Validity and inter-observer reliability of the TURN 180 test to identify older adults who reported falls

Background: Falls while turning are associated with increased risk of hip fracture in older adults. Reliable and clinically valid methods for turn ability assessments are needed. Objectives: To explore the inter-observer reliability and known group validity of the TURN 180 test. Methods: We divided 78 independent older adults (mean age 76.6 ± 6.5 years) into three groups: non-fallers, infrequent fallers (1–2 falls per year), and recurrent fallers (> 2 falls per year). Participants underwent performance-based tests: Timed Up and Go (TUG), Performance Oriented Mobility Assessment (POMA), and Berg Balance Scale (BBS). TUG was videotaped for later analysis of the TURN 180 test by two blinded observers. Results: A significant difference was found in the TURN 180 test parameters among the groups (P < 0.04). TURN 180 was highly correlated with TUG (r = 0.81–0.89, P < 0.001) and BBS (r =-0.704–0.754, P < 0.0001) and moderately with POMA (r =-0.641–0.698, P < 0.0001). The number of steps was found to be the strongest parameter to determine fallers among older adults (specificity 96.3%, sensitivity 40%). Inter-rater reliability (intraclass correlation coefficient 0.91–0.96, P < 0.0001) was found to be excellent for the number of steps, time taken to accomplish a turn, and total test score categories. Conclusions: The TURN 180 test is highly reliable and can identify the older adults who fall. Our results show that the TURN 180 test can serve as a good performance-based examination for research or clinical setting.