File
Moore B.1,2
1New York University, Physical Therapy, New York, United States, 2California State University, Physical Therapy, Sacramento, United States
Background: It is well known that an individual's performance can be influenced by the presence of an observer. This phenomenon is commonly referred to as the Hawthorne or observer effect. However, there is limited research on how the presence of an observer may affect the reliability and validity of gait speed measurement across patient populations.
Purpose: The aim of this study is to determine if there is agreement between the comfortable gait speed of individuals with vestibular dysfunction when they have knowledge of being observed, compared to their comfortable gait speed in a circumstance where they are unaware of being observed.
Methods: A methodological design was used to assess comfortable gait speed of 31 participants with vestibular dysfunction under two conditions: 1) the current, standardized method with instruction to walk at your normal pace, (known observer condition), and 2) a condition where the participant was unaware of the testing condition (unknown observer condition). Testing of both conditions were repeated over a 7-day interval to determine test-retest reliability.
Results: Test re-test reliability of gait speed measurement was excellent for the unknown observer condition (ICC value of 0.96) and the known observer condition (ICC value of 0.92). Standard error of the measurement (SEM) values were small (0.018 to 0.024 m/s) and the minimal detectable change (MDC95) ranged between 0.04 to 0.06 m/s. Participants, on average, walked significantly faster in the unknown observer trials, with a mean gait speed of 1.31 m/s (SD=0.17), than participants did in the known observer trials, mean gait speed of 1.23 m/s (SD=0.16), p .001.
Conclusion(s): There is low agreement between the comfortable gait speed of individuals with vestibular dysfunction when they have knowledge of being observed, compared to their comfortable gait speed when they are unaware of being observed. When measuring the mean difference between the two conditions, 80% percent of the participants exceeded the MDC95, which shows the presence of an assessor produces a clinically relevant difference in comfortable gait speed measurement for the majority of individuals with vestibular dysfunction. Future studies could investigate the presence of or degree to which an observer introduces bias on comfortable gait speed in other patient populations.
Implications: This report provides evidence that a consistent comfortable gait speed can be measured over a 7-day interval in individuals with vestibular dysfunction, in either an unknown or known observer condition. However, the potential for the observer effect to bias an individuals comfortable gait speed raises serious questions about the validity of the current, standardized method used to assess comfortable gait speed in individuals with vestibular dysfunction.
Funding acknowledgements: This work was supported by the Sacramento State Research and Creative Activity Faculty Awards Program.
Topic: Neurology
Ethics approval: This study was approved by the Institutional Review Board of the New York University Langone Medical Center.
All authors, affiliations and abstracts have been published as submitted.
