Integrative Informatics Approach to Predict Readmissions and Improve Outcomes in COPD

点击翻译按钮获取中文摘要

Project Summary This proposal for a mentored career development award consists of a training and research plan devised to facilitate Dr. David Jacobs' transition to an independent investigator focusing on the implementation of medication use interventions during the transition from hospital to home for high-needs patients. Dr. Jacobs is a clinical pharmacist with an advanced degree in epidemiology and health services and has significant experience conducting clinical research. The candidate's current research is focused on transition of care interventions. Given the resource intensive nature of these interventions, the sustainability of these programs will heavily depend on identifying and targeting patients at high-risk for hospital readmission. To anticipate such admissions, predictive models have been developed; however, our ability to predict hospital readmissions remains poor. This is likely because detailed social information, which is disproportionately represented in high-needs populations, is typically absent in risk stratification tools. Therefore, the long-term research objective is to build predictive models that combine social information with rich clinical data to individualize care management interventions and reduce readmissions. The specific aims are: (i) to determine social risk factors driving hospital readmissions by conducting semi-structured interviews with patients, their caregivers, and clinicians; (ii) to develop a risk- prediction model using advanced informatics techniques; and (iii) to develop and test the feasibility of a pilot intervention aimed at improving transition strategies. This work will focus on a singular complex condition, chronic obstructive pulmonary disease (COPD), in developing a risk stratification tool in order to improve its predictive performance for identifying high-risk patients. COPD will serve as the model condition since it is one of the major readmission diagnoses, and there exists a high level of complexity following hospital discharge at patients' care transition. This award will provide the applicant with mentor-guided didactic and experiential learning to address the following career development objectives: 1) gain experiential learning in qualitative methodology, 2) advance his knowledge in biomedical informatics and develop predictive models integrating high-dimensional electronic health record data, and 3) increase his understanding of implementation science and gain practical experience in conducting a pragmatic clinical trial. The primary mentor, Dr. Sanjay Sethi, and the mentorship team will work closely to monitor his progress toward independence and will provide him with the guidance and the resources to guarantee his success. The proposed study leverages the extensive resources available at the University at Buffalo to address an important public health issue. Achieving the proposed aims and acquiring these advanced skills will position the candidate to submit successful R01s testing the proposed clinical prediction model and transition strategy in real-world settings. In summary, a comprehensive career development plan in the context of a well-defined training, research, and mentorship structure will allow Dr. Jacobs to emerge as a highly successful, independent clinician-investigator in health services research.