TY - JOUR PY - 2016// TI - A comparison of self-report and performance-based balance measures to predict recurrent falls in people with Parkinson disease: a cohort study JO - Physical therapy A1 - Almeida, Lorena R. S. A1 - Valença, Guilherme T. A1 - Negreiros, Nádja N. A1 - Pinto, Elen B. A1 - Oliveira-Filho, Jamary SP - 1074 EP - 1084 VL - 96 IS - 7 N2 - BACKGROUND: Balance confidence and fear of falling are factors associated with recurrent falls in individuals with Parkinson disease (PD). However, the accuracy for falls prediction based on self-report measures has not been widely investigated.

OBJECTIVE: To compare the accuracy of the Activities-specific Balance Confidence Scale (ABC) and Falls Efficacy Scale-International (FES-I) with the Berg Balance Scale (BBS), Dynamic Gait Index (DGI), Functional Reach Test (FRT) and Timed Up and Go (TUG) in identifying recurrent fallers with PD; to explore the ability of combinations of up to three tests to predict recurrent falls.

DESIGN: Prospective cohort study with individuals with PD (n = 225).

METHODS: Participants were assessed with ABC, FES-I, BBS, FRT, TUG and DGI. Those who reported ≥2 falls in the 12-month follow-up period were classified as recurrent fallers. Areas under the ROC curves (AUC) were determined and Akaike information criterion (AIC) was used to select the best predictive model.

RESULTS: Eighty-four (37.3%) participants were classified as recurrent fallers. The AUC for the ABC was 0.73, for the FES-I 0.74 and for the TUG, FRT, DGI and BBS was 0.72, 0.74, 0.76 and 0.79, respectively. Two-test models provided additional discriminating ability when compared to individual measures and had similar AIC as three-test models, particularly BBS with FES-I. LIMITATIONS: There is not an external validation sample.

CONCLUSIONS: ABC and FES-I demonstrated moderate accuracy in predicting recurrent falls, showing similar predictive ability as performance-based balance measures, especially FRT and TUG. Two-test models showed similar performance as three-test models, suggesting that a combination of two measures may improve the predictive ability of recurrent falls in PD. Specifically, the utilization of BBS along with FES-I may be considered.

Language: en

LA - en SN - 0031-9023 UR - http://dx.doi.org/10.2522/ptj.20150168 ID - ref1 ER -