|Home | About | Journals | Submit | Contact Us | Français|
The following are personal and candid reflections concerning the selection, implementation, and administration of an electronic health record (EHR). My perspective is primarily that of a member of a leadership and oversight group for information technology at an academic health center. Although I am director of the Dartmouth College's Biomedical Libraries, my primary role in this group is not at all that of a library administrator, but rather of a general institutional leader of information technology.
Dartmouth-Hitchcock Health (DHH) <http://www.dartmouth-hitchcock.org> is the dominant health care system in northern New England. The flagship facility is the Dartmouth-Hitchcock Medical Center in Lebanon, New Hampshire. There, I am a member of the DHH Information Systems Steering Committee (ISSC), a group reporting to the board of governors. ISSC develops and nurtures DHH's vision and strategy for information technology to further our mission to “advance health through research, education, clinical practice and community partnerships, providing each person the best care, in the right place, at the right time, every time.” One of the principal tasks of the ISSC initially was to oversee the selection, implementation, and administration of a marketplace EHR solution to replace Clinical Information System (CIS), our homegrown system. Our timeline, at its broadest, is:
My reflections here are not from my role as director of the Dartmouth Biomedical Libraries but as a member of the ISSC. (I also chair ISSC's messaging, collaboration, and information technology support subcommittee, which oversees the selection, implementation, and administration of email, messaging, calendaring, and groupware solutions for DHH.)
My first thesis concerning libraries and EHRs is that the librarian community may have a bucolic, rosy, and—perhaps—naive view of how our clinicians use library resources and services in the context of an EHR. Clinicians, in their patient encounters, take histories, do physical exams, examine lab test results and imaging, talk with the patient about possible diagnoses and treatment plans, and develop treatment and care plans, including the “order set” detailing how the patient will be treated. The health care “system” does not permit the time to do—or even any encouragement to do—full-blown, comprehensive “literature searches” or otherwise locate best evidence on which to base treatment and care. This, again, is my personal observation: that clinicians do not, during the actual patient encounter, search for “evidence” versus rely on whatever instruction that may be baked into the EHR. Patient encounters are not “learnable moments” for clinicians to apply new evidence to treatment and care.
My second thesis is based on my first. The question is not, “How do we integrate [name your library information resource] in the EHR?” Sure, we could embed some kind of “library toolbar” or “information button” in the EHR interface or even make selected words in patient records hyperlink to [name your library information resource]. But I venture these would not be generally successful, because these information-seeking behaviors would not be a seamless part of the treatment-and-care process. Clinicians just want the order set. The information clinicians want or are able to use has to be seamlessly integrated into the process of providing treatment and care. I think the question is, rather, “How can evidence be logically integrated into the EHR so it is part of the treatment-and-care process?”
EHRs are very much about increasing clinical productivity, about advancing the business of health care. The vision of a successful EHR does not include that clinicians get bombarded with every available piece of evidence—every available piece of information—potentially relevant to the need at hand. Rather, the vision is that clinicians are given just the information they need to advance health, to provide each person the best care, at the right time, every time. For example, the core goals of DHH's new EHR implementation are:
At every level, each clinician and staff should be supported and workflows designed in such a manner as to optimize a) quality of care, b) patient satisfaction, c) clinician and staff productivity, d) appropriate charge capture, e) clinician and staff job satisfaction and f) advancement of our academic mission. [Internal presentation to board]
Librarians should ask ourselves how we can contribute our knowledge, skills, and experience to an EHR implementation, not how the “library should be integrated in the EHR.” We should understand that just inserting a “library button” in the EHR probably is not going to provide fundamental, systematic benefit. Even if clinicians push the button—which they very well may not—if it does anything other than provide information exactly integrated to the treatment and care process at hand, it will be of little use. Information—some of which, but not all, may be the responsibility of the library to manage—has to be an integral part of the real process of providing care.
When you talk with your clinical leaders about EHRs, talk about how evidence—not “the library”—should be integrated into the treatment and care process.
*Based on a paper presented at the Association of Academic Health Sciences Libraries symposium: “Electronic Health Records (EHR) and Knowledge-Based Information: State-of-the-Art and Roles for Libraries in Health Information Technology”; Boston, MA; November 7, 2009.