Cancer Prevention Research Centre, School of Population Health, The University of Queensland, Brisbane, Australia.
Med Sci Sports Exerc. 2011 Nov;43(11):2127-33. doi: 10.1249/MSS.0b013e31821b94f7.
With evidence that prolonged sitting has deleterious health consequences, decreasing sedentary time is a potentially important preventive health target. High-quality measures, particularly for use with older adults, who are the most sedentary population group, are needed to evaluate the effect of sedentary behavior interventions. We examined the reliability, validity, and responsiveness to change of a self-report sedentary behavior questionnaire that assessed time spent in behaviors common among older adults: watching television, computer use, reading, socializing, transport and hobbies, and a summary measure (total sedentary time).
In the context of a sedentary behavior intervention, nonworking older adults (n = 48, age = 73 ± 8 yr (mean ± SD)) completed the questionnaire on three occasions during a 2-wk period (7 d between administrations) and wore an accelerometer (ActiGraph model GT1M) for two periods of 6 d. Test-retest reliability (for the individual items and the summary measure) and validity (self-reported total sedentary time compared with accelerometer-derived sedentary time) were assessed during the 1-wk preintervention period, using Spearman (ρ) correlations and 95% confidence intervals (CI). Responsiveness to change after the intervention was assessed using the responsiveness statistic (RS).
Test-retest reliability was excellent for television viewing time (ρ (95% CI) = 0.78 (0.63-0.89)), computer use (ρ (95% CI) = 0.90 (0.83-0.94)), and reading (ρ (95% CI) = 0.77 (0.62-0.86)); acceptable for hobbies (ρ (95% CI) = 0.61 (0.39-0.76)); and poor for socializing and transport (ρ < 0.45). Total sedentary time had acceptable test-retest reliability (ρ (95% CI) = 0.52 (0.27-0.70)) and validity (ρ (95% CI) = 0.30 (0.02-0.54)). Self-report total sedentary time was similarly responsive to change (RS = 0.47) as accelerometer-derived sedentary time (RS = 0.39).
The summary measure of total sedentary time has good repeatability and modest validity and is sufficiently responsive to change suggesting that it is suitable for use in interventions with older adults.
有证据表明长时间久坐会对健康造成有害影响，因此减少久坐时间是一个潜在的重要预防健康目标。需要高质量的测量方法，尤其是针对最久坐的老年人群体，以评估久坐行为干预的效果。我们研究了一种自我报告的久坐行为问卷的可靠性、有效性和对变化的反应能力，该问卷评估了老年人常见的行为所花费的时间：看电视、使用计算机、阅读、社交、交通和爱好，以及一个总结性的衡量标准（总久坐时间）。
在一项久坐行为干预的背景下，非工作的老年人（n = 48，年龄 = 73 ± 8 岁（平均值 ± 标准差））在两周的时间内三次完成问卷（两次之间间隔 7 天），并且在两个 6 天的时间段内佩戴加速度计（ActiGraph 模型 GT1M）。在干预前的 1 周内，使用 Spearman（ρ）相关系数和 95%置信区间（CI）评估测试-重测信度（个体项目和总结性衡量标准）和有效性（自我报告的总久坐时间与加速度计得出的久坐时间相比）。干预后的反应能力变化使用反应性统计量（RS）进行评估。
看电视时间（ρ（95%CI）= 0.78（0.63-0.89））、使用计算机时间（ρ（95%CI）= 0.90（0.83-0.94））和阅读时间（ρ（95%CI）= 0.77（0.62-0.86））的测试-重测可靠性非常好；爱好时间（ρ（95%CI）= 0.61（0.39-0.76））的测试-重测可靠性可接受；社交时间和交通时间的测试-重测可靠性较差（ρ < 0.45）。总久坐时间的测试-重测可靠性可接受（ρ（95%CI）= 0.52（0.27-0.70）），有效性适中（ρ（95%CI）= 0.30（0.02-0.54））。自我报告的总久坐时间与加速度计得出的久坐时间变化的反应能力相似（RS = 0.47）。
总久坐时间的总结性衡量标准具有良好的可重复性和适度的有效性，并且对变化的反应足够敏感，这表明它适合在老年人干预中使用。
