Sunday June 30, 2019 from 07:30 to 09:00
Does the learning healthcare system work: A systematic review
Raees Cassim1, Viviane Grandpierre1, Beth Potter2, Pranesh Chakraborty3,4, Melise Keays1.
1Division of Urology, Department of Surgery, Children's Hospital of Eastern Ontario, Ottawa, ON, Canada; 2School of Epidemiology and Public Health, University of Ottawa, Ottawa, ON, Canada; 3Newborn Screening Ontario, Children's Hospital of Eastern Ontario, Ottawa, ON, Canada; 4Department of Pediatrics, Children's Hospital of Eastern Ontario, Ottawa, ON, Canada
Introduction: The Institute of Medicine describes a learning healthcare system (LHS) as the alignment of science and informatics with the goal of improving care and generating new knowledge. With the increased uptake of electronic medical records (EMR) at healthcare centres, we hypothesized that LHS have been developed and are improving healthcare and research outcomes. Our objective was to determine if existing LHCS are effective in increasing research efficiency, provider efficiency/satisfaction, cost-effectiveness, and quality improvement.
Methods: We searched MEDLINE, Cochrane CENTRAL Trials Registry, CINAHIL, and EMBASE for articles related to LHS. We included articles describing systems that automatically collected/analyzed data. Articles had to describe effects on quality improvement, research efficiency, provider efficiency/satisfaction, or cost-effectiveness. We screened abstracts using the liberal accelerated method (i.e., two reviewers to exclude, one reviewer to move to full text). After double reviewer screening of full text, we extracted data from each article, including outcome measures, EMR used, disease area, and publishing institution.
Results: We screened 1258 abstracts, 257 of which moved to full text screen. Eighteen articles met our inclusion criteria. Of the included articles, 12 reported on quality improvement, six on research efficiency, five on provider efficiency/satisfaction, and one on cost-effectiveness. The majority of the articles described improvements on the aforementioned outcomes, with no articles describing any detrimental effect. Only three articles specified the EMR used (EPIC, GE Centricity). Most LHS focused on a specific disease area, with oncology being the most common. Most of the articles were published in the U.S. with some published in Europe and Asia.
Conclusions: Current publications suggest successful implementation of several LHS in the world. These systems have been described as effective at improving outcomes in specific disease areas.