EHR data and comparative effectiveness research

Comparative effectiveness research received a significant boost from Recovery Act (ARRA) money - $1.1 billion has been earmarked for this type of research. ARRA provides $300 million for the Agency for Healthcare Research and Quality (AHRQ), $400 million for the National Institutes of Health (NIH), and $400 million for the Office of the Secretary of Health and Human Services (HHS). These funds are to support studies assessing the comparative effectiveness of health care treatment and strategies, through efforts that (1) conduct research on clinical outcomes, effectiveness and appropriateness of treatments for various disease and conditions, and (2) encourage the development and use of clinical registries, data networks and other forms of electronic data to generate or obtain outcomes data.

Why the urgency? Health care resource utilization and clinical quality measures are far from standardized, with great variability across the country. The Dartmouth Atlas is a resource for looking at these patterns of variation, and has demonstrated wide variations in health spending from region to region. In a recent article in the New England Journal of Medicine, Jason Sutherland, Elliot Fisher and Jonathan Skinner showed that only about 30% of the regional variation in spending is attributable to variations in individual health and socioeconomic status. The remainder has more to do with inefficiencies in the health care delivery system – discretionary decisions by physicians that are influenced by the local availability of hospital beds, imaging centers and other resources, as well as by a payment system that rewards growth and higher utilization.

Driven by these challenges, the vision reflected in the ARRA legislation recognizes that many factors will need to be addressed in order to get a handle on health care spending. One of these factors is the development of a robust body of evidence-based data on “what is the best way” to treat a given condition – the conclusions of comparative effectiveness research. The AHRQ is one of the bodies that conducts such research, and is earmarked for funding by ARRA money. The AHRQ’s Effective Health Care Program conducts (1) research reviews, comparing completed scientific studies in order to make head-to-head comparisons of different health care interventions (and also show where more research is needed); (2) original research reports, which use healthcare databases and other resources to look at practical questions about effectiveness (benefits and harms) of various treatments; and (3) summary guidelines, as short, plain-language guides to be put into practice.

Clinical Decision Support – a desired feature of Electronic Health Records (EHRs) – is intended to be a mechanism whereby the kinds of recommendations that come out of comparative effectiveness research can be presented to rank-and-file physicians at the point of care. By doing so, some of the regional variations in health care currently seen can be reduced, and quality of care can be normalized. Of course, other factors – like changing compensation methodologies so as not to reward higher utilization, or developing a more robust primary care workforce in order to implement patient centered medical homes and accountable care organizations, or tort reform to reduce the pressure to practice “defensive medicine” – will also be needed if these efforts are to make a difference in the bottom line.

The AHRQ recently announced funding for $44 million in grants for what it calls the PROSPECT (Prospective Outcome Systems using Patient-specific Electronic data to Compare Tests and therapies) Studies. This effort is to establish an infrastructure and improve the methodology for prospective collection of clinical data contained in electronic databases. The projects intend to collect data from multiple patient settings – ambulatory practices, hospitals, and long-term care facilities – and address one of the topic areas identified by the Institute of Medicine (IOM), or one of the 14 priority conditions identified by the AHRQ, or fall into one of the AHRQ identified evidence gaps.

Clearly, the direction for collecting outcomes data will come from increasingly-connected clinical electronic databases. For developers of EHRs (ambulatory- and hospital-oriented), interconnectivity is key. No longer will self-contained “all inclusive” but separate systems be the “state of the art.” Systems that talk to each other – hospital systems, ambulatory systems, long-term care systems – are the things being encouraged by these kinds of AHRQ grants. Web-based ambulatory systems, like Practice Fusion, are already quite far down this road – clinical chart sharing between different practices is already being rolled out. Hospital interconnectivity is more challenging, mainly resulting from the heterogeneity of hospital IT systems, but connecting such systems to ambulatory ones will likely be one of the salient areas of growth in 2010.

There is still much work to be done. Our sleeves are rolled up, and we continue to find ways to “build the bridges” between disparate systems, so that the kind of vision for a transformed health care delivery system can be realized.


Robert Rowley, MD – Chief Medical Officer, Practice Fusion, Inc.