Date of Award

Fall 1990

Project Type


Program or Major


Degree Name

Doctor of Philosophy

First Advisor

William Baum


Researchers have been unable to discern the relationship between energy expenditure and obesity because there has been no reliable, unobtrusive method to capture daily activity. Single-plane accelerometers have been developed, using microcomputer technology, which may provide a means for measuring expenditure in human subjects. The Caltrac Personal Activity Computer$\sp\circler$ has shown high test-retest reliability and a strong relationship to several measures of expenditure in some laboratory tests. In Experiment 1, 40 university women (mean age = 19.15 yr) wore Caltracs during waking hours for either three or six consecutive days, with the devices programmed to count activity regardless of individual metabolism. Pearson product-moment correlation coefficients showed a moderate negative relationship between Caltrac counts and Body Mass Index (BMI) (r = $-$.47, $p<$.01) for the second three-day period (N = 26) but there was no correlation between Caltrac counts for the first three days and any measure of body weight or fat. Subjects may need a few days to become accustomed to the device before counts are analyzed when it is used in natural, free-ranging settings.

There was no correlation between estimates of weekly caloric expenditure determined from the Paffenbarger Survey of Physical Activity and measures of body weight and fat or the Caltrac. There was a moderate positive correlation between skinfold measures and the total amount of weight lost and gained over the past year (r =.46, $p<$.001), indicating that fatter subjects had more labile weight.

In Experiment 2, 15 women (mean age = 19.4 yr) who were at least 30 percent fat, with weight ranging from 139-234 lb (mean weight = 178.37 lb) wore Caltracs for four weeks which were programmed to reflect individual expenditure in calories. Subjects recorded caloric intake into the Caltrac which displayed energy balance in the form of net calories (intake minus output). Mean weight loss was $<$1 lb, not significantly different from control subjects (N = 5) who weighed daily and graphed their weight. The Caltrac may be an improved method for estimating energy expenditure but it did not help subjects lose weight.