Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website
FORUM - Translating research into quality health care for Veterans

» Back to Table of Contents


Getting the Most Out of an EHR System Requires Research and Development

Key Points


  • Lessons from today’s EHR system can inform uses of tomorrow’s.
  • Some lessons from local EHR implemen­tations are also relevant to distant ones.
  • Decision-making, usability, interoper­ability, surveillance, risk prediction, population management, telehealth, patient-generated data, and support of patients’ preferences are a few key areas of needed research and development.
  • The Cerner EHR system contains research-oriented tools, such as alerts about research participants and their visits.
  • Many study designs targeting EHR systems will beneft from thoughtful selection of comparison groups.
  • The process of technical development should not slow the planning of EHR research.
  • The Offce of EHR Modernization may facilitate study of the EHR.

Since the electronic health record (EHR) has become such a powerful element in the process of healthcare delivery, many health services researchers have undertaken projects directly examining its roles, effects, and potential to improve care—“the HSR of EHR.” Beyond VistA and the Computerized Patient Record System, VA has seen decades of companion projects, additional developments, and iterations of the EHR system itself: VistA Evolution, VistA Web, the Enterprise Health Management Platform, the Joint Legacy Viewer, and others. Typically, skilled VA-employed developers have undertaken these projects in-house, at times working with selected contractors. As VA “begins” a transition to a commercial EHR system through its contract with Cerner Corporation—slated to cost at least $10 billion over the next ten years—questions arise about VA’s research and development in informatics: why, what, how, where, and when? This article provides suggestions for some of the answers to these questions.

A fundamental question is why we should study the EHR system, whichever one it is. Won’t everything change next year (month, week), rendering today’s studies unhelpful? Everything will change next year, because that is the natural course of technology and always has been, inside and outside VA. Therefore, we keep moving: study what we have and where we think the technology is headed next. As with other aspects of healthcare, which also change every year, we make our best educated guesses, learn from the practices, processes, and policies in effect, and design and test innovations that hold promise for improvement. Veterans and VA professionals need improved EHR structures, processes, and outcomes today, and we can help with that.

Won’t our studies rooted in a (select: commercial, public, old, new, special, large) EHR system fail to be generalizable to other sites and systems? Not necessarily. The question of generalizability comes up in almost all of the grant proposals I have seen that include an aspect of the EHR. The naïve critic offers the comment, “they are studying just one system, so it’s not generalizable.” Although the scientists might study one system or more than one, a basic concept still holds: all systems are unique. Even with a single commercial EHR product, every site has a unique instance and implementation of that product. This results in unique templates (even at the user level), features, and functions. Should we then throw up our hands in defeat? Not in the least. Our goal is to study the common ground or even the special and potentially benefcial features, identifying which approaches work, why they work, and how to change current practices to improve outcomes. You can perhaps see the parallels between a unique EHR environment and a unique healthcare environment: a unique Veteran seeking care in a unique medical center, with unique healthcare professionals. Such systems—human, clinical, organizational, and technical—can, should, and must be studied.

The topic of what to study is too large to capture in this brief article and will be covered elsewhere. Many of the essential elements of biomedical informatics—both clinical and more basic aspects—need study in  the VA environment. These include how the EHR can most effectively facilitate clinical management, shared decision making, effciency, and safety through improved usability, interoperability to facilitate cross-institutional and community-based care, automation of surveillance and risk prediction, population management, telehealth, capture of patient-generated data, and the support of Veterans’ preferences for care. There are many more topics! No one reading this article could run out of important ideas to pursue in an entire career.

Implementation of EHR innovations has always required numerous steps in VA and elsewhere. We need approvals—sometimes local, regional, and national—and people who can do the work, not to mention training, monitoring, safety assessments, etc. We will still need those basic elements in a Cerner or other commercial EHR environment. Cerner’s presence in VA might make aspects of technical development more challenging, since VA can’t get “inside the box” easily or quickly (but were we doing that anyway?). Our hope is that VA will help the HSR&D community work with Cerner to extract data for detailed retrospective studies and to design, implement, and test aspects of the EHR system prospectively. Fortunately, Cerner’s system actually contains certain tools to facilitate research. In addition, modular and interoperable applications, such as through technologies like Fast Healthcare Interoperability Resources, hold promise for development and study in conjunction with modern EHR systems.

Where are the best places to study the EHR? Everywhere. We need studies of current technologies as well as planned or imagined ones. We need studies in small and large systems, urban and rural ones, north and south ones, more and less electronic ones. For obvious reasons relating to our confdence in causality and the design of improvements, having comparison groups will be a key element of many of the best studies.

You know, they are just about to start something new at your facility, right? Should you wait for Cerner or (xyz) before getting started? Um, of course not—do not wait—unless you are planning for a very active retirement. The Cerner implementation process is unfolding in stages, starting now, and extending for approximately one decade. My basic approach to engaging in technology is do not wait for technology, because it changes constantly. Get started now, learn what we can learn, and do what we can do. “Won’t the politics of the Cerner implementation kind of get in my way?” If you receive a paycheck from any organization, then you are involved in a political system, so get over it and keep moving. I’m not saying that we won’t face roadblocks, but we will not overcome any of them without trying. As always, we will uphold our values in improving care for Veterans.

VA now has an Offce of EHR Modernization (OEHRM). It is working with the Department of Defense and includes a Governance Integration Board, EHR Councils, and a Legacy EHR Modernization “Pivot Working Group.” OEHRM’s principles include standardizing clinical and business processes across VA, designing a Veteran-centric system focusing on quality and safety, pursuing a fexible and open solution, accommodating scalability, and reengineering clinical business processes. The roadmap for platforms, solutions, and services includes research tools such as Cerner’s Health Facts and PowerTrials (thanks, Jim Breeling, for many of these details). Health Facts is a HIPAA-compliant electronic database with rich clinical EHR data available in de-identifed form, representing the care of nearly 50 million patients in at least 90 health systems that contribute to the database. PowerTrials provides EHR indicator fags of patients’ participation in clinical research, shows protocol and study contact information, and notifes the study’s contact when a research subject is scheduled for a clinical visit. It will also ultimately have features designed to enhance recruitment using screening tools based on a study’s inclusion and exclusion criteria. VA’s governance group will facilitate VA scientists’ understanding of EHR capabilities, and engagement in research about VA’s EHR system.

So remember, when it comes to the HSR of EHR, the time is now, and the person is you. A tip for grant writers: many reviewers will not read this article, so help them when needed, through reminders about any of the key points.

Next