Falls in long-term care facilities are a major problem. Becker and colleagues conducted a prospective, cluster-randomized controlled trial to study a multifaceted, nonpharmaceutical intervention that would allow participants to self-select their fall prevention modalities.

Three long-term care facilities were randomized to the intervention group and three to the control group. End points included fall density rate, frequent falls, and fractures. Participating staff received training in fall prevention, residents were given written educational materials on fall prevention, and an environmental hazard assessment was conducted by the staff. Residents also were given the option to participate in an exercise program and to wear hip protectors.

Of the 981 study participants, 509 were assigned to the intervention group. Of these, 127 participated in more than one exercise class (mean number of classes: 33). Of the 160 residents who decided to wear hip protectors, 108 had a 100 percent rate of adherence. Determining adherence to environmental corrections was not feasible.

Incidence density rate of falls was greater for frequent fallers (relative risk [RR]: 0.56) than for all fallers (RR: 0.75), with no difference in hip fractures between the intervention group and the control group (RR: 1.11). Time to first fall declined in the intervention group after six months.

The study was underpowered to detect a difference in hip fractures because of an unexpectedly low incidence of hip fractures in the control group. The authors conclude that a learning curve is associated with fall prevention in nursing home residents, with both staff and resident involvement required. To be successful, fall prevention programs may require longer time frames than those currently studied.