From clinical trials to clinical practice: bridging the GAP.

Abstract

EVIDENCE-BASED MEDICINE. THE CONCEPT SEEMS SO simple. Take the best available information from clinical trials and observational studies and apply the results in clinical practice. What in medicine could be more rational or straightforward? Yet despite a rich evidence base for management of cardiovascular disorders, study after study has demonstrated disconcertingly low rates of compliance with widely disseminated evidence-based treatment guidelines for managing common cardiovascular conditions including coronary heart disease, heart failure, and hypertension. The manifest difficulty in translating clinical trials into clinical practice suggests the presence of multiple barriers to implementation. Although there is substantial overlap, these barriers fall into 3 general domains: physicianrelated, patient-related, and health system–related. Physicianrelated barriers include lack of knowledge of the “best” current evidence, which is not surprising given the plethora of studies that have been completed, with new studies reported every week, as well as time constraints and the overriding desire to avoid iatrogenic complications. Patientrelated barriers include polypharmacy, time and financial constraints, and difficulties engaging in health-modifying behaviors such as smoking cessation, exercise participation, and dietary restriction. Health system–related barriers include the high number of uninsured and underinsured individuals, the lack of systematic approaches to the care of chronic illness, and practical concerns about the high cost of health care, including the reality that few interventions actually reduce costs. The complexity of issues involved mandates a comprehensive and collaborative approach involving physicians and other health care professionals, patients and their families or other support systems, and the health care system itself, if the myriad barriers to implementing evidence-based care are to be overcome successfully. In this issue of THE JOURNAL, Mehta and colleagues describe just such an approach to improving care for acute myocardial infarction (AMI)—the Guidelines Applied in Practice (GAP) initiative in Southeast Michigan. The GAP program was developed in collaboration with the American College of Cardiology, the Center for Medicare and Medicaid Services (CMS), and the Michigan Peer Review Organization (MPRO). To facilitate adherence to established treatment guidelines for AMI, the GAP investigators worked closely with a consortium of 10 local hospitals that were selected from 22 hospitals that volunteered to participate in the project. Selection of participating hospitals was not performed by randomization, but did allow for study of a diverse group of institutional cultures. Each participating hospital had a local physician “champion” and the program was publicized through “grand rounds” lectures and other materials. In addition, each GAP facility was provided with the GAP “tool kit,” which comprised educational resources for staff and patients, a clinical pathway, standardized admission and discharge order forms, and reminders or “ticklers” (eg, chart stickers) to promote guideline adherence. To evaluate the effectiveness of the intervention, adherence to 11 quality indicators for AMI care was measured before and after the intervention at participating hospitals, as well as at 11 hospitals that volunteered for the program but were not selected. The principal findings of the study were that at participating hospitals, compliance with 9 of the 11 quality indicators improved following implementation of the GAP program, with absolute gains ranging from 4% to 12%, and with 4 of the improvements achieving statistical significance. However, favorable changes also occurred in several of the quality indicators at non-GAP hospitals, such that the only significant difference in change over time was in prescription of aspirin at hospital discharge, which improved to a greater extent at GAP hospitals than at nonGAP hospitals. Conversely, improvements in the use of -blockers within 24 hours of admission and at discharge tended to be greater at non-GAP hospitals (16.6% vs 10.8% and 16.1% vs 5.6%, respectively), although these differences were not statistically significant. Notably, improvements were consistently greater when there was evidence for use of the GAP tools, and also tended to be greater among subgroups that typically exhibit greater disparities in guideline adherence, eg, racial minorities and elderly persons.