Using computers to reduce falls

Patrick Hardigan, PhD, executive director for Health Professions Division research at NSU, developed a predictive modeling algorithm based on information gathered from unidentified patients in Florida, including their age, height and weight. The physical data was then combined with each individual's clinical diagnosis and medication history to determine whether they were more likely or less likely to fall. "Our goal is to develop a multidisciplinary fall prevention program and ultimately reduce the number of deaths and serious injuries due to falls," Hardigan said in a press release.

Gary S. Margules, ScD, vice president for research and technology transfer at NSU, added, "Statistical models like this offer significant inherent advantages for clinicians because they can correctly register the simultaneous importance of a dozen or more factors."

It is hoped that this new algorithm will be incorporated into a easy-to-use template for use in  long-term care settings, hospitals and clinics so that caregivers can develop personalized fall prevention strategies based on resident-specific risk profiles.