| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
* From the Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, University of Michigan Health System, Ann Arbor, MI.
Correspondence to: William F. Bria, II, MD, Department of Internal Medicine, Division of Pulmonary and Critical Care Medicine, University of Michigan Health System, 1500 E. Medical Center Dr, 3916 Taubman Center, Ann Arbor, MI 48109-0360; e-mail: wbria{at}umich.edu
Abstract
The world of applied medical informatics is changing rapidly. This is the first of a three-part series of articles on applied medical informatics that will bring the practicing chest physician up to date on the structure, function, benefits, and drawbacks of the electronic medical record and all of its components, including the virtual ICU and the daily practice of medicine.
Key Words: applied medical informatics • computerized physician order entry • electronic medical record
As I sat down to begin this series of articles on applied medical informatics (the science of processing data for storage and retrieval), I had an epiphany. What the world does not need right now is yet another stolid treatise on information technology, saying how wonderful it all is and how it should be used. The current reality is that in the United States of America, we are at the bottom of the list of developed nations in use of the electronic medical record (EMR)! One must ask, Why is that? What is the problem? Should I care that this situation change and, if so, how fast? One of the key missing elements in the national debate on adoption of the EMR12 has been straight talk on what this is all about, what is the reason for the gap in use, and how we get over it and move forward. That is the purpose of these articles. Parts 2 and 3 will appear in subsequent issues of Chest. In the course of this discussion, each part of the EMR will be examined and reviewed for its positive and negative attributes, to understand the forces that have lead to this current state of affairs in the United States, and to explain how we can become better informed consumers of the available health-care information technology (including the latest virtual ICU and other technologies).
The Plan
This series of articles will address the following questions: (1) What is the EMR (anatomy and physiology); (2) Why do three out of four American chest physicians find themselves without access to a complete EMR?; (3) As a chest physician, what can/should be done to get the benefits of an EMR in my practice?; (4) What about my patients? What can this informatics do for them? The personal health record; and (5) What is next: the virtual ICU.
What Is the EMR (Anatomy and Physiology)
In the wake of the September 11, 2001, attacks and Hurricane Katrina, the EMR in America has taken on a broader perspective than its traditional hospital-centric role. A recent review by Stead et al3 describes three dimensions of health-care information: (1) health-care provider dimension; (2) personal health dimension; and (3) population health dimension (Fig 1 ). Understanding this broader context of health information indicates the need for an electronic health record being a combination of a provider-centric EMR and a patient-centric personal health record (Fig 2 ). First, the EMR as the introductory piece of the world of applied medical informatics will be discussed.
|
|
The EMR has undergone innumerable descriptions and redefinitions over the last 35 years. As a practical matter, the EMR can be understood as containing three major components: (1) A results-reporting information system (RRIS); (2) a computerized physician order-entry system (CPOE); and (3) a clinical decision support system (CDS). To begin to be an informed medical consumer and participant in this EMR discussion, an understanding of these parts is necessary
RRIS: What Is it?
The RRIS can be understood as the essential first component of the EMR, including presentation of clinical test results and information from all allied health departments across the continuum of care, inpatient, outpatient, acute, and chronic care settings. In the most recent Healthcare Informatics Management Systems Society Survey4 of American chief information officers, > 50% indicated implementation of an RRIS. The RRIS includes several key functional characteristics: accuracy, timeliness, patient centricity, comprehensiveness, and effective presentation.
RRIS: What Does It Do?
An RRIS can integrate all important patient information in a familiar Web-based interface (CareWeb; University of Michigan [not for sale]; Fig 3
). Effective presentation must include demonstration of data trends either tabular or graphic format (Fig 4
). Multimedia (bronchoscopy images or video, audio) also is now an expectation of any RRIS (Fig 5
), permitting virtual chart reviews when and wherever needed (this is a key enabler for a future discussion of virtual ICU). Access security standards are now well described,5 enabling effective distribution of this element of the EMR throughout a health system (or region).
|
|
|
|
Now after this understanding of the RRIS, the most important thing for the practicing chest physician to realize is its ease of use. This component of the EMR is far and away the simplest piece to realize value from, to learn, and the one that (unlike the next component, CPOE) is installed in the majority of American hospitals today. Having said that, there are many (how many is unknown) physician offices in America that still do not have a comprehensive RRIS (results from office patients and participating hospitals) installed. The era of smudgy faxes and poorly copied medical records is far from over yet. The good news is that there is legislation pending in Congress (expected to be passed in the next year) that will temporarily suspend the Stark Amendment6 permitting health systems to distribute computers and RRIS systems to regional physicians offices to at least begin the process of building the EMR around the country. So the bottom line is that this is the one component that without a doubt should be in your practice, today.
CPOE: What Is it?
CPOE systems are traditionally thought of as inpatient order communication systems allowing instant creation and transmissions of orders throughout a health system. After the RRIS, CPOE is the oldest component of the EMR yet has been the most difficult to perfect and implement in this country (more detail on this in part 2 of this series).
Typical anatomy of a CPOE system includes a pick list function, usually by department (such as pharmacy, laboratory, radiology) of all services, products, and functions available for patient care in any given health system. The structure of these pick lists is quite important, since studies have demonstrated that clinicians will often select the first item in a list (for example antibiotics) which, if alphabetical listings are utilized, might not necessarily be the ideal recommendations from the Pharmacy and Therapeutics Committee perspective.
Also typical for CPOE systems is the order set. Order sets allow a system to take a static view of workflow for a particular diagnosis (eg, empiric therapy for community-acquired pneumonia), a procedure (before/after bronchoscopy orders), or even a snapshot of a particular moment in the care process (admission/discharge order sets). In the early days, order sets were held up as examples of cookbook medicine, and it was feared that interns and residents trained at an institution with order sets would not be able to function without them in their future careers. These concerns have turned out to be unfounded. However, order sets are essentially electronic versions of preprinted order forms. As a medical director knows, preprinted forms get out of date, need regular monitoring, and by their existence, do not constitute a standard of care that all clinicians will adhere to. Hence, it is not so much how an order set is constructed, but how the data within them are managed over time. Additionally, order sets, for most commercial EMR systems, are tied directly to the institutions version of the EMR software only. The implications of that are, with few exceptions, the idea of sharing order sets automatically between institutions/practices is just a great idea. There is hope that the Office of the National Coordinator for Health Information Technology (ONCHIT) efforts at EMR certification and standardization may have impact here, since this is one of the most time consuming and clinically surprising drawbacks of this component of the EMR.7
One additional aspect of CPOE systems that needs to be discussed is interfaces. Interfaces are electronic links between systems allowing rapid transmission of data. For many reasons—the free market forces in American business, the rapidly changing technologies in information technology—the difficulty of connecting many systems (even from the same vendor) is a lot more trouble than most nontechnicians might expect. When interfaces exist, they may be either one way or two way. The implications of this for the physician are numerous. For example, a pharmacy system that is not interfaced with a CPOE system would require someone to actually collect the printout orders from the CPOE system, and (if there was a separate pharmacy system) type those orders into that system, thereby increasing the chance that transcription errors could occur. Furthermore, if alerts appeared in the pharmacy system only, then the ordering physician would either miss those alerts or be notified in traditional methods with accompanying delays. The bottom line is ask. Ask if the particular CPOE being considered has in production an interface with the key departmental systems (laboratory, radiology) and ask to directly contact customers at that site to find out how well things are going. Some up-front, firsthand investigation in this key dimension of a CPOE system can save tremendous problems later on, especially if one of the justifications of the CPOE system is improved patient care efficiency and safety.
As mentioned above, CPOE usually applies only to inpatient order communication systems; however, the practice reality in 2005 of inpatient orders that continue in the ambulatory arena and vice versa (the so-called continuum of care) is just now being more aggressively addressed by products in the marketplace. Yet, due to system costs and complexity, projects usually start in one setting, usually inpatient, then extend to the ambulatory setting, sometimes involving two competing vendor products (see interfacing above!). e-Prescribing, the specific automation of ambulatory care medication ordering and administration, is a subset of this dynamic, and is outside the scope of this part of the series.
CDS
CDS has been defined as a system that uses two or more items of patient data to generate case-specific advice.8 In practical terms, CDS includes a wide range of EMR subsystems, including definable rules, alerts, and workflow tools that allow the EMR to try to aid in improved medical decision making, improved efficiency of patient care and communications, and enhancement of patient safety. Rules are, at their core, built on if/then logic statements that allow a tremendous amount of flexibility and power to be added to EMR systems, allowing the introduction of guidelines and evidence-based medical protocols into the regular workflow of on-line patient care.
In practice, however, in most commercial systems today, the business end (the user experience) of CDS is the alert box. Growing numbers of studies9 are beginning to reveal the critical limitation of alerts that, by design, interrupt the clinician’s workflow; in particular, in order entry. The primary reason for this limitation is that currently available CDS systems alert post hoc after the clinician has requested a particular thing (drug dosage, test). Interestingly, some recent literature10 suggests that first-year interns are much more likely to ignore alerts in their rush to finish their many tasks than their senior resident and fellow colleagues. The notion of passive alerts has been raised, where, for example, medications to which the patient is known by the system to be allergic to are designated with special color or symbols to avoid clinician ordering at the outset. Of the listed components of the EMR, CDS is the youngest, and many new and exciting developments are on the way, further adding to the real clinical value of the EMR. That CDS is key, however, to the overall realization of value from the EMR has been demonstrated in evaluation of return-on-investment studies of some of the most well-established EMRs in the country.11 Several excellent recent publications review the structure, function, and future of CDS and are well worth reading, especially for those involved with evaluation and management of the CDS component in hospital/offices EMR.11
Next Installment
The reasons for the current state of the art of the EMR in America will be reviewed, as well as key lessons learned, all essential for success.
Footnotes
Abbreviations: CDS = clinical decision support system; CPOE = computerized physician order-entry system; RRIS = results-reporting information system
Received for publication November 29, 2005. Accepted for publication December 3, 2005.
References
This article has been cited by other articles:
|
|
 |
|
  W. F. Bria II Applied medical informatics for the chest physician: information you can use!--Part 3. Chest, April 1, 2006; 129(4): 1057 - 1060. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. F. Bria II Applied Medical Informatics for the Chest Physician: Information You Can USE!--Part 2 Chest, March 1, 2006; 129(3): 777 - 782. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
