Occupational physical activity: Reliability and comparison of activity levels

Background: In 2001, the Behavioral Risk Factor Surveillance System (BRFSS) included a new occupational physical activity (PA) question. This article evalu-ates the reliability of this survey question. Methods: Forty-six subjects were followed for 3 wk, answered 3 PA surveys by telephone, and completed daily PA logs for 1 wk. Kappa statistics determined the reliability of occupational activities (sitting/standing, walking, and heavy lifting). A descriptive analysis compared the time in specific occupational activities. Results: Eighty percent of the respondents reported “mostly sitting or standing” at work; and test–retest reliability was moderate (k = 0.40 to 0.45). The occupationally inactive sat/stood for 85% (mean hours = 5.6) of the workday, whereas the occupationally active sat/stood for 53% (mean hours = 3.9) of the workday. Conclusions: The BRFSS occupational activity question has moderate reliability, distinguishes between occupationally active and inactive persons, and can be used in surveillance systems to estimate adult occupational PA