Effect of investigator observation on gait parameters in individuals with and without chronic low back pain

Despite the ubiquity of gait assessment in clinic and research, it is unclear how observation impacts gait, particularly in persons with chronic pain and psychological stress. We compared temporal spatial gait patterns in people with and without chronic low back pain (CLBP) when they were aware and unaware of being observed. This was a repeated-measures, deception study in 55 healthy persons (32.0±12.4 yr, 24.2±2.7kg/m²) and persons with CLBP (51.9±17.9 yr, 27.8±4.4kg/m²). Participants performed one condition in which they were unaware of observation (UNW), and three conditions under investigator observation: (1) aware of observation (AWA), (2) investigators watching cadence, (3) investigators watching step length. Participants walked across an 8.4m gait mat, while temporal spatial parameters of gait were collected. The Medical Outcomes Short Form (SF-12), Beck Depression Inventory (BDI), State Trait Anxiety Inventory (STAI), and Oswestry Disability Index (ODI) were completed. Significant condition by group interactions were found for velocity and step length (p<0.05). Main effects of study condition existed for all gait variables except for step width. Main effects of group (healthy, LBP) were significant for all variables except for step width (p<0.05). Regression analyses revealed that after accounting for age, sex, and SF-12 mental component score, BDI scores predict velocity changes during walking from the UNW to AWA conditions. These findings show that people change their gait patterns when being observed. Gait analyses may require additional trials before data can reliably be interpreted and used for clinical decision-making.