Annually, the Behavioral Risk Factor Surveillance System (BRFSS) survey is administered by the Centers for Disease Control and Prevention (CDC). This article uses 2016 SMART BRFSS data to predict the likelihood a person will get a health checkup and it identifies which factor(s) influence the decision to obtain a checkup. Patterns of individual decision making were analyzed using various supervised data mining techniques. The best predictive model, with a predictive accuracy of 80%, can improve future BRFSS surveys by better understanding the responses and provide insight into the factors affecting decisions. The model was scored on new data to verify its accuracy. These findings supplement ongoing research to identify how behavior leads to better decision making related to medical checkups. The model can help identify poor decision-makers in high-risk groups. This research can also be used by healthcare professionals to improve clinical prevention services. Potentially, the research can be extended by combining the BRFSS data with ICD-10 and CPT codes. Better knowledge of diagnosis (ICD-10) and the cost associated with diagnosis (CPT) will help to understand a person’s health behavior. In the United States, expenditures on healthcare are rising significantly every year. Health decisions of individuals determine the overall health of a nation. Therefore, the U.S. Government should initiate health programs that encourage individuals to make better health decisions.
Singh, Neetu; Kanthwal, Apoorva; Bidhuri, Prashant; and Vijaykumar Munnolli, Anusha
"Using Individual Decision, Economic, and Health Status Data to Predict Health Checkup Behavior,"
Journal of the Midwest Association for Information Systems (JMWAIS): Vol. 2020:
1, Article 3.
Available at: https://aisel.aisnet.org/jmwais/vol2020/iss1/3