THE UNIVERSITY OF BRITISH COLUMBIA
Research Specific Projects
IMPLEMENTING PREDICTIVE ANALYTICS TOWARDS EFFICIENT COPD TREATMENTS (IMPACT)
ABOUT:
With a phased implementation of a harmonized electronic health records (EHR) system across all tertiary care centres, British Columbia (BC) is in the midst of a transformational change in health care delivery. Our 23-member research team comprised of patients, care providers, knowledge users, administrators, researchers, and implementation sciences and knowledge translation experts, is ideally positioned to lead this unique IMPACT (IMplementing Predictive Analytics towards efficient COPD Treatments) study. Our objective is to implement and evaluate Precision Medicine tools for the care of patients with Chronic Obstructive Pulmonary Disease (COPD). Our proposed intervention is a Clinical Prediction Model (CPM), integrated within EHR and powered by advanced Natural Language Processing to provide a user-friendly human-computer interface. It can assess the risk of COPD exacerbations, communicate risk, and recommend personalized pharmacotherapies and lifestyle modifications via shared decision making. As part of our integrated knowledge translation (iKT) approach, we will engage patients and care providers as research partners to co-design and integrate this tool into clinical workflow. We will implement the finalized tool in two of the largest teaching hospitals in BC, and will conduct a pragmatic stepped-wedge cluster randomized controlled trial (RCT) to evaluate its impact on: prescription patterns, medication adherence, patient-reported outcomes, exacerbation rates, and sex- and gender-disparities in COPD care. The multidimensional data from this study will inform a previously validated Canadian health economics model of COPD, allowing us to assess the cost-effectiveness of large-scale implementation of such an intervention. This project will create a lasting collaboration among an interdisciplinary team to support a continuous program of promoting predictive analytics in clinical decision-making. Through this, IMPACT will not only improve COPD care and outcomes, but can also serve as a template for other chronic diseases.
Team:
Principal Investigators:
Management team:
Co-investigators:
Collaborators:
Knowledge users:
PARTNERS:
PUBLICATIONS:
IMplementing Predictive Analytics towards efficient COPD Treatments (IMPACT): protocol for a stepped-wedge cluster randomized impact study.
Read more
Uncertainty and the value of information in risk prediction modeling
Read more
Value-of-Information Analysis for External Validation of Risk Prediction Models
Read more
Non-parametric inference on calibration of predicted risks
Read more
ACCEPT 2·0: Recalibrating and externally validating the Acute COPD exacerbation prediction tool (ACCEPT)
Read more
Generalizability of risk stratification algorithms for exacerbations in COPD
Read more
Moving beyond AUC: decision curve analysis for quantifying net benefit of risk prediction models
Read more
RELATED IPRESS WEBAPP:
MORE RESEARCH