This study seeks to expand the medical application of Geographic Information Systems (GIS) by bringing this technology into the primary care setting, where it can directly support pediatricians in 'real-time'as they counsel patients about achieving health behavior changes. We propose to do this by coupling GIS with a computer based decision support system (CDSS) previously developed by our group. This combination of technology will enable the pediatrician to direct patients to resources in their community that can help address their health needs. Because these resources will be selected on the basis of being near the patient's residential address, they will have the benefit of being more accessible, trusted and culturally relevant than agencies that are not local. We propose to demonstrate the abilities of this new system by pilot testing three types of referrals: 1) pediatric dental referrals for caries prevention, 2) physical activity programs for obesity prevention, and 3) academic support services to prevent school failure. We will also use this same technological approach to enhance a pediatrician's capability to screen children for environmental factors that represent health risks. In this case, we will evaluate the usefulness of GIS to improve lead screening. Our GIS application will process geographic data such as surveillance results and age-of-housing-stock that is specific to a patient's neighborhood. Such comprehensive information, because it encompasses entire counties, would be difficult for physicians to utilize in screening decisions. To our knowledge, no one has studied using GIS to identify high-risk children in need of lead screening, at the point of care. This proposal is innovative in that it proposes a way for GIS to be intelligently inserted into the health care delivery process. By utilizing spatial data, clinical information stored in an electronic medical record, and data captured directly from the patient/parent at the time of a clinic visit, our system can generate highly tailored follow-up questions and recommendations for the physician as well as personalized educational materials for parents/patients.
The specific aims for this study are to: (1) Expand and modify an existing computer-based decision support system (CHICA), to include a geographic information system (GIS);(2) Demonstrate both the feasibility and effectiveness of CHICA-GIS in a) altering physician behavior in regards to referring patients to community resources and b) altering patient (child/family) behavior in regards to utilizing community resources to which they are referred;and (3) Demonstrate both the feasibility and effectiveness of using CHICA-GIS to identify children for targeted environmental screening based on patient specific data.
Aims 2 and 3 of this study will include a randomized trial conducted in six pediatric clinics. We hypothesize that the coupling of CDSS with GIS will allow pediatricians to more easily and accurately refer children to community resources near their home, and in turn will increase the rate at which patients access these resources. We also hypothesize that CHICA-GIS will allow pediatricians to better screen children for environmental hazards such as lead exposure.
This study couples geographic information systems (GIS) with computer based decision support, creating an electronic medical record that can directly support pediatricians in 'real-time'as they 1) counsel patients about achieving health behavior changes and 2) screen children for environmental factors that represent health risks. This innovative application of GIS will help physicians become aware of neighborhood characteristics that constrain behavior, and also inform patients about community-based resources that can support behavior change. The system can also improve screening for health risk by delivering geographically relevant information to physicians, such as spatial clusters of disease or environmental pollution near a patient's home.
| Bauer, Nerissa S; Carroll, Aaron E; Saha, Chandan et al. (2016) Experience with decision support system and comfort with topic predict clinicians' responses to alerts and reminders. J Am Med Inform Assoc 23:e125-30 |
| Dugan, T M; Mukhopadhyay, S; Carroll, A et al. (2015) Machine Learning Techniques for Prediction of Early Childhood Obesity. Appl Clin Inform 6:506-20 |
| Pan, Gang; Wan, Mei Hua; Xie, Kun-Lin et al. (2015) Classification and Management of Pancreatic Pseudocysts. Medicine (Baltimore) 94:e960 |
| Whipple, Elizabeth C; Odell, Jere D; Ralston, Rick K et al. (2013) When Informationists Get Involved: the CHICA-GIS Project. J Escience Librariansh 2: |
