Validity of Two Self-Report Measures of Sitting Time

Introduction Evidence suggests that time spent in sedentary behaviours (usually defined as time spent sitting) is an independent risk factor for a number of adverse health outcomes [1] and there have been calls for the explicit measurement of sedentary behaviour in population surveillance studies [2]. To date, limited evidence exists on the validity of self-reported measures of sedentary behaviour. This study examined the criterion validity of two different self-report measures of sitting time.

Methods 44 healthy volunteers (30% male, age = 41.5±12.8 years, BMI = 24.8±4.7 kg/m²) wore an ActiGraph accelerometer (the criterion measure) for seven consecutive days. During each day of monitoring, participants recorded their daily sitting time by responding to the single-item question “how long have you been sitting for today?” in a daily diary. Participants completed this specific-day question on a daily basis just before going to bed each night. After seven days, participants completed a new domain-specific sitting time questionnaire that requires participants to retrospectively report sitting times in different domains (i.e. time spent sitting whilst travelling to and from places, at work, watching TV, using a computer at home, and during leisure time), on a usual weekday and weekend-day [3]. Total sitting times recorded from the single-item question and from the domain-specific questionnaire were compared with mean accelerometer-determined sedentary time (minutes/day of <100 counts/minute) for weekdays and weekend-days.

Results Accelerometer-determined sedentary times (mean ± SD) for weekdays and weekend-days were 639.7±133.9 and 612.4±132.7 minutes respectively. Total sitting time calculated from the domain-specific questionnaire did not differ significantly from accelerometer-determined sedentary time on either weekdays (mean difference = -13.7 minutes [95% CI = -69.2 to 41.8]) or weekend-days (-4.2 minutes [95% CI = -91.7 to 83.4]). Sitting time was significantly underestimated using the single-item specific-day question on weekdays (-173 minutes [95% CI = -207.5 to -138.5]) and weekend-days (-219 minutes [95% CI = -262.9 to -174.2]). From the domain-specific questionnaire, the greatest contributor to daily sitting time was sitting at work on weekdays and watching TV on weekends.

Discussion and Conclusion When assessed via self-report, the estimation of total sitting time is improved by summing sitting times reported across different domains. Estimates of sitting time are more precise for weekdays than for weekend-days. The continued improvement of self-report measures of sitting time, particularly for weekend-days, will be important if we are to further our understanding of the links between sedentary behaviour and health.

References [1] Katzmarzyk PT, Church TS, Craig CL, Bouchard C. Sitting time and mortality from all causes, cardiovascular disease, and cancer. Med Sci Sports Exerc, 2009; 41: 998- 1005. [2] Owen N, Leslie E, Salmon J, Fotheringham MJ. Environmental determinants of physical activity and sedentary behavior. Exerc Sport Sci Rev, 2000; 28: 153-8. [3] Marshall AL, Miller YD, Burton NW, Brown WJ. Measuring total and domain-specific sitting: a study of reliability and validity. Med Sci Sports Exerc, 2010; 42: 1094-102.