CROSS-SECTIONAL AND LONGITUDINAL COMPARISON OF SELF-REPORT VERSUS OBJECTIVE MEASUREMENT IN PHYSICAL ACTIVITY AMONG WOMEN

Abstract

Physical activity improves health while combating the obesity epidemic. However, quantifying physical activity through self-report questionnaires or objective measures can provide varying results. The purposes of these studies were to determine if time, body mass index, or treatment assignment could affect the validity of physical activity measurements. The data were part of a larger physical activity promotion study conducted at the Miriam Hospital/Brown Medical School in Providence, Rhode Island and in communities in Southeastern Massachusetts from 2002 to 2005. In this trial, 280 women, with a mean age of 47.1 years, were randomly assigned to one of three intervention groups: Choose to Move (n=93), Jumpstart (n=95) and Wellness (n=92). A randomly selected sub sample of participants simultaneously wore an ActiGraph accelerometer and completed a 3-Day Physical Activity Recall questionnaire at baseline, 3 months and 12 months. Body mass index and treatment assignment were also used in-group comparisons. The results indicated that all components of time, BMI and treatment assignment influenced the accuracy of self-reported measurements when compared to objective accelerometer data. Additional research is essential to uncover the independent aspects considered influential to these physical activity measurements to enhance study design and participant outcomes in future trials.