1) The RIS/PACS Experience at UCSD
2) Integrating the Healthcare Enterprise

The RIS/PACS Experience at UCSD

Using imaging informatics to improve patient safety

By Meg Richman, MD

Preventing morbidity and mortality has always been important. The health care industry, however, has fallen short in its goal of assuring patient safety. For example, both the Institute for Healthcare Improvement and the Canadian Patient Safety Institute recently looked at error rates in different fields of human endeavor. The only event to end up in the column labeled “dangerous” was hospitalization. Everything else typically thought of as dangerous (driving, coal mining, construction, bungee jumping, and rock climbing, for example) all had lower error and death rates compared to hospitalization.

If one compares reliability among industries, one finds, not surprisingly, that the airline industry is the most reliable, with an extremely low fatality rate. Unfortunately, at the opposite end of the scale are a number of medical services. Even tax advisors and airline baggage handlers score better on reliability than health care does.

Radiology, of course, is but one component of health care. Still, those of us in radiology must do more to reduce error rates and protect patient safety. One of the most useful tools for that purpose is PACS. This has allowed us to make giant strides toward improving patient safety. Another useful tool is the RIS. A robust RIS can greatly improve patient care, by making sure that the patient receives the correct exam and that the exam reaches the PACS quickly.

PACS and Its Impact on Clinicians

A PACS, in its most elemental form, allows referring physicians and clinician teams to view images and consult with radiologists entirely at the point of care. Assume, for instance, that a CT scan of the head shows blood herniating the brain from right to left. The neurosurgeon would once have needed to go to the radiology department to review this image. With a PACS in place, the neurosurgeon can now see the image in the operating room while planning a burr hole to evacuate the blood.

Expediting the availability of critical diagnostic images with the use of a PACS can help save lives. Three clinical examples make this clear. In the first, a patient with a bullet wound was treated in the trauma unit. The bullet had entered through the victim’s side and passed across the body cavity to the opposite side without exiting. To save the patient’s life, the trauma team needed to determine quickly how the bullet traveled from its entry point to its final (unknown) position. The answer was provided with the help of a PACS, which swiftly loaded and displayed the cross-table lateral image, showing that the bullet did not pass through any vital organs of the abdomen, but actually skirted them, and was relatively harmlessly lodged in the skin of the back.

The second case involved severe injuries due to a motor-vehicle accident. A C-collar had been placed on the patient’s neck prior to transport to the trauma center. Because she was burned in the accident, the C-collar was greatly distressing to the patient. The trauma-center nurses were about to remove the collar to provide relief, but they did not do so because the PACS delivered images to the point of care showing the existence of a serious cervical spine fracture of the 5th vertebral body. Owing to the rapid availability of those images, an error that might have caused quadriplegia was avoided.

In the third case, a chest radiograph captured a feeding tube in crisp detail. It was so readily visible because it had been inserted, by mistake, into the lung. This is a common problem when clinicians must place tubes and lines without the benefit of image guidance. As long as these incorrectly placed tubes can be removed quickly, the patient will not suffer harm. Thanks to the PACS, every image can be checked promptly, even if a radiologist is covering multiple hospitals, greatly reducing the potential for inadvertent harm from incorrect positioning of tubes and lines.

PACS Features and Their Impact on Radiologists

Multiple features available in PACS may seem commonplace, but actually greatly impact patient care. These features include: comparison studies, multiplanar reformat (MPR) capability, and 3D capability.

Viewing comparison studies, which a PACS facilitates, makes a tremendous difference to patient care. For example, a woman came in with marked focal enlargement of her aorta, seen on a routine chest x-ray. We were worried that she had developed a pseudoaneurysm (pocket of blood that can burst at any moment). By looking at her prior chest x-ray, we knew immediately that this was a new finding. We conducted a chest CT scan, which confirmed the diagnosis and convinced the cardiovascular surgeons to operate urgently. If we did not have those prior images, we would have incorrectly assumed that her abnormalities were chronic, and she would not have received the care she needed. Indeed, had this pseudoaneurysm ruptured, the patient could have died in an instant. The Impax 6.2 PACS from Agfa HealthCare, Greenville, SC, allows radiologists to see key images from prior studies immediately. It also makes reference studies available at the push of a button. We can know the patient’s history and access all information that will help us make the right diagnosis promptly.

Apart from making comparison studies available quickly, the PACS also offers huge patient care benefits by executing multiplanar reconstruction of images. For example, a CT scan of the abdomen of a cancer patient showed a metastatic lesion in the spine. On the traditional axial cross-sectional views, the spine involvement was subtle and could have been missed, even by a seasoned radiologist. The MPR capability allows the radiologist to see the sagittal view, and, suddenly, it was easy to see that one of the vertebral bodies was completely eaten away, with tumor extending into the spinal canal.

In contemporary imaging, we need all three vantage points (axial, coronal, and sagittal) to be available quickly and easily. Not long ago, we imaged a child’s head and detected a subdural hematoma (blood layering on top of the brain). This would have been almost impossible to diagnose from the axial view, but not with the sagittal-coronal view incorporated. In this particular case, the diagnosis was especially important because this subdural hematoma in this child was caused by physical abuse. By making this diagnosis, we were able to report the abuse and have the child rescued.

Woman with marked focal enlargement of her aorta seen on a routine chest x-ray (above right) was suspected of having a pseudoaneurysm. With immediate access to key images from prior studies (above left) via Agfa’s Impax 6.2 PACS, UCSD radiologists ruled out a congenital defect and confirmed that this was a new finding. Cardiovascular surgeons were convinced to operate urgently.

Three-dimensional (3D) imaging is another aspect of radiology in which a PACS can prove itself invaluable. Historically, to generate 3D images, the radiologist had to move from one workstation to another, one that was able to generate 3D images. This, of course, was highly inefficient. Certain PACS solutions can solve this problem by providing an environment in which 3D images can be generated within the PACS itself. Do 3D images merely provide “pretty” pictures? No—3D images can change patient management in an instant. For instance, a 34-year-old man was ejected from his car during a crash. What looked like an innocent slight bulge along the thoracic aorta on the axial images was correctly interpreted as an aortic transection, or tear, on the 3D images. Such an injury causes death within 1 hour 90% of the time. The 3D images allowed us to save his life by making the accurate diagnosis.

Information Integration

Equally effective is the combination of a PACS and a RIS at a single workstation. In an enlightened radiology department, it is no longer acceptable to operate the PACS and the RIS as separate systems; the University of California at San Diego recognizes the necessity of blending the respective strengths of the two into an integrated whole. In the past, before our integrated RIS/PACS became available, radiologists were obliged to use a separate computer system to look up a patient’s history, costing many valuable minutes. Today, images and pertinent patient information are accessible from one workstation, giving us all the information we need to do a better, faster job of evaluating studies. Similarly, the integrated RIS/PACS spares us both the accidental loss of important work and the resulting patient callbacks by making reference images and other materials available on-screen.

Workflow itself is a constant source of concern for radiologists. At least 15 different steps must take place from when the clinician orders the study to when the clinician receives the report. Unfortunately, the potential exists for patient harm to occur at every one of those steps. Integration of the PACS and RIS is of great utility in reducing such possibilities.

A facet of workflow that causes problems for many radiology departments is patient scheduling. Lengthy waits for service after arrival at the radiology department’s reception area engender some of the strongest complaints from patients. An integrated RIS/PACS solution reduces this problem dramatically. In many instances, patients are taken to the examining room within minutes after they check in at the front desk. This is possible because, with an integrated RIS/PACS, sophisticated radiology departments are no longer waiting for paper to move throughout the department. The impact on patient satisfaction alone is remarkable. At one site, the introduction of RIS/PACS integration caused patient satisfaction scores to jump, within a month, from 35% to 86%.

Another problem eliminated by RIS/PACS integration is that of studies going unread for unacceptably long periods. An integrated RIS/PACS warns of any studies sitting in the queue for which reports have not yet been dictated and can help ensure that every study will be read within 24 to 48 hours.

Meanwhile, of the eight Joint Commission patient-safety goals (four of which pertain to radiology), one easily addressed with the help of the integrated RIS/PACS is the safe use of medications (in this case, contrast agents). Administering contrast to someone with impaired renal function can cause permanent kidney damage. Many situations arise in which we ordinarily think it is imperative to give contrast; for instance, a patient who has been beaten in the abdomen would clearly benefit from the use of contrast to exclude organ injury. If the suspicion of injury is low, however, and the person has impaired renal function, then the study must be obtained without contrast. All too often, radiologists do not have easy access to patients’ laboratory values and can inadvertently protocol a study with contrast when a patient should not receive contrast. The remedy for situations of this sort is RIS/PACS integration. Now, radiologists can look at laboratory values, histories, and prior images, all at the same time and all within the PACS workstation.

Talking Points

In our department, digital dictation is a fully integrated system feature. Consequently, the process of dictation now is vastly more efficient, and reports are generated far sooner. Prior to implementation of our newest digital transcription system, 50% of our reports were turned around in less than 24 hours; a month after that system’s debut, that percentage jumped to 85% (and 5 months later, it stood at 90%). Faster report turnaround times mean that the clinicians receive the reports and act on their findings much more rapidly. All of this translates into better and more accurate patient care.

Also, newly integrated with our PACS is the Veriphy critical test-result management system from Vocada Inc, Dallas. This is an enterprise solution for swiftly communicating critical test results from hospital diagnostic departments to ordering clinicians. It employs a proprietary hosted system to automate and verify the communication of critical test results in real time, which has the effect of enhancing patient safety and boosting physician productivity. Now, we no longer face the problem of being unable to reach the patient’s clinician team in the event that our findings require immediate attention (a situation that, in a major medical institution, can easily occur 20 times a day). The system automatically pages the physician and, if there is no response, works its way up the chain of command, all the way to the department chair, if necessary, until it reaches someone who answers the page. It also maintains, for medicolegal auditing purposes, a log of page attempts and callbacks.

Meshing With the Mission

Without question, PACS is growing in importance to radiology departments. The reasons are many, not the least of which is the fact that imaging volume is growing at exponential rates. PACS helps us do a better job of getting images in for interpretation quickly and prioritizing them so that the most crucially important images are read first.

In order to maximize productivity, however, a department’s PACS ought to be elegantly and logically structured, making it extremely user friendly. It should have on-screen buttons where they naturally belong so that the tools needed for each task can be found easily and quickly. When it comes to this type of informatics technology, a radiology practice, group, or department should be certain that the PACS that it chooses offers the most feature-rich environment, the fastest retrieval speeds, the best integration, and the best reliability.

Our mission, in radiology, is to take the best possible care of patients. Having a full-function, fully integrated solution incorporating PACS, RIS, and digital dictation meshes perfectly with that mission. It enables us to meet the needs of clinicians and their patients to an extent that we previously could not have imagined. The PACS is a tool that allows us to find, read, and prioritize images and then generate a report that can be promptly delivered to the referring physician. We could not live without PACS.

Meg Richman, MD, is director of Thornton Radiology at the University of California at San Diego, and vice chair of education in the university’s Department of Radiology.

Integrating the Healthcare Enterprise

IHE can increase efficiency, improve workflow, and positively affect patient care

By Tim Kaschinske

Integrating the Healthcare Enterprise (IHE) is an initiative undertaken by health care professionals and the industries that serve them to improve the way that computer systems in health care share information. Originally involving the Healthcare Information and Management Systems Society and the Radiological Society of North America, the project has grown to include the American College of Cardiology and many other organizations worldwide.

IHE is intended to serve as the link between the standards now in use for information exchange and their practical application to the delivery of health care. The ultimate result will be improved patient care because:

• IHE permits various medical information systems to communicate;

• clinical workflow improves, and the repetition of data-related tasks on various systems is eliminated; and

• the integrity of patient information is protected, along with its continuity.

IHE promotes the coordinated use of established standards for information exchange, particularly Health Level 7 (HL7), Digital Imaging and Communications in Medicine (DICOM), the Kerberos network authentication protocol, and various Internet-related protocols. IHE accomplishes this coordination through the use of three entities: integration profiles, actors, and transactions.

An integration profile is used to establish broad categories of compatible function, to describe clinical function and workflow needs, and to define actors that participate within the integration profile. The integration profile also defines transactions that occur between actors within an integration profile. Examples of integration profiles established by IHE to date include scheduled workflow (see Figure 1 below), patient information reconciliation, consistent presentation of images, presentation of grouped procedures, patient identifier cross-referencing for the master patient index, and retrieval of information for display.

IHE actors are information systems or applications that produce, manage, or act on information. An actor supports a specific set of IHE transactions and is represented as a functional unit. A given information system may support one or more IHE actors, but partial implementations of actors are not allowed.

IHE transactions are exchanges of information between these actors. They use messages based on established standards, such as HL7 and DICOM. Each transaction is defined with reference to a specific standard and provides additional detailed information.

The concept of multiple domains is a recent addition to the IHE initiative. IHE began in a single domain: radiology. Information-technology infrastructure (ITI) then became necessary as the project progressed, so an ITI domain was formed to define ITI integration profiles. Next, the cardiology domain was formed to define interoperability within cardiac catheterization and echocardiography environments. Thus, there are currently three domains, each defining its own integration profiles, actors, and transactions. A fourth domain, laboratory, is in the process of being formed to define interoperability for laboratory data.

Figure 1. An example of scheduled workflow.

All of these IHE components are brought together for testing in the IHE Connectathon. This is an event to which the vendor community is given an open invitation to bring relevant systems for real-time, face-to-face testing of interoperability using advanced testing tools. The testing is organized and supervised by the IHE project management team. Dozens of vendors and systems are involved, thousands of cross-vendor tests are performed, and the results are recorded and published. Impax 6 from Agfa HealthCare, Greenville, SC, was successfully tested at the Connectathon, having implemented several actors within several integration profiles, including image manager/archive and image display within the scheduled workflow and patient information reconciliation integration profiles. Complete details on the actors and integration profiles supported are available in the Impax 6 integration statement.¹

Why IHE Matters

IHE matters to radiology because it can improve service to patients and referring physicians. In addition, it can increase efficiency, improve workflow, enhance the academic mission, and modernize information systems. IHE integration profiles have been designed to solve problems and address issues within specific operational areas.

The scheduled workflow integration profile, for example, addresses several issues involved in providing service within radiology. How does one order and fulfill a request for imaging service? How many steps are involved? How much paper is used? How does one know when a procedure is finished? How does one know what was actually done (versus what was planned, but possibly not done)? What happens when someone makes a mistake?

The patient information reconciliation integration profile addresses issues concerning unidentified patients. How does one order and fulfill a request for imaging service in the absence of key information? How does one deal with trauma patients? How does one deal with John Doe patients? What if the admission-discharge-transfer/patient registration actor is not working? What if the order placer is down?

The charge posting integration profile addresses the issues associated with billing for radiology services. How does one make sure that the billing system receives timely and accurate information on what is being done? Does one recode procedures for what was done (instead of what was planned)? Does one track charges at each step of the process?

The consistent presentation of images integration profile addresses the issues involved in image viewing. How does one control image quality in a distributed electronic environment? How can one make sure that what is visible to the radiologist is visible to the clinician? How does one make sure that an image looks the same on different workstations and on film? How does one manage image quality across a large range of devices?

The access to radiology information integration profile addresses the issues concerning the distribution of radiology data. How does one let other information systems access images, key image notes, gray-scale presentation states, and reports?

Issues of security are clearly associated with such access; the basic security integration profile addresses them. For example, how does one coordinate auditing and message logging between heterogeneous information systems? How does one time synchronize systems? How does one recognize authorized information systems? Can someone spoof the scanner or the PACS?

The Request for Proposal and IHE

Given the importance to radiology of the problems that the integration profiles address, many of the buyers compiling requests for proposals (RFPs) for PACS are now asking specific questions about IHE (see Table 1 at right). For example, purchasers are asking vendors to describe, in detail, the relationship between the system being considered for purchase and IHE. Understanding and specifying this relationship in advance can save the purchaser considerable money, time, and frustration, in addition to ensuring that a newly acquired system is actually providing the purchaser with up-to-date capabilities.

In many cases, however, the IHE-related questions incorporated in the RFP are the product of inadequate consideration and/or prepurchase investigation. Some, in fact, indicate an unmet need for education concerning IHE on the part of the potential purchaser. Although it may be helpful to ask general questions, the points that become part of the RFP should be based on the specific needs of the facility, not on a set of generic principles that may quickly become inapplicable or out of date. It should always be helpful to ask which IHE profiles and actors are supported by each component of the system, however, as well as to ask the vendor to identify the system’s ability to support any options identified by IHE.

There is much disparity in IHE knowledge in the user community. The vendor has an important role to play in helping the purchaser understand how the company’s products implement IHE. Of course, the vendor must have a thorough understanding of IHE to answer the user’s questions correctly. No boilerplate answers will work everywhere in the current IT environment, so it is a great advantage to work with a vendor whose representatives can help buyers understand not only the proper use of IHE terminology, but the importance of this initiative and how it benefits real-world radiologists, clinicians, and patients.

Tim Kaschinske is a product manager for connectivity products, including PACS Broker and Connectivity Manager, at Agfa HealthCare. He has been involved with the IHE initiative starting with Radiology Year 1.

Reference

1. Impax 6 integration statement. Available at: www.agfa.com/healthcare/ihe. Accessed January 21, 2007.