| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
AMIA Position Paper |
Affiliations of the authors: Healthvision, Waltham, MA (JMT); Department of Emergency Medicine, Brigham and Women's Hospital and Harvard University, Boston, MA (JMT); Thomson Micromedex, Denver, CO (JAO); Department of Medicine, University of Pennsylvania Health System, Philadelphia, PA (JAO, EAP); Northwest Permanente, PC, Portland, OR (DFS); Department of Medical Informatics and Clinical Epidemiology, Oregon Health & Sciences University, Portland, OR (DFS); Department of Medicine, Cedars-Sinai Medical Center, University of California, Los Angeles, CA (RAJ).
Correspondence and reprints: Jonathan M. Teich, MD, PhD, Department of Emergency Medicine, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115; e-mail: <jteich{at}harvard.edu>.
Received for publication: 03/08/05; accepted for publication: 03/23/05.
 |
Abstract |
|---|
 Top Abstract Executive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
|
Executive Summary and Primary Recommendations |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
|
|
|
|
White Paper Purpose |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
The DHHS plays a major role in financing and regulating health care in the United States as well as in improving its quality. Ensuring that clinicians and consumers/patients use high-quality, timely, relevant medical information to guide their health care decisions is essential for improved quality of care, patient safety, and appropriate use of resources. The DHHS therefore has a strong interest in the availability and intelligent delivery of this medical information through CDS. More specifically, the Medicare Modernization Act of 2003 (MMA)1 calls for the Secretary of the DHHS to develop standards and guidelines for eRx systems that will be supported under the MMA. Appropriately developed and disseminated, CDS is an important ingredient in achieving the care improvements that these systems are expected to deliver.
This white paper provides recommendations for actions at a national level to help optimize the value and increase the use of CDS, particularly in eRx systems. Specifically, it discusses
This white paper focuses on benefits that can be realized specifically by CDS features, as opposed to those that accrue strictly from implementation of the underlying eRx infrastructure (such as legible prescriptions). Users and beneficiaries of the CDS interventions discussed in this report include clinicians, patients, pharmacists, pharmacy benefit managers, and payers.
Definition of Clinical Decision Support
Clinical decision support has been defined somewhat differently by different authors. Braden et al.,2 following Langton et al.,3 define it as "computer software employing a knowledge base designed for use by a clinician involved in patient care, as a direct aid to clinical decision making." Perrault and Metzger4 emphasize the relationship of knowledge to data in their definition: "a set of knowledge-based tools that are fully integrated with both the clinician workflow components of a computerized patient record, and a repository of complete and accurate data." In previous work,5 we have adapted these and other writings to establish a definition in functional terms: "providing clinicians or patients with clinical knowledge and patient-related information, intelligently filtered and presented at appropriate times, to enhance patient care." This includes not only the familiar reactive alerts and reminders (such as alerts for drug allergies and drug–drug interactions), but also many other intervention types, including structured order forms that promote correct entries, pick lists and patient-specific dose checking, proactive guideline support to prevent errors of omission (such as ensuring that appropriate patients are placed on aspirin), medication reference information for prescribers and patients, and any other knowledge-driven interventions that can promote safety, education, workflow improvement, communication, and improved quality of care.
A detailed treatment of clinical decision support in eRx, including practical issues of classification, usability, implementation, and evaluation, is presented as a chapter in the eHI consensus report Electronic Prescribing: Toward Maximum Value and Rapid Adoption.6 That report describes and references several ways of classifying CDS interventions7,8 based on when in the process the logic is executed, how it is delivered, and the global impact that it has on the process. A conceptual framework for evaluating outpatient eRx applications based on functional capabilities recently proposed by Bell et al.9 is an important step toward understanding variable CDS in this domain.
Clinical Decision Support Benefits
There are well-documented problems with the appropriate, safe, and cost-effective use of medications in health care.10,11,12 The very structure of most eRx applications, such as using standard drug dictionaries, selecting parameters from lists, and having required fields, can alleviate some of the problems associated with generating and filling medication prescriptions.6 However, supplementing this structure with CDS interventions aimed at those who enter, edit, and manage prescriptions offers greater leverage for achieving optimal patient care (Table 1).
|
|
Current and Desired State |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
Patient X is a 62-year-old woman with diabetes, borderline kidney failure, and high blood pressure. She has been seeing her primary care physician, Dr. Smith, for the past three years and has generally been pleased with her care. She arrives at the office for a visit, checks in at the front desk and then is ushered into an examination room. A few minutes later, Dr. Smith walks into the room to see her. He is carrying her paper chart, and he flips through it as they discuss her current issues. After some discussion and a brief examination, Dr. Smith determines that patient X has a sinus infection. He glances at the medicines that she is taking and his last written note about drug allergies and then handwrites a prescription for an antibiotic.Patient X then leaves the office with the written prescription and takes it to her pharmacy. The pharmacist puts the prescription into his computer and then informs patient X that the antibiotic is not covered on her benefit plan. Patient X goes back home and places a call to Dr. Smith's office. She speaks to a nurse who has a brief conversation with Dr. Smith, who prescribes an alternative antibiotic; the nurse then calls the new prescription in to the pharmacy. The next day, after a difficult night dealing with the symptoms of sinus infection, patient X goes back to the pharmacy. She receives some instructions from the pharmacist about how to take the drug and then returns home.
That evening she takes the first dose of the drug, and an hour later, she develops severe vomiting. Patient X calls her doctor's office again to report the new problem. When the message reaches Dr. Smith, he considers that perhaps the drug was given in too high a dose given her age and kidney function. He prescribes an antinausea medicine and yet another antibiotic. The antinausea medicine eventually controls her vomiting but makes her very sleepy, so much so that when she gets up that evening to go to the bathroom, she stumbles and falls, breaking her hip. She is taken to the hospital by ambulance and undergoes surgery the next morning to have her hip pinned.
When we first wrote this scenario, we were concerned that it was overly dramatic. However, we were quickly able to identify many real cases with consequences that were just as serious or even more so. Serious problems, leading to hospital admission and increased morbidity and mortality, occur frequently because of medication prescribing problems. The current state of medicine relies far too heavily on the memory of the practicing physician, both for important patient data and for relevant clinical knowledge. When Dr. Smith prescribed the first antibiotic, he needed to know the significance of the other drugs that the patient was taking, details about the dosing of that antibiotic for an older diabetic with kidney problems, the up-to-date formulary list of her medication benefit plan, and any details of her medication history that might preclude the use of a given medication. Given that physicians (and other prescribers such as nurse practitioners and dentists) are making these complex decisions several times per day in an environment where the number, complexity, and toxicity of drugs continue to expand rapidly, it is easy to see how the practicing physician needs more support.
Patient X arrives for her office visit. The nurse brings her back to the examination room and puts a preliminary diagnosis of "sinus infection" into the computer. Dr. Smith arrives to see her a few minutes later. After examining her and confirming the preliminary diagnosis, Dr. Smith clicks a button to reveal an evidence-based recommendation on the best antibiotic options for this condition. The computer returns a list of three antibiotic choices; next to each choice is an icon indicating whether that medication is covered on patient X's plan. The first antibiotic is off-formulary, so Dr. Smith selects the second antibiotic. The computer checks the patient's other active medications, and an alert window pops up indicating that the drug may interact with one of her diabetes drugs, resulting in vomiting (in fact, it was this interaction, not the patient's age or kidney function, which was responsible for patient X's vomiting in the first scenario; in that scenario, the physician never did make this connection).Dr. Smith contemplates giving her the adjusted dose of the drug and treating through the risk of vomiting. To be sure, though, he clicks a button revealing her drug history over the past 3 years. He notes that one of his partners gave a similar drug to her last year and the result was, indeed, severe nausea and vomiting. Armed with this highly relevant history, Dr. Smith cancels the drug order and selects the third antibiotic. No warnings appear this time, but the computer does recommend an adjusted dose based on her age and last measured kidney function, which Dr. Smith accepts. He confirms the prescription with a click, which directs the prescription to be electronically transmitted to the patient's local pharmacy and which also prints a concise patient's guide to the drug and its potential side effects. He reviews the prescription, dose, and potential side effects with patient X and prepares to discharge her from the office.
Before sending her home, however, he notes that the computer, which includes a full electronic health record as well as an eRx function, is recommending that the patient be placed on a cholesterol-lowering drug, based on her most recent cholesterol and LDL results and her diagnosis of diabetes; the system again shows which of the applicable drugs is on the patient's plan formulary. With two clicks, Dr. Smith prescribes this medication as well, again following the computer's recommended adjustment for age and kidney function. The computer also recommends a follow-up blood test (creatine kinase) after four weeks of therapy because of the potential risk of muscle inflammation with this family of drugs. With one click, Dr. Smith orders this blood test and instructs the patient to return next week to get the test done. The rest of patient X's course remains uneventful, and she recovers rapidly from her sinus infection without further incident.
|
Current State of Clinical Decision Support–enabled Electronic Prescribing |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
Features
Bell and colleagues9 and Schiff and Rucker20 have attempted to identify and classify the significant elements of eRx systems, including some elements of CDS. More recently, Wang et al.21 led a RAND Corporation field study that assessed the capabilities of ten commercially available eRx systems in 2002 and 2003. The data collected from this study provide information on implementation levels for 28 of the recommended CDS system features discussed later in this white paper. Preliminary results indicate that, in the mean, each product included 64% of the "Basic 2006" features (those features deemed by our panel to be essential for all systems by 2006), 34% of the "Advanced 2006" features, and only 12% of the "Basic 2008" and "Advanced 2008" features.
Taking all these studies together, we can conclude that eRx is growing in popularity but is still only found in a relatively small minority of U.S. practices, and even where it is used, available systems have many, but not all, of the most basic essential CDS features; advanced, higher value features are found in only a minority of commercially available systems. Thus, a majority of U.S. patients are not yet reaping the safety and quality benefits that can come from eRx with CDS.
|
Removing Barriers |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
Table 5 (available as an online data supplement at www.jamia.org) contains a more detailed listing of these barriers, along with potential high-level solutions.
In general, there are three areas where action is necessary to bring the current state of CDS closer to the desired state:
The next sections present detailed recommendations in each of these areas.
|
Recommendations |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
to help ensure that the recommendations in this report reflect broad input from the many different stakeholders in the prescribing and medication management process as well as from experts on clinical quality and informatics and from representatives of major health care information thought leadership organizations. The CDS Workgroup compiled an initial draft list of recommendations for the tables in this white paper. Expert panelists reviewed early drafts and provided comment. About 25 of the panelists convened in a half-day meeting at the Medinfo conference in San Francisco on September 9, 2004, during which the recommendations were discussed extensively, resulting in additions, deletions, reassignments, and clarifications of many items. The resulting version went through two more rounds of review with the panelists via e-mail, including editing between each round by CDS Workgroup members, to yield the final recommendations presented here. These recommendations have been presented in preliminary form to the Subcommittee on Standards and Security of the National Committee on Vital and Health Statistics for its use as it considers standards and guidelines for rule making pursuant to the MMA. In addition, because the recommendations are clearly applicable to potential certification of eRx and electronic health record systems, they have been shared in preliminary form with the newly formed CCHIT.
Core Features to Support Clinical Decision Support
Certain features of eRx systems can help ensure that knowledge and data are effectively used for safe, high-quality, cost-effective medication management. These recommendations fall into four areas:
The recommendations in each of these areas are divided into features expected of basic (minimally acceptable) and advanced eRx systems, and they are further divided to indicate features expected of eRx implementations in 2006 and those expected by 2008. Essentially, basic CDS functionality would be expected of all capable eRx systems implemented on or after the target date. Advanced functionality is that which clearly adds to the effectiveness and benefit of CDS; systems containing several elements of advanced functionality should be considered for increased favor through additional incentives.
These recommendations could be used as part of the health information technology certification process as it evolves. CCHIT is not specifically working on eRx in its first phase. Commission members, including the chair, have expressed interest in making use of this white paper's recommendations in ongoing CCHIT work, and in facilitating ongoing collaboration between CCHIT and the joint CDS Workgroup. In addition, these recommendations are intended to help guide requirements for participation in federal eRx activities under the MMA, such as demonstration projects and pay-for-performance programs.
The infrastructure required to fulfill the recommendations in the "organizational" column will vary from one site to another, but there are common themes and guidelines that can help. In particular, the Clinical Decision Support Implementers' Workbook5 contains a step-by-step guide to identifying stakeholders, understanding communications channels, setting goals, and establishing the necessary organizational structures for CDS implementation.
Standards, Structures, and Enablers
In addition to requiring specific features in individual eRx systems, there are other crucial elements of common infrastructure needed to support effective CDS nationwide.
Standards were explored in great detail in the eHI report6 that was presented to NCVHS on March 30, 2004. Using this report and many other sources of information in its deliberations, the NCVHS Subcommittee on Standards and Security provided initial recommendations for standards adoption in its letter of September 2, 2004, to the Secretary of DHHS. The recommendations here expand on those by adding more detailed needs and requirements and by proposing government actions to promote adoption and implementation of these standards.
Medical societies, health care organizations, informatics groups, knowledge vendors, and other stakeholders could all contribute to providing content to such clearinghouses. Government agencies could be important content contributors as well. However, rather than having a single government-controlled source of knowledge, the favored model would permit the publishing of multiple knowledge sets or clearinghouses by different agencies and groups, using a common structure. Local clinicians and managers would be able to select and configure specific interventions that are applicable to their situation.
Some specifics related to this concept have been briefly explored by the panel, including required elements, authorization, indicating level of evidence, organizational endorsements, and exchange standards. Considerable additional thought has been given to the concept by the CDS Workgroup, and the Workgroup has begun laying the foundation for further collaborative discussions and follow-on work, involving a variety of stakeholders.
Table 3 lists the recommendations for structures, standards, and other enablers that should be developed in a centralized or collaborative fashion to support effective, widely available CDS. Along with the specific suggested action items, we list possible government actions to promote and accelerate each item, and the time frame (based on Table 2) when they are needed.
Incentives and Related Issues
It is widely believed that adoption of eRx itself needs to be driven by financial, regulatory, and/or accreditation incentives. This is because providers bear a disproportionate share of the cost of implementing and using an eRx system, relative to the intrinsic financial benefits that accrue from its use (as outlined in the Center for Information Technology Leadership's report on ambulatory computerized physician order entry10). Specific incentive programs have been discussed in the eHI eRx report6 and expanded further in the March 2004 white paper produced by Rosenfeld et al.22 for the same organization. These reports contain substantial information on the foundation and the business case for eRx and CDS; we have used them as the jumping-off point for this brief discussion of practical action items.
Recommendations in Table 4 focus on three areas that the panel considered to be feasible, to address significant barriers to adoption, and to be specific to the use of effective CDS:
In addition, the CDS Expert Review Panel discussed mechanisms for carrying out certification of individual systems. One important controversy here is the question of whether certification should be based on a review of documented and validated system specifications, by performance in a test suite, or by performance and/or outcome metrics from actual use. The first method is easier to undertake but may not accurately reflect real-world performance; the second and particularly the third methods more closely characterize system benefits but are more difficult to implement. We recommend that the first method should be used for the initial stage of certification implementation but that there should be steady and prompt progress toward test case and actual occurrence reporting (see Table 4). Additionally, evaluating performance and outcomes of CDS-enabled eRx in actual practice may be dependent on local clinical conditions and patient mix. We have ceded this discussion to the newly formed CCHIT, which is specifically charged with deciding such issues; however, CDS Expert Panel consensus opinion on these various options has been shared with CCHIT commissioners, and we are maintaining an ongoing discussion with them. We have also shared preliminary versions of the CDS feature recommendations as potential elements for functionality certification.
As in the previous table, each incentive in Table 4 is described with its essential details and accompanied by recommendations for government action to promote its development along with an implementation timeline to keep pace with the recommendations of the previous tables.
|
Next Steps |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
Primary Use
Review and Dissemination
Further Work
|
Footnotes |
|---|
The authors thank the entire CDS Expert Review Panel for their helpful input and gratefully acknowledge the additional support and personal contributions of Gail Arnett, Doug Bell, Jeff Blair, Dasha Cohen, Kelly Cronin, Mike Fitzmaurice, Karen Greenwood, Zeba Kimmel, Nancy Teich, and Pat Wise.
Approved by the AMIA Board of Directors as an official white paper on March 14, 2005.
* The terms EMR and EHR are in a state of evolution. In this paper, we use the most current common usage available, specifically, an EHR is a collection of all person-centric health information; an EMR is a specific application primarily used in ambulatory care for clinical documentation, orders, data review, and workflow. 
Bruce Bagley, Marion Ball, David Bates, Douglas Bell, Jeff Blair, Jennifer Covich Bordenick, Suzie Burke-Beebe, Kelly Cronin, Don Detmer, Carol Diamond, Robert Elson, Michael Fitzmaurice, Mark Frisse, Tejal Gandhi, Peter Geerlofs, Lynne Gilbertson, Patricia Hale, Kathy Hollinger, Zebadiah Kimmel, Robert Kolodner, Gil Kuperman, Mark Leavitt, Michael Lake, Stuart Levine, Jane Metzger, Blackford Middleton, Arnold Milstein, Stuart Nelson, Eduardo Ortiz, Marc Overhage, Stan Pestotnik, Helga Rippen, Karen Trudel, Emily Welebob. Full affiliations of the panel members are available online as a data supplement at www.jamia.org.
|
References |
|---|
|
TopAbstractExecutive Summary and Primary...White Paper PurposeCurrent and Desired StateCurrent State of Clinical...Removing BarriersRecommendationsNext StepsReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. S. Field, P. Rochon, M. Lee, L. Gavendo, S. Subramanian, S. Hoover, J. Baril, and J. Gurwitz Costs Associated with Developing and Implementing a Computerized Clinical Decision Support System for Medication Dosing for Patients with Renal Insufficiency in the Long-term Care Setting J. Am. Med. Inform. Assoc., July 1, 2008; 15(4): 466 - 472. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. M. Kilbridge and D. C. Classen The Informatics Opportunities at the Intersection of Patient Safety and Clinical Informatics J. Am. Med. Inform. Assoc., July 1, 2008; 15(4): 397 - 407. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Grossman, A. Gerland, M. C. Reed, and C. Fahlman Physicians' Experiences Using Commercial E-Prescribing Systems Health Aff., May 1, 2007; 26(3): w393 - w404. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Osheroff, J. M. Teich, B. Middleton, E. B. Steen, A. Wright, and D. E. Detmer A Roadmap for National Action on Clinical Decision Support J. Am. Med. Inform. Assoc., March 1, 2007; 14(2): 141 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. J. Kuperman, R. M. Reichley, and T. C. Bailey Using Commercial Knowledge Bases for Clinical Decision Support: Opportunities, Hurdles, and Recommendations J. Am. Med. Inform. Assoc., July 1, 2006; 13(4): 369 - 371. [Full Text] [PDF]
|
|
|
|
|
|
R. Tamblyn, A. Huang, Y. Kawasumi, G. Bartlett, R. Grad, A. Jacques, M. Dawes, M. Abrahamowicz, R. Perreault, L. Taylor, et al. The Development and Evaluation of an Integrated Electronic Prescribing and Drug Management System for Primary Care J. Am. Med. Inform. Assoc., March 1, 2006; 13(2): 148 - 159. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Berner, T. K. Houston, M. N. Ray, J. J. Allison, G. R. Heudebert, W. W. Chatham, J. I. Kennedy Jr., G. L. Glandon, P. A. Norton, M. A. Crawford, et al. Improving Ambulatory Prescribing Safety with a Handheld Decision Support System: A Randomized Controlled Trial J. Am. Med. Inform. Assoc., March 1, 2006; 13(2): 171 - 179. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. F. Bria II Applied medical informatics for the chest physician: information you can use! Chest, February 1, 2006; 129(2): 446 - 451. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. S. Berner and J. Moss Informatics Challenges for the Impending Patient Information Explosion J. Am. Med. Inform. Assoc., November 1, 2005; 12(6): 614 - 617. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. E. Detmer and C. Safran AMIA's White Paper Policy Series on Timely Issues in Informatics J. Am. Med. Inform. Assoc., July 1, 2005; 12(4): 495 - 495. [Full Text] [PDF]
|
|
|
|
|
|
R. A. Miller, R. M. Gardner, K. B. Johnson, and G. Hripcsak Clinical Decision Support and Electronic Prescribing Systems: A Time for Responsible Thought and Action J. Am. Med. Inform. Assoc., July 1, 2005; 12(4): 403 - 409. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
