Accuracy of Physical Activity Measures in Cancer Studies

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): Methodological challenges inherent in the assessment of physical activity in epidemiological studies introduce measurement errors that limit our understanding of the role of physical activity in reducing the risk of cancer and other diseases. The primary goal of this proposal is to identify and quantify the measurement error associated with two self administered physical activity (PA) questionnaires used to assess recent activity in two on-going cancer cohort studies (Life After Cancer Epidemiology Study, funded by the NCI, and the California Men's Health Study, funded by the California Cancer Research Program). To achieve this goal, we propose to conduct a validation study involving comparisons of the two test questionnaires with 1) 14 days of physical activity diaries (2 7-day periods, 5-6 months apart); 2) 14 days of accelerometer recordings, simultaneous with the PA diaries; 3) aerobic capacity as determined from a sub-maximal graded exercise treadmill test; and 4) body size measurements (body mass index, % body fat from bioelectric impedance, and waist circumference). The validation study will also include evaluation of 2-4 week test-retest repeatability of the two questionnaires. The study participants will be randomly drawn from the Northern California Kaiser Permanente (KP) Cancer Registry and general membership files and will consist of 400 women and men, ages 45-65, diagnosed with breast or prostate cancer 24-36 months prior to recruitment, and 400 women and men of similar age but with no history of cancer. The sample is expected to reflect the racial/ethnic diversity of the KP membership. In addition, we propose to develop an application program for a hand-held computer (personal digital assistant or PDA) that will be used for the collection and management of the PA diaries. We will evaluate the comparability of the PA diaries collected on the PDAs with traditional paper and pen diaries in a pilot study prior to the validation study. For the pilot study, 50 men and women, randomly drawn from the same sampling frame to be used in the validation study, will record their activities for a 4-day period in each of 3 consecutive weeks, with half using PDAs for two of the periods and paper and pen for one period, and the other half doing the reverse. The pilot study participants will also complete a cognitive interview to provide insights into the usability of the PDA software and the thought processes required in the recall of physical activity by the two test questionnaires. Sources of measurement error in the questionnaires will be identified and quantified, and statistical methods for adjusting for measurement error will be applied to the parent studies to estimate the risk of disease (breast cancer recurrence or incident prostate cancer) associated with physical activity. Because the validation will be conducted in a large, diverse, randomly selected sample, the findings may be potentially generalizable to other studies of physical activity and cancer, and may have broad applicability for studies of physical activity and other health outcomes as well.