Four Square Step Test Performance in People With Parkinson Disease

Background and Purpose: The Four Square Step Test (4SST), a quick and simple test of multidirectional stepping, may be useful in predicting falls in people with Parkinson disease (PD). We studied the reliability of the 4SST and its ability to discriminate between freezers and nonfreezers, between fallers and nonfallers, and factors predictive of 4SST performance in people with PD. Methods: Fifty-three individuals with idiopathic PD completed the full protocol, including the 4SST as well as measures of balance, walking, and disease severity on anti-PD medication. Results: Interrater (intraclass correlation coefficient [ICC] = 0.99) and test-retest reliability were high (ICC = 0.78). The median 4SST performance was 9.52 seconds. There was a significant difference between 4SST time on-medication versus off (P = 0.03), while differences between fallers and nonfallers (P = 0.06) and between freezers and nonfreezers (P = 0.08) did not reach significance. All outcome measures were significantly related to 4SST time. In an exploratory, simultaneous regression analysis, 56% of the variance in 4SST performance could be accounted for by 3 measures: Mini-Balance Evaluation Systems Test (Mini-BESTest), Five Time Sit to Stand, and Nine Hole Peg Test. The 4SST cutoff score for distinguishing fallers from nonfallers was 9.68 seconds (Area under curve = 0.65, sensitivity = 0.73, specificity = 0.57). The posttest probability of an individual with a score greater than the cutoff being a faller was 31% (pretest probability = 21%). Discussion and Conclusions: The 4SST is a reliable, quick test that can distinguish between on-and off-medication conditions in PD but is not as good as other tests (eg, Mini-BESTest) for distinguishing between fallers and nonfallers.