opportunity, and personalcircumstances. CONCLUSIONS: This study demonstrates the capacity of GANNsto examine all available data and then to identify the best 16 variablesfor predicting falls. The risk profile complements risk factors in thecurrent literature identified by use of standard and conventionalstatistical methods. Additional data about environmental factors mightenhance the sensitivity of the GANN approach and help identify those olderpeople who are at risk of falling.

This publication has 0 references indexed in Scilit: