Disclosure: Consultant: Michigan Stroke Registry and Quality Improvement Program.
Pub Date: Tuesday, March 9, 2010
Author: Mathew J. Reeves, PhD
CitationAmbardekar AV, Fonarow GC, Dai D, Peterson ED, Hernandez AF, Cannon CP, Krantz MJ., Quality of care and in-hospital outcomes in patients with coronary heart disease in rural and urban hospitals (from Get With the Guidelines-Coronary Artery Disease Program)., The American Journal of Cardiology, 105 (2) 139-43. View in PubMed
Article TextThis edition of Get With the Guidelines Science News highlights another set of publications that illustrate the extraordinary breadth and high value of the information being generated from GWTG. Topics include the clinical effectiveness of implantable defibrillators in Medicare patients with heart failure , regional differences in the quality of care and outcomes in patients with acute coronary syndrome , gender differences in the care of coronary artery disease (CAD) patients , and inhospital outcomes in heart failure patients who are nonadherent to medications or dietary recommendations. For this editorial, I have chosen to focus on one publication that examined rural-urban differences in quality of care and outcomes in the GWTG-CAD program. The topic of rural versus urban differences in the quality of care delivered by hospitals has obvious public health significance and offers the potential for both significant opportunities and challenges for GWTG programs. The paper authored by Ambardekar and colleagues, compared inhospital outcomes, length of stay, and quality of care between 71 rural hospitals and 477 urban hospitals using data from more than 350,000 coronary heart disease patients enrolled in GWTG-CAD between 2000 and 2008.
In unadjusted analyses, the study found moderately higher inhospital mortality in rural hospitals (5.7% vs. 4.4%) and slightly lower conformity with 6 CAD performance measures (PM); the defect-free composite score was 74.7% in rural hospitals compared with 90.5% in urban hospitals. However, this study also illustrated that rural patients differed from urban patients in several important ways. Rural CAD patients were slightly older (67 vs. 66 years old), were more likely to be white (77% vs. 66%), were more likely to have health insurance (especially Medicare), and were more likely to have atrial fibrillation and heart failure. Interestingly, rural patients were less likely to have a past medical history of dyslipidemia. These factors ended up being important because, after the data were adjusted using statistical models, the significant differences previously observed in inhospital mortality and PM conformity went away. So the first learning point from this study is that adjustment for patient-level factors matters when examining differences between rural and urban hospitals.
It is self-evident to state that rural hospitals differ substantially from urban hospitals in terms of their structure and resources. Rural hospitals are much smaller, are rarely teaching hospitals, and have fewer personnel and infrastructure resources. In the study by Ambardekar and colleagues, adjustment for these hospital differences had little impact on their findings once patient-level data had been accounted for. The fact that adjustment for hospital characteristics made little difference might seem a little surprising because we know that hospital characteristics, such as bed size and teaching status, are known to be associated with patient outcomes and quality of care. However, I suspect that this finding probably reflects the fact that these characteristics do not vary much across rural hospitals, which limits the impact of statistical adjustment. Moreover, it could be argued that analyses examining differences between rural and urban hospitals should not adjust for the structural differences between the hospitals anyway. So a second learning point is that comparisons between rural and urban hospitals should probably not adjust for the obvious structural differences between the rural and urban hospitals. A study examining the impact of hospital characteristics should probably be performed just within the rural hospitals, rather than by comparing them with urban hospitals.
The finding that the rural hospitals in GWTG-CAD provided similar care and had similar outcomes as urban hospitals is obviously the study's most newsworthy finding. As pointed out by the authors, this finding runs counter to most of the previous studies, and can be interpreted as another example of how participation in GWTG can reduce or eliminate differences and disparities in care. Of course, one must also consider the fact that the rural hospitals that chose to participate in GWTG-CAD are fundamentally different from other rural hospitals that do not participate, and it may be these unmeasured factors (i.e., confounders) that may be responsible for the better performance in care and outcomes, rather than participation in GWTG per se. I suspect that both factors, participation in GWTG and the hospital's inherent commitment to quality improvement (QI), are contributing to the positive findings.
Of the approximately 5,000 community hospitals in the US, approximately 40% are defined as rural according to the American Hospital Association 2008 survey. The fact that only 13% of the hospitals in the GWTG-CAD data used in the Ambardekar study were from rural hospitals indicates that, not surprisingly, rural hospitals are under-represented in this GWTG program. This fact illustrates both the opportunity and challenges that rural hospitals bring to GWTG programs. Most studies show that smaller rural hospitals struggle to provide equivalent care as urban hospitals, and so the ability of QI programs such as GWTG to improve quality would seem like a perfect opportunity to improve the lives for the one-fourth of Americans who live in rural areas. However, herein "lies the rub" with rural hospitals that lack the resources to participate in programs such as GWTG. Because of their small size, limited resources, and wide geographic distribution, rural hospitals are difficult to enroll in QI initiatives. It has been our experience in enrolling rural hospitals into stroke registries in Michigan that participation is substantially lower compared with larger urban hospitals. The most frequent reason given for not participating is the lack of available staff and the distance required to travel to collaborative trainings. Even if rural hospitals do participate, their smaller case volumes mean that quality of care and outcomes data will be measured more imprecisely, and so it will take a longer time to accumulate enough data to be able to demonstrate the impact of the program.
As GWTG continues to expand, the next frontier has to involve smaller rural hospitals where substantial opportunities for QI undoubtedly lie. GWTG should consider strategies to increase participation of rural hospitals; perhaps data can be generated to demonstrate the amount of resources that a hospital needs to commit to participate in these programs. Perhaps participation is more "doable" than many of these rural hospitals initially realize. Of course, modern communication technologies are helping to remove the barriers that geographic isolation has dictated for so long.
Rural hospitals certainly create challenges in terms of participation in QI efforts such as GWTG, but they almost certainly represent the next great frontier on the battle to improve quality of care and outcomes for all Americans. By studying the care delivered by rural hospitals that the paper by Ambardekar has highlighted, these data present an issue that too often is ignored because of the substantial practical challenges. It is time for a renewed effort to extend the benefits of GWTG into rural areas.
- Hernandez A, Fonarow G, Hammill B, et al. Clinical effectiveness of implantable cardioverter-defibrillators among medicare beneficiaries hospitalized with heart failure. Circulation: HF (online) Circ Heart Fail 2010;3:7-13.
- Laskey W, Spence N, Zhao X, et al. Regional differences in quality of care and outcomes for the treatment of acute coronary syndromes: an analysis from the American Heart Association's Get With the Guidelines Coronary Artery Disease Program. Crit Pathw Cardiol 2010;9(1):1-7.
- Lewis WR, Ellrodt AG, Peterson E, et al. Trends in the use of evidence based treatments for coronary artery disease among women and the elderly: findings from the get with the guidelines quality-improvement program. Circ Cardiovasc Qual Outcomes 2009;2:633-641.
- Ambardekar A, Fonarow G, Dai D, et al. Quality of care and in-hospital outcomes in patients with coronary heart disease in rural and urban hospitals (from the Get With the Guidelines-Coronary Artery Disease Program). Am J Cardiol 2010;105:139-143.
-- The opinions expressed in this commentary are not necessarily those of the editors or of the American Heart Association