Falls are significant risks to mortality and morbidity in the elderly. Now, university researchers have developed a wireless device able to predict when a person might fall, long before the fall actually happens. The device uses a series of miniature sensors composed of accelerometers and gyroscopes to analyze trends in posture and gait. It then issues an alert if there is a break in routine.

Such a technology could benefit potentially unstable geriatric patients, as well as patients with balance issues from disorders like Parkinson's disease, epilepsy and dementia. The alarm would instruct patients to grab hold of something or sit down when an impending fall is calculated.

"We're creating an inexpensive device to help us understand fall dynamics in different geriatric demographics and, ultimately, to incorporate algorithms that will lead to early warning and even fall prevention," says Donald Lie, chairman of electrical and computer engineering, Texas Tech University.

The best placement of the device on the body is still under consideration. Researchers have tried leg-mounted inertial sensors and slippers with pressure sensors in their soles, as well as placing the tiny device on the spine between the shoulder blades.

C. Gresham Bayne, M.D., founder and chairman of Call Doctor Medical Group, San Diego, and a leader in reviving the notion of house-call medicine, believes the Texas Tech device has potential in home care since risk of falls is common among the elderly because of medications and physical changes.