AN AUTOMATED IDENTIFICATION OF INDIVIDUALS AT HEALTH RISK BASED ON DEMOGRAPHIC CHARACTERISTICS AND SELF-REPORTED PERCEPTIONS

Dejan Magoc, Department of Health Studies, Eastern Illinois University, Tanja Magoc, Ph.D., Center for Bioinformatics and Computational Biology, University of Maryland, Joe Tomaka, Ph.D., Department of Public Health, University of Texas at El Paso


Abstract

The risks of developing diabetes, high blood pressure, and cardiovascular disease could
be reduced by increasing the number of individuals receiving adequate levels of physical activity
(PA). Centers for Diseases Control and Prevention (CDC) has reported that about 30% of
Americans do not engage in any PA and about 40% engage in some levels of PA, but still not
meeting the recommended levels defined by the American College of Sports Medicine (ACSM).
Studies have shown that the greatest declines in PA occur during the transitions from high school
to college and beyond. Thus, it is important to identify students at young age that are at health
risk due to lack of PA, so that specific steps could be taken toward helping these individuals
develop a healthier lifestyle. We used data on 100 college students to develop a preliminary
computer program (using a backpropagation multilayer neural network approach) to
automatically identify individuals at risk of being not sufficiently physically active. Besides
various types of demographic variables, data included information on the association between
students’ self-reported levels of PA and Social Cognitive Theory (SCT) constructs (e.g., selfefficacy,
self-regulation, social support, expectations), as predictors of participation in PA. The
results of this study indicated that the backpropagation multilayer neural network identified and
classified individuals at risk of being not sufficiently physically active into right categories (atrisk
individuals or not at-risk individuals) 77% of the time. Collecting additional data points that
contain more at-risk individuals will improve the neural network’s prediction of at-risk
individuals.

Keywords: automated identification, physical activity, college students

 



References
Booth, M. L. (2000). Assessment of physical activity: An international perspective. Research
Quarterly for Exercise and Sport, 71(2), s114-120.
CDC. (2005). Trends in leisure-time physical inactivity by age, sex, and race/ethnicity: United
States, 1994-2004. Morbidity and Mortality Weekly Report, 54(39), 991-994.
Dunn, M. S., & Wang, M. Q. (2003). Effects of physical activity on substance use among college
students. American Journal of Health Studies, 18(2/3), 126-132.
Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., & Witten, I. H. (2009). The
WEKA data mining software: an update. SIGKDD Explorations, 11(1).
Haskel, W., Lee, I. M., Pate, R. R., Powell, K. E., Blair, S. N., Franklin, B. A., Macera, C. A.,
Heath, G. W., Thompson, P. D., & Bauman, A. (2007). Physical activity and public
health: Updated recommendations for adults from the American College of Sports
Medicine and the American Heart Association. Journal of American Heart Association,
116, 1081-1093.
Magoc, D., Tomaka, J., & Thompson, S. (2010). Overweight, obesity and strong attitudes:
Predicting participation in physical activity in a predominantly Hispanic college
population. Health Education Journal, 69(4), 427-438.
Automated identification of individuals at health risk
Mounton, C. P., Calmbach, W. L., and Dhanda, R. (2000). Barriers and benefits to leisure-time
physical activity among older Mexican Americans. Archives of Family Medicine, 9, 892-
897.
Petosa, R. L., Suminski, R. R., & Hortz, B. (2003). Predicting vigorous physical activity using
social congnitive theory. American Journal of Health Behavior, 27(4), 301-310.
Pratt, M., Macera, C. A., & Blanton, C. (1999). Levels of physical activity and inactivity in
children and adults in the United States: Current evidence and research issues. Medicine
and Science in Sports and Exercise, 31(Nov), S526-533.
Rovniak, L. S., Eileen, S. A., & Winett, R. A. (2002). Social cognitive determinants of physical
activity in young adults: A prospective structural equation analysis. Annals of Behavioral
Medicine, 24(2), 149-156.
Rusell, S. & Norvig, P. (2010). Artificial intelligence, a modern approach. Prentice Hall.
Sallis, J. F., Grossman, R. M., Pinski, R. B., Patterson, T. L., & Nader, P. R. (1987). The
development of scales to measure social support for diet and exercise behaviors.
Preventive Medicine, 16, 825-836.
Sallis, J. F., Pinski, R. B., Grossman, R. M., Patterson, T. L., & Nader, P. R. (1988). The
development of self-efficacy scales for health-related diet and exercise behavior. Health
Education Research, 3, 283-292.
Tan, P., Steinbach, M., & Kumar, V. (2006). Introduction to data mining. Addison Wesley.
Submitted 22 April, 2011
Accepted 15 June, 2011