(Radiographics. 2001;21:1015-1018.)
© RSNA, 2001
Computerized Physician Order Entry and Decision Support: Improving the Quality of Care1
Ramin Khorasani, MD
1 From the Department of Radiology, Brigham and Women’s Hospital, 75 Francis St, Boston, MA 02115. From the Plenary Session, Friday Imaging Symposium: Challenges to Radiology in the New Millennium, at the 2000 RSNA scientific assembly. Received March 12, 2001; revision requested April 2 and received May 7; accepted May 7. Address correspondence to the author (e-mail: rkhorasani@partners.org).
Index Terms: Computers • Quality assurance • Radiology and radiologists, departmental management • Radiology and radiologists, design of radiological facilities
 |
Introduction
|
|---|
Information technology is transforming the practice of medicine and radiology. There may be only a handful of radiology departments in this country that are not familiar with picture archiving and communication systems (PACS), radiology information systems, or speech recognition technologies. However, equating radiology information technology with these entities alone does not fully exploit the power of information technology to change the process of care in radiology and thereby create real value in patient care.
In this article, we will (a) review the process of care in radiology and suggest some of the reasons that change is needed, (b) discuss the importance of extending the radiologist’s role in helping improve physician test ordering practices to improve quality of care, (c) define an important information technology infrastructure (computerized physician order entry [POE]) necessary to improve the quality of care, and (d) propose ways to use this infrastructure to create value in radiology.
|
Process of Care in Radiology: Where Does It Begin?
|
|---|
For most radiologists, the process of care in our departments or practices begins when we receive a requisition from a referring physician. We then perform the examination and generate a report. The report has traditionally been distributed to our referring physicians by mail, by fax, or (in some cases) by means of electronic transmission through hospital information systems. Not infrequently, we call our referring physicians to ensure that they are aware of abnormal or important findings. With respect to information technology, this part of the process has received the most attention in our practices. We are rapidly moving from a film environment to a PACS environment. More and more departments are depending on speech recognition technology to generate near-real-time reports that can then be distributed with the corresponding images to physicians outside the department, by means of either the World Wide Web or the electronic infrastructure of hospital information systems. However, the great majority of outpatient text reports still travel to our referring physicians on paper.
Radiologists are not typically involved in the care process before requisitions are generated. Infrequently, referring physicians consult their favorite radiologists to obtain guidance as to the imaging strategy that is best suited to their patients in a particular clinical setting. We believe that engaging radiologists in this part of the process to help choose appropriate imaging strategies creates value and is therefore important to our patients, to referring physicians, and thus to our specialty. We will use the term medical decision support to refer to this activity. Automating this activity with POE could result in significant additional value for both nonradiologists and radiologists, as we will demonstrate. The success of order entry is greatly enhanced by its seamless integration into the information management infrastructure of the enterprise. It should be clear that such integration would play a key role in enhancing the value of radiology. POE can be integrated seamlessly into a PACS, resulting in the distribution of electronic multimedia information to users of imaging services. The reporting could be enhanced with use of notifications and reminders for referring physicians and surrogate outcomes tracking, which could help generate new content. The ultimate goal of all these potential changes for radiology is comprehensive information management. Some have suggested that information technology is the next modality for radiology to embrace.
|
Medical Decision Support
|
|---|
Imaging accounts for 5%–10% of healthcare expenditures, and many believe that this figure will increase due to the aging population and to rapid innovation in imaging technologies such as computed tomography (CT), magnetic resonance imaging, and positron emission tomography. At the same time, there is significant concern—particularly outside radiology—over the inappropriate use of these technologies, which could increase spending and reduce the quality of care. There are at least two major hurdles to ensuring more appropriate use of imaging: (a) lack of agreement on what constitutes "appropriate" use of imaging studies and how it contributes to improved patient outcomes, and (b) the difficulty in changing physician behavior, which is magnified by the lack of data regarding the appropriateness of imaging in a specific clinical context. Medical decision support refers to improving physician test ordering behavior to reduce costs and improve the quality of care. The goal is appropriate use of imaging resources. It is important to emphasize that, although reducing the number of unnecessary studies can save money in radiology, there are instances in which increasing the number of imaging studies would be more appropriate, resulting in cost reduction outside radiology. For example, it has been suggested that routine CT of the appendix in emergency department patients will reduce overall healthcare costs by reducing the number of unnecessary appendectomies (2).
Tools for changing physician behavior include education (eg, expert opinion, consensus-based guidelines, and, most important, evidence-based guidelines or decision support), feedback (physician profiling), rationing (eg, precertification programs), and penalties (eg, capitation, withholding). Education is generally believed to be the most important element of any intervention designed to change physician test ordering behavior (1).
Several rules of thumb should be considered in devising an optimal strategy for improving physician decision making. If physician behavior can’t be "measured," change or improvement is unlikely. Such change requires accurate and detailed data collection methods. We should take advantage of the synergistic relationship among the different tools described earlier: a combined use of the tools is much more effective than any one tool implemented alone (1). Intervening at the time of clinical decision making has significant advantages. In addition, information should be directed to those physicians making the clinical decisions; targeting interns when in fact a more senior physician makes the clinical decision may be suboptimal (3). To impact radiology overall, interventions must be implemented for inpatients as well as outpatients. Finally, interventions to change physician behavior must be scalable, modular, and reasonable in cost.
It is not difficult to envision, then, how POE can provide the infrastructure necessary to implement an optimal strategy for changing physician test ordering behavior. This is not new to medicine; many such interventions have been successfully implemented in other domains such as pharmacy and have actually been shown to improve care and even save lives by reducing medical errors (4). However, radiology interventions have been few and, so far, of limited effectiveness (3), partly because the necessary infrastructure has not been widely available to take advantage of the synergistic effects of interventions that involve multiple tools (eg, education, profiling). For inpatients, POE is typically made available with laboratory and pharmacy programs through hospital information systems. The significant void in outpatient POE will ultimately be filled by components of electronic medical record products that are being increasingly deployed throughout integrated delivery systems. Although the ultimate solution for outpatient POE will depend on the institution, it is becoming clear that extensive use of Web technology will likely reduce the overall cost of deployment and provide greater opportunity for integration into the relevant information technology infrastructure. To provide an effective platform for radiology medical decision support, information technology solutions must address outpatient radiology, which accounts for 60%–70% of all imaging tests performed and 10%–50% of which is "leaked" to radiology providers outside the enterprise. The loss of information from "leaked" studies may adversely affect patient care, thereby proving even more critical than the revenue that could have been retained within the enterprise had such "leakage" not occurred.
|
Our Experience
|
|---|
Brigham and Women’s Hospital funded the development of a Web-enabled modular software program for outpatient radiology medical decision support (RIMS; Medicalis, Chestnut Hill, Mass) to complement its extensive inpatient POE. This program is increasingly being integrated into the information management infrastructure of our enterprise and is methodically being implemented throughout our affiliated practices, beginning with the primary care practices. To persuade physician practices to use the application, several "carrots" were built in. We made possible the online scheduling of radiology studies and built in sophisticated risk management tools (notifications, reminders) for ordering physicians. The application also allows text and image distribution with Web technology. Physicians or their office staff can choose radiology examinations and their indications from structured menus. These indications can then be mapped to the appropriate International Classification of Diseases, 9th Revision or Current Procedural Terminology billing codes to improve billing efficiency and significantly reduce billing errors. Inappropriate indications could then potentially launch real-time, context-specific decision support to aid the requesting physician in choosing the best imaging strategy. Physician profiling and precertification programs can also be incorporated into such applications when necessary. Data generated from the use of these applications can help identify physician test ordering behavior and can then be used to devise targeted interventions to improve physician behavior and, ultimately, to improve care. Evolving data warehousing technologies are being used to create new content for decision support. These data could also provide a sophisticated health services research infrastructure for radiology (5). Wide variation exists in outpatient radiology test ordering behavior among primary care physicians, even those in the same practice (Fig 1). Understanding the reasons for this variation and devising interventions with POE to reduce the degree of variation could result in improved quality of care.
View larger version (19K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1. Histogram illustrates the variability of outpatient radiology test ordering among primary care physicians (P < .0001). Each bar represents an individual physician. Physicians in the same practice are represented by bars of the same color.
|
|
|
Return on Investment
|
|---|
The use of POE will be of great benefit to radiology. There are significant operational gains within radiology if such a system is implemented effectively. The accuracy of information (e-requisitions) from the referring physician is improved, which should help radiologists make more accurate and meaningful diagnoses. Mapping of appropriate billing codes will improve billing efficiency within the radiology department and the entire hospital. Online scheduling will improve service. Risk management features will increase the value of radiology to our users. Finally, POE allows the deployment of real-time, context-specific information as well as real-time implementation of other medical decision support tools to provide "just-in-time" information at the point of care. The research opportunities, as described earlier, are truly endless!
|
Availability and Implementation of POE in Clinical Practice
|
|---|
A recent Institute of Medicine report (6) and subsequent White House action have set the stage for mandated medical error reduction in healthcare. Although POE has clearly been shown to help improve physician behavior and reduce medical errors (4), the majority of U.S. hospitals do not make POE available. In a report by Ash et al (7) in 1998, two-thirds of hospitals surveyed did not provide POE, and, in the one-third that did, POE was not widely and routinely used. Successful implementation of POE is a complex process (8). With further improvements in bandwidth, hardware, and software and increasing content offered through POE, expansion of its role in clinical practice is certain. Successful implementation requires detailed attention to workflow and process reengineering on a larger scale than what is needed for a complex PACS implementation because of the sheer number and variety of users and the more widely dispersed physical plant of a hospital compared with a radiology department. POE will be a required element of the electronic medical records. It should be noted that the availability and implementation of electronic medical record programs involves less than 10% of hospitals—probably closer to 5% at present—but its prevalence is growing. Brigham and Women’s Hospital has been one of the pioneers of POE. For example, every inpatient radiology examination performed at our hospital must be requested through POE (over 150,000 studies per year).
The level of functionality available through POE and its integration into the information management infrastructure of the enterprise are critical variables in its successful deployment and use. The more value that POE creates for ordering physicians and their office staff, the more likely they will be to use it. Thus, detailed attention to functionality that improves operations in a physician’s office is critical in distinguishing inadequate from optimal POE applications.
|
Value Proposition for Radiology
|
|---|
How does radiology enhance its value in the process of care? Should it be service value or content (knowledge-based) value (Fig 2)?
View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2. Diagram illustrates how information technology initiatives in radiology can add service value (italicized concepts) and content or knowledge value (underlined concepts) to the process of care. Integration into the information system infrastructure of the enterprise will be a prerequisite for success in most if not all cases.
|
|
It is understood that improved services (eg, greater ease of scheduling, improved patient satisfaction, shorter report turnaround times, improved image availability) are important determinants of the day-to-day success of a practice or department. In the long run, however, content value will be the sustainable competitive advantage for our specialty. This content value will derive from expanded information about appropriate use of testing, improved diagnosis with the aid of computers, more timely notification and reminders for referring physicians, and improved content in radiology reports similar to that which is possible with structured reporting or standardized taxonomy (eg, the American College of Radiology Breast Imaging Reporting and Data System for mammography [9]). The value of the content is enhanced significantly if it is delivered "just in time" and is context specific.
Computerized POE will be critical in further enhancing the value of radiology and will allow radiologists to be fully integrated participants in the information management initiatives of the enterprise. The complexity of the task should not deter us from taking advantage of this great opportunity!
|
Footnotes
|
|---|
Abbreviations: PACS = picture archiving and communication system, POE = physician order entry
|
References
|
|---|
-
Greco PJ, Eisenberg JM. Changing physicians’ practices. N Engl J Med 1993; 329:1271-1274.[Free Full Text]
-
Rao PM, Rhea JT, Novelline RA, Mostafavi AA, McCabe CJ. Effect of computed tomography of the appendix on treatment of patients and use of hospital resources. N Engl J Med 1998; 338:141-146.[Abstract/Free Full Text]
-
Harpole LH, Khorasani R, Fiskio J, Kuperman GJ, Bates DW. Automated evidence-based critiquing of orders for abdominal radiographs: impact on utilization and appropriateness. J Am Med Inform Assoc 1997; 4:511-521.[Abstract/Free Full Text]
-
Bates DW, Leape LL, Cullen DJ, et al. Effect of computerized physician order entry and a team intervention on prevention of serious medication errors. JAMA 1998; 280:1311-1316.[Abstract/Free Full Text]
-
Rothschild JM, Khorasani R, Silverman SG, Hanson RW, Fiskio JM, Bates DW. Abdominal cross-sectional imaging for inpatients with abnormal liver function test results: yield and usefulness. Arch Intern Med 2001; 161:583-588.[Abstract/Free Full Text]
-
Institute of Medicine. Crossing the quality chasm: a new health system for the 21st century Washington, DC: National Academy Press, 2001.
-
Ash JS, Gorman PN, Hersh WR. Physician order entry in U.S. hospitals. Proc AMIA Symp 1998; :235-239.
-
Ash JS, Gorman PN, Lavelle M, Lyman J. Multiple perspectives on physician order entry.. Proc AMIA Symp 2000; :27-39.
-
Kallergi M, Gavrielides MA, He L, Berman CG, Kim JJ, Clark RA. Simulation model of mammographic calcifications based on the American College of Radiology Breast Imaging Reporting and Data System, or BIRADS. Acad Radiol 1998; 5:670-679.[Medline]
This article has been cited by other articles:
|
|
|
|
 
R. H. Gottlieb
Imaging for Whom: Patient or Physician?
Am. J. Roentgenol.,
December 1, 2005;
185(6):
1399 - 1403.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
K. Juluru and J. Eng
Internet-Based Radiology Order-Entry, Reporting, and Workflow Management System for Coordinating Urgent Study Requests During Off-Hours
Am. J. Roentgenol.,
March 1, 2005;
184(3):
1017 - 1020.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Introduction
Process of Care in...
