Prognostic factors that modify outcomes of vestibular rehabilitation in elderly patients with falls.

Background: Our previous study has shown that vestibular rehabilitation (VR) is an effective technique to reduce falls in elderly patients. It would be interesting to establish patients' clinical characteristics in which vestibular rehabilitation is expected to be more effective.

Aims: Evaluate factors that could modify rehabilitation outcomes in elderly patients with previous falls.

Methods: Fifty-seven patients randomized to one of the intervention group (computerized dynamic posturography-CDP-training, optokinetic stimulus or exercise at home) and with previous falls were analyzed. Patients were assessed with objective outcome measures (sensorial organization test and limits of stability-LOS-of CDP, modified timed up and go test-TUG-and number of falls) and with subjective outcome measures (dizziness handicap inventory and Short falls efficacy scale-international-Short FES-I) during a 12-month follow-up period.

Results: In the logistic regression model, a worse score in the maximum excursion (MXM), and a shorter time in the TUG significantly associated with a reduction > 50% of falls. Also, association with a higher score in the Short FES-I was close to a statistical significance. There was no statistical significance association with other covariables.

Discussion: In patients with reduced limits of stability, VR seems to be more effective and they should be encouraged to perform it. But on the other hand, patients with longer time in the TUG show worse outcomes and may benefit more with gait training.

Conclusions: VR in elderly people with previous falls is effective regardless of their age and gender.