In modeling human behavior and social structures several factors can emerge over time this can be attributed to the availability of new data, increased complexity, changes to the organizational structure, interventions, introduction of innovative technology or services and due to improved knowledge and treatments. We hypothesize a new class of emergent decision support systems that continually evolve to account for this "Causal Drift". In this work, we illustrate the application of the Emergent Approach to Systems and Intervention (EASI™) methodology with the example of Community Intervention Activity Model (CIAM) to reduce the rate of diabetic hospitalization at the local/ county level. A key contribution of this work is the design of an efficient theoretically informed emergent data collection system. A second key contribution of this work is that it offers practitioners a methodology to systematically determine data that needs to be collected and then model the collected data. Thus EASI™ methodology supports the efficient capture of data that has utility in decision making. To ensure applicability of this work publicly available Behavioral Risk Factor Surveillance System (BRFSS) and Social Vulnerability Index (SVI) data sets have been utilized. The EASI™ method has four significant advantages: a) the prediction is based on theoretically informed causal structure; this allows it to be used as a basis for evaluation of interventions as opposed to deep learning and other machine-based structure learning methods which are susceptible to spurious associations, b) existing data is utilized to evaluate clinical relevance of predictions, c) leveraging high dimensional synthetic observational health data to model health objectives, and d) provides guidance on transformation of system from the emergent basis to the new emergent system as new knowledge is gained. The dissertation proposes, implements, and evaluates the EASI™ methodology as applied to a CAIM for the reduction in diabetic hospitalizations.
| Identifer | oai:union.ndltd.org:ucf.edu/oai:stars.library.ucf.edu:etd2020-2716 |
| Date | 15 December 2022 |
| Creators | Renduchintala, Chaithanya |
| Publisher | STARS |
| Source Sets | University of Central Florida |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | Electronic Theses and Dissertations, 2020- |
Page generated in 0.0019 seconds