Bogle Thorbahn and Newton performed a prospective cohort study to determine if the Berg Balance Assessment was a reliable measure at predicting an elderly patient’s fall risk. According to the evidence, the Berg Balance Assessment had low sensitivity for identifying those who would fall, but was highly specific at correctly identifying individuals without a history of falls. In addition, the Berg Balance Assessment proved useful at predicting a person’s use of an assistive device with a sensitivity of 76% and specificity of 94%. The authors concluded that falls are multi-factorial and individuals should be assessed in the environment in which they operate in order to fully determine if a person is at risk for falling. The evidence did not support the use of the Berg to identify fallers, as shown by the low sensitivity; however, did support the use of the test to rule out individuals as non-fallers. The sole use of self-report measures to report fall history and the use of a convenience sample may be limitations of the study.
Blasco, Paige, "Is The Berg Balance Assessment A Reliable Predictor Of Fall Risk In Older Adults?" (2014). Critically Appraised Topics (CAT). 5.