Linear model for physical activity prediction in healthy middle aged men

Introduction: The usage of valid, reliable, rapid and low cost instruments for measurement of physical activity (PA) volume are essential in epidemiological and clinical studies. The purpose of this study was to develop a new physical activity prediction model using physiological and anthropometric variables in middle aged men. Methods: Daily physical activity was measured with pedometer throughout the day for 2 consecutive week in 101 40-65 years old men. Anthropometric variables, total daily energy expenditure and Vo2rest were measured. Linier regression was used to obtain prediction model with independent variables. Results: A significant correlation was found between average steps per day and independent variable P<0.001). Daily steps was predictable with BMI, BMR, TDEE and VO2rest (r2=0.961, SEE=767 steps/day, P<0.001). A significant correlation was found between the measured and predicted PA (steps/day) in samples (P<0.001). Conclusion: This study show that BMI, BMR, TDEE and VO2rest are useful variables in predicting daily PA volume. It seems that regression linear equation can be applied at clinical and epidemiological studies