Objective: To explore whether clinical balance and gait tests can predict instability (falls and/or near falls) in people with Parkinson’s disease (PD).

Background: Current PD-studies suggest that multiple balance tests should be used in order to predict falls. However, few studies have included near falls when investigating falls prospectively as recommended.

Methods: The study included 74 people with PD (mean age and PD-duration, 69 and 4.9 years, respectively) visiting a neurological clinic during 2006–2010. Those >80 years of age, requiring support in standing or did not understand the instructions were excluded. Assessments included: the Berg Balance Scale (BBS, 0–56 points), Nutt Retropulsion test (NRT, dichotomized; 0 = "normal”, 1 = “abnormal”), tandem gait test (TG, dichotomized; 0 = “normal”, 1 = “abnormal”), 10-meter walk test (fast speed, m/s), and Timed Up & Go test (TUG, s). All assessments were conducted in the “on” condition. Participants then registered all falls and near falls by using a diary for six months.

Results: Mean score for UPDRS III was 14 (SD 7.5). Thirty-six participants (49%) experienced 1 fall and/or near fall (“unstable” group), whereas 38 (51%) had no incidents at all (“stable” group). Simple logistic regression analyses (controlling for age and gender) showed that (P0.05 in all instances) the NRT was the strongest predictor (OR = 5.70) followed by TG (OR = 3.45). Better BBS-scores (OR = 0.88) and gait speed (OR = 0.26) were associated with a decreased risk of instability. The longer time to perform TUG, the higher risk of being unstable (OR = 1.14). When considering all five variables (i.e. tests) simultaneously only BBS was found significant (OR = 0.91, P 5 0.04).

Conclusions: Clinical balance and gait tests can predict a future instability in people with PD. Further studies using larger samples are needed for firmer conclusions and to establish sensitivity/specificity and cut-off values for these tests.