Picture Archive Communication Systems (PACS)

Picture Archive Communication Systems (PACS)

The Effects of Picture Archiving Communications Systems (PACS) and Computerization on Radiology Workflow and Turnaround Time

Medical diagnostic imaging has experienced numerous innovations since the discovery of X-rays in 1895 that allowed physicians to look inside the human body for the first time without using a scalpel. Perhaps some of the most important of such innovations to date have been the introduction of computer-based radiology information systems and picture archiving communications systems. While these technologies are still being refined and new uses for them are being identified, it is clear that these innovations represent the beginning of a ubiquitous computing element in radiology. The vast majority of organizations planning to implement picture archiving and communications systems (PACS) are acutely aware of the need to integrate the hospital information system (HIS) and radiology information system (RIS) with the PACS; however, relatively few are aware of the fundamental challenges associated with its implementation and administration. To determine the effects of these innovations, this paper will provide an analysis of how the advent of radiology information systems (RIS) and picture archive communication systems (PACS) have affected today's radiology department, including reduced turnaround times required for examinations, improved and streamlined workflow, immediate delivery and accessibility of digital images virtually anytime anywhere. A summary of the research will be provided in the conclusion.

Review and Discussion

Background and Overview. According to Geracimos (2004), "Radiology is considered a branch of medicine that uses radioactive substances, electromagnetic radiation and sound waves to create images of the body, its organs and structures for the purpose of diagnosis and treatment"; properly used, these systems can help clinicians understand how effectively the body and its internal organs and structures are operating (B01). Radiographers today are increasingly relying on computer-based digital imaging equipment to produce images of the tissues, organs, bones, and vessels of the body (Holcomb 17).

Innovations in the supporting technology in recent years, though, have provided radiographers with the ability to provide these services faster than ever before, with the digital results being available to clinicians almost as soon as the images are taken: "The equipment is all computer-based these days," reports Cathy Parsons, vice president of the American Society of Radiologic Technologists; "You put the patient on the table, make the exposure, and the image is almost immediate" (Holcomb 17). In this dynamic environment, it is little wonder that clinicians and patients alike may be confused by the rapid advances taking place around them, a factor that may constrain wider acceptance until they have proven themselves over time. In this regard, Stanney (2002) points out that in the field of radiology in particular, "A number of new technologies will not be accepted by physicians or patients because of unfamiliarity, conventional prejudice, or outright ignorance of the benefits" (954). Clearly, though, understanding and applying these new technologies has assumed greater importance than ever in many medical applications as new uses for them continue to be identified. To this end, the benefits of these innovations in radiological applications are discussed further below.

Benefits of Radiology Information Systems (RIS). Information that is required for radiological applications is increasingly being managed by several autonomous medical information systems including hospital information systems (HIS), radiological information systems (RIS), and picture archiving and communications systems (PACS) (Breant, Taira & Huang 88). According to Albensi, Ilkanich, Dini, and Janigro (2004), "Simply put, an image is a visual representation of some measurable property of an object, person, or phenomenon. Imaging systems create and record images for the eye to view and the mind to experience. Some imaging systems create a visual map of what the eye and brain can already see and perceive"; still other types of imaging systems transduce visual data that is frequently indiscernible to the human eye into visible forms (1127). In their essay, "Evidence-based Radiology: Requirements for Electronic Access," Bui, Taira, Dionisio, Aberle, El-Saden, and Kangarloo (2002) report that, "Imaging is an effective method for screening (e.g., mammography), for documenting the presence or absence of a disease or condition (i.e., testing of a medical hypothesis), and for managing many medical conditions (e.g., follow-up of cancer).

Part of the challenge in this process in the field of radiology is to avoid misrepresenting the essential character of the object or phenomenon. "Therefore, it is important to choose an imaging system that will do the job and assist in solving a scientific question, rather than a system that is not appropriate for the application or that simply produces beautiful graphics and serves little practical use" (Albensi et al. 1127). Since their introduction in recent years, Abelha, Machado, Alves and Neves report that PACS have enjoyed the same type of reaction to new technologies that their counterparts have in other industries by providing their users with a taste of what can be accomplished with such innovations while still being constrained by a number of limitations. To date, these authors report that a great deal of attention has been focused on better understanding these constraints, and in identifying solutions to them. "The overall trend has been away from seeing the process as one of encoding heuristics derived from an expert," they say, "towards modeling the domain on which the expertise operates" (5).

One company that recognized that practical applications of these new technologies is Ris Logic Inc.; Ris introduced its radiology software at a Cleveland Convention Center trade show in May 2002 (Robinson 4). The client server software, called Ris Logic CS, is designed to help users comply with the Health Insurance Portability and Accountability Act of 1996 which was intended to protect health insurance coverage for people when they change or lose their jobs, and to bring uniformity to the electronic transmission of health information (Robinson 4). The company's director of marketing, Christine Boehm, reported that the company believes demand for the new software will help Ris Logic duplicate the growth pattern that emerged after its launch at its Radiology Society of North America show of Version 4.0 of its software the previous year. That version of the software automated record-keeping functions such as billing, appointment scheduling and patient reports for outpatient radiology centers (Robinson 4).

In 2001, Ris Logic more than doubled its customer base to 33 clients from 16 the previous year, with four new customers signing on shortly after the Radiology Society show, Ms. Boehm said. The new software will make managing a group practice with multiple office sites easier, Ms. Boehm said. The integrated patient information means basic data for each person and procedure is entered into the system once and everyone in the radiology practice has access to it. Ms. Boehm said that feature saves time and reduces the possibility of mistakes (Robinson 4).

Ris Logic has grown to 24 employees from three at its founding in 1999, and it isn't finished yet. Ms. Boehm said the company wants to hire another salesperson as well as someone to work on product implementation. One of the company's newest customers is Kentucky Diagnostic Center, a three-office diagnostic imaging business in northern Kentucky. Sam Grippa, president and CEO of Kentucky Diagnostic, said he needed to replace his old DOS-based scheduling system as his business grew. Mr. Grippa said he chose Ris Logic because its software is Windows-based and easy to use. "I did look at some other (software systems), one that I'm not sure even I could learn," Mr. Grippa said. "I look at it this way - if I have to put someone on hold to look up the information they need because I have to figure out how to access my computer system, I've wasted too much time" (Robinson 4). Grippa reported also appreciating the licensing terms for the Ris Logic software. "I don't have to call for permission every time I add something," he said (Robinson 4).

In their article, "Just the right prescription," the editors of Communications News (2005) report that the Kansas Spine Hospital in Wichita specializes in neurosurgery, spinal surgery, pain management and radiology; this facility is also one of the first hospitals in the country to completely digitalize patients' medical records, including prescriptions and clinical records to x-rays and other radiology images, and make them accessible by computer (24). The goals for the completely digital system were established by the hospital's development team and required that a digital telecommunications system before the institution opened in 2003. According to the healthcare provider's CIO, Michael Knocke, the initiative's objectives included ensuring that there was a reliable and durable digital telecommunications system in place that featured open standards that would allow seamless integration and convergence with its data network; the information system was also required to be one that would permit future growth and migration to new technologies and capabilities (Just the right prescription 24).

To meet these objectives, the hospital collaborated with Great Plains Communications of Wichita, one of the area's providers of communications solutions for the healthcare industry. Following a needs assessment, Great Plains suggested that a Toshiba Strata CTX670 business-communications system with Toshiba Stratagy IES 12-port voice mail, and unified messaging would be an appropriate choice for the hospital. This RIS includes Toshiba digital desktop telephones, plus 24 SpectraLink wireless telephones; these wireless phones were programmed to extend the features and capabilities of the users' desktop telephones to anywhere they roam at the 22-bed hospital; however, other wireless technologies such as cellular, could not be used due to the interference with sensitive medical equipment (Just the right prescription 24). The hospital's chief operating officer, Darryl Thornton, reported, "Being able to be mobile, yet still receive all our telephone calls, has greatly improved our efficiency and productivity. Toshiba's SpectraLink solution was the only one we found that would extend our desktop telephones to the palms of our hands" (Just the right prescription 24). The RIS also accommodated the needs of the hospital's remote users: "The system is so flexible that we were able to connect our remote users and still have it look like they are located at our corporate offices," Thornton added. "Both incoming and outgoing calls are routed via the remote users' extensions through our main system at the hospital, so it's totally transparent that they aren't here at the hospital" (Just the right prescription 24).

The hospital also had unique requirements for the Strata CTX670, including paging that could be restricted by area and the ability to restrict long distance on specific telephones, such as those in the lobby and in patient rooms, which required special programming. To deliver the hospital's integrated voice and data communication, Great Plains partnered with TelCove, a provider of business-critical telecommunications services to enterprise customers and carriers, and NetVision Technologies, a provider of data networking services and technology consulting, all with offices in Wichita. TelCove delivered the external network, while NetVision handled the internal network. TelCove installed a fiber connection at the hospital connecting it to TelCove's synchronous fiber-optic network (SONET), to provide both local network services and long distance. The SONET topography design enables all elements on the SONET ring to continuously communicate with each other, with information being routed in both directions, so if an element is inactive for any reason, the network stays active. At the hospital, the optical network terminates the voice services on ISDN, using a primary rate interface (PRI) that runs through the Toshiba switch via PRI cards; the system also delivers a full T-1 of Internet service (Just the right prescription 24).

The reliability of the optical network was, of course, a high priority for the hospital; the system's reliability was tested in the school of hard knocks when a tornado destroyed two miles of the network cabling; however, in spite of the damage, the network did not experience any downtime. NetVision installed the hospital's internal voice and data network, including setting up the wireless local area network, multitiered security and a high-capacity, fully gigabit infrastructure. "The key to the success of both the data networking and voice applications is having a wireless network that is free of interference and fully compatible with both applications" reported Brent Burdick, NetVision president (Just the right prescription 24). NetVision was heavily involved in the overall design of the network, from placements of cabling and wireless access points to infrastructure issues such as heating and cooling. Planning the system before the hospital was built was a critical element to the system's deployment, Thornton says, as was having a battery backup in place as the system was deployed (Just the right prescription 24).

Finally, the hospital's new wireless telephones have proven to be more useful than the planners first imagined: "The ability to be mobile and receive your telephone calls at the same time is one of the biggest benefits I've ever seen in a hospital telephone system. Our telephones have given us a huge advantage in communicating with patients, families, doctors, other medical facilities and each other" (Just the right prescription 24). "Being an open system, this system will let us add voice over IP or other capabilities as our needs change and as we grow," Knocke concluded (Just the right prescription 24).

Benefits of Picture Archive Communication Systems (PACS). According to Creighton (1999), picture archiving and communication systems (PACS) were originally developed to combine viewing of modality images, archiving, and distribution of images (138). "When PACS is integrated/interfaced with radiology information systems (RIS) or hospital information systems (HIS)," he says, "it can merge patient demographics, medical records, and images" (Creighton 139). In their essay, "Computers in imaging and health care: now and in the future," Arenson Andriole, Avrin and Gould (2000) report that, "Early picture archiving and communication systems (PACS) were characterized by the use of very expensive hardware devices, cumbersome display stations, duplication of database content, lack of interfaces to other clinical information systems, and immaturity in their understanding of the folder manager concepts and workflow reengineering" (145). These earlier systems were implemented historically at large academic medical centers by biomedical engineers and imaging informaticists. PACS were nonstandard, home-grown projects with mixed clinical acceptance. However, they clearly showed the great potential for PACS and filmless medical imaging. Filmless radiology is a reality today (Arenson et al. 145).

The introduction of efficient softcopy display of images provides a means for dealing with the ever-increasing number of studies and number of images per study. Computer power has increased, and archival storage cost has decreased to the extent that the economics of PACS is justifiable with respect to film. Network bandwidths have increased to allow large studies of many megabytes to arrive at display stations within seconds of examination completion. PACS vendors have recognized the need for efficient workflow and have built systems with intelligence in the management of patient data. Close integration with the hospital information system (HIS)-radiology information system (RIS) is critical for system functionality. Successful implementation of PACS requires integration or interoperation with hospital and radiology information systems (Arenson et al. 145).

In an effort to address a number of emerging issues involved with communication between PACS and RIS, as well as to facilitate interface development and provide faster turnaround time for clinicians, a number of healthcare organizations have developed standards for the formatting and transfer of clinical data (Creighton 139). In addition, research continues to identify better approaches to resolving these issues. For example, Creighton cites the example of Communication protocol Health Level 7 (HL7) as a standard application protocol that is being used for electronic text data exchange in healthcare by most RIS today. The imaging communication protocol for PACS is the Digital Imaging and Communications in Medicine (DICOM) standard specification protocol; this protocol describes the means of formatting and exchanging images and associated information (Creighton 139).

In their journal article, "DICOM modality worklist: an essential component in a PACS environment," Gale and Gale (2000) discuss the emergence of this evolving communications protocol and report that, "The development and acceptance of the digital communication in medicine (DICOM) standard has become a basic requirement for the implementation of electronic imaging in radiology" (101). These authors point out that DICOM is now evolving to provide a standard for electronic communication between radiology and other parts of the hospital enterprise (Gale & Gale 101). In his essay, "HIS/RIS/PACS integration: getting to the gold standard," Boochever (2004) reports that the technology for acquiring, storing, retrieving, displaying, and distributing images has experienced enormous innovations in recent years. The demand for such enterprise-wide imaging management solutions in which digital images from radiology and other services are integrated in a seamless fashion together with information from clinical information systems and other databases; such digital images can then be accessed seamlessly from a single point of end-user interaction.

Calling it a "gold standard" of system integration, the ideal approach to this integration, Boochever says, would be to provide an integrated platform that provided improved workflow, patient throughput and patient safety, as well as decreased cost. "Unfortunately," he says, "the gold standard remains elusive in most healthcare environments, even those with new systems" (16).

One of the earliest problems that confronted software developers in this regard was the need for a protocol that would allow communication between the HIS/RIS/PACS systems and between Health Level-7 (HL7) and DICOM. "This barrier was solved by the broker -- a software and hardware device that accepts HL7 messages from the RIS then translates, or maps, the data to produce DICOM messages for transmission to the PACS" (Boochever 16). According to this author, technologist workflow requires patient and examination data from the RIS to be communicated to the modality; the broker then provides support for this by exploiting the DICOM Modality Worklist (DMWL); however, two fundamental constraints remain inherent to most brokered configuration: 1) workflow remains paper-driven and 2) RIS information flows in one direction only, resulting in redundant databases (Boochever 16). Resolving the limitations of HIS/RIS/PACS connectivity will require industry-accepted communication protocols and rules; Boochever notes that the Integrating the Health Care Enterprise (IHE) initiative was developed to achieve this level of seamlessness within the HIS/RIS/PACS systems. "The goal of IHE," he says, "is to provide end-users improved access to critical patient and clinical information across all systems within the healthcare delivery network. While the IHE initiative began to facilitate more efficient, predictable, and functional integration between disparate systems, vendors still had technology hurdles to overcome" (Boochever 16). In addition, system integration remains a major constraint, with many practitioners experiencing the inevitable effects of turf battles and office politics as the technology continues to be refined. In fact, it would appear that there will not be much to fight about in the very near future since developers continue to provide solutions to the numerous problems reported by users of previous incarnations of these systems. According to Boochever, "In response to these challenges, several vendors have begun to offer consolidated RIS/PACS solutions and/or HIS/RIS/PACS solutions. Consequently, the prospect of the gold standard appears to be on the horizon. Single vendor consolidated systems are not, however, feasible for deployment in many healthcare organizations, and they are not necessarily the panacea" (16).

In a completely integrated filmless radiology department, there are three core computer systems, the picture archiving and communication system (PACS), the hospital or radiology information system (HIS, RIS), and the acquisition modality. "Ideally, each would have bidirectional communication with the other two systems. At a minimum, a PACS must be able to receive and acknowledge receipt of image and demographic data from the modalities. Similarly, the modalities must be able to send images and demographic data to the PACS" (Gale & Gale 101). Today, the fundamental DICOM communication protocols for query or retrieval, storage, and print classes have become established through both conformance statements and intervendor testing; as a result, there has been an increase in interest in enhancing the functionality of communication between the three computers. "Historically, demographic data passed to the PACS have been generated manually at the modality despite the existence of the same data on the HIS or RIS. In more current sophisticated implementations, acquisition modalities are able to receive patient and study-related data from the HIS or RIS" (Gale & Gale 101). The DICOM Modality Worklist represents the missing electronic link that serves to transfer this critical information between the acquisition modalities and the HIS or RIS (Gale & Gale 101).

Picture archiving and communication systems (PACS) are described by Cuadros and Sim as being "formal applications that can effectively address the needs of clinicians to communicate" (1). Emerging data and knowledge interchange standards have demonstrated enormous potential in allowing clinicians to exchange comprehensive healthcare information through universally interoperable PACS and RIS applications in the future; however, these systems have not been extensively used in most outpatient settings because of their high cost as well as the technical and organizational difficulties that have typically been associated with implementation. So-called "light weight" solutions to clinical communication may offer more immediate accessibility clinicians, particularly those in small outpatient settings. These authors conclude that it is important that these partial solutions eventually be interoperable with more comprehensive solutions that are expected to be readily available in the near future (Cuadros & Sims 4). Integrating multiple subsystems such as HIS/RIS/PACS integration, though, means that there must be a ready exchange of the workflow control information, and consistency of the information between subsystems (Osada & Nishihara 103).

In her report, "Image of cancer," Geracimos (2001) notes that the introduction of an early picture archive communication system (PACS) called RapidScreen by Georgetown Medical Center and University of Maryland Medical Systems in Baltimore represented some of the first applications of this technology in the United States. According to Geracimos, this early version was "retrofitted" to use existing imaging techniques and then digitalize them; however, future versions would eliminate this step by incorporated the digital images provided by newer machines. "The cancer nodule in the patient's left lung is a blurry gray polka dot that is barely visible on the screen even though it is marked by a bright white circle," Geracimos says. "The nodule may be vague, but the image and the circle together represent an important step forward in the battle for early detection of a cancer that claims more American lives than any other - some 178,000 new cases each year, leading to about 160,000 deaths, according to the American Cancer Society" (1). To date, the disease has been highly lethal largely because of the difficulties of finding cancer traces sufficiently early enough for treatment to begin in time to prevent the cancer from spreading (Geracimos 1).

According to Dr. Matthew Freedman, associate professor at Georgetown University Medical Center, intervention in later, more critical, stages is not usually effective enough to effect a cure. "It's a race against time because most cancers of any kind take five years before they are detectable in any form," Dr. Freedman notes. Freedman has worked for a number of years collaborating with the Rockville firm responsible for creating a machine that identifies deviant tissue that would most likely be missed on standard X-ray film ("even by skilled radiologists"); the system identifies these suspicious areas by placing rings around them in the process (Geracimos 1).

The trademarked RapidScreen RS-2000 is an extension of a computer-aided detection (CAD) system designed by Deus Technologies; the system digitizes and analyzes chest X-rays effectively enough to have won U.S. Food and Drug Administration approval after a six-month clinical research project headed by Dr. Freedman, also a consultant for Deus (Geracimos 1). The company reports that this PACS (which was approximately the size and shape of an office copier) could detect up to 14% more lung cancers than were able to be detected by existing methods. Research Corp., which has performed high-tech research and development for clients including the National Aeronautics and Space Administration and the National Institutes of Health, wanted its initial product in the area of digital radiography to be one that was sorely needed in the medical field (Geracimos 1). Like other emerging technologies, though, the high cost of this early version of RapidScreen was prohibitively expensive for most institutions (in 2001, the system cost $170,000); however, also like all of the other technologies reviewed, the price of this PACS was expected to drop as demand increased in the future. "The more sales, the faster the company is compensated for the total costs of research, said by Dr. Freedman to be "in the hundreds of thousands of dollars" (Geracimos 1).

By late 2001, Georgetown Medical Center and University of Maryland Medical Systems in Baltimore were the only facilities in the United States to have these systems installed (Geracimos 1). The author notes that Maryland Medical Systems was allowed to evaluate the system prior to purchase and at the time of this report, Georgetown had been using its system for just two weeks. Deus provided grant funds to the hospital and supplied it with a machine on a permanent basis with the understanding that patients would not be charged directly for its use under their insurance plans. Dr. Freedman is adjusting the machine for different-size X-ray film before it goes into routine use. "You will select which patients should have this done routinely. For example, you might say everyone over age 40 or 35. Everyone who smokes might qualify. I don't know yet," he says (Geracimos 1).

The clinical trials performed at the hospital used blind testing on X-ray films of past patients, some of whom had cancer and some of whom did not. Dr. Freedman employed 15 radiologists throughout the Washington area for the project, working with 80 cases of lung cancer and 160 non-cancer cases. In nearly every instance, the radiologists were able to detect a nodule with an average size of 15 millimeters (three-fifths of an inch) that previously had been overlooked. That is the size, he says, that radiologists most often miss, as has been shown in two studies; false positives exist, he says, but not to an alarming degree (Geracimos 1).

Physicians used the earlier version of this PACS by initially reviewing chest radiographs (X-rays) and making an initial interpretation. The X-ray film was then loaded onto the RapidScreen's vertical scanner and the physician was required to enter a code on the keyboard that sent a signal for the software to digitize the image which was then viewable on a monitor; this image indicated whether any cancer nodules were present. "Eventually, the film will be acquired digitally and will be reprinted digitally, thereby speeding the process" (Geracimos 1).

Dr. Freedman was using the machine one recent afternoon to see whether it could be adapted for detecting tuberculosis, which he calls "the worst scourge worldwide. There is great need for a system like this for TB" (Geracimos 1). "Meanwhile," he says, "If we could get this machine implemented widely, a half-million small lung cancers could be found around the world each year that should be curable. At present, most lung cancers can't be cured" (Geracimos 1).

The company's Web site, www.deustech.com, asserts that lung cancer can be found in one out of 200 nonsmokers and one out of 100 smokers and that 85% of early-stage cancers can be treated successfully. Dr. Freedman flags smokers with a dire warning: "Remember, you are at risk 15 years after stopping if you have smoked heavily. And a moderate smoker, even one using filtered cigarettes, is at risk smoking just 20 packs a year for 20 years" (Geracimos 1).

In his article, "NAS thriving in SMB market," Ortiz (2005) points out that an increasing number of hospitals and clinics are implementing picture archive communication systems (PACS) that employ network-attached storage (NAS) connected disks in order to store medical images from X-ray, CT, MRI, ultrasound and other imaging modalities. "The most advanced medical imaging studies may involve a stream of high-resolution images in which the view is rotated in three dimensions, requiring the storage subsystem to store large image files and deliver them as high-resolution streaming video images" (Ortiz 1-2).

While PACS have traditionally used DAS or SAN storage, their requirements are particularly well suited for high-performance, cost-effective NAS implementations; consequently, these systems are being used for both the short- and long-term storage of medical images (Ortiz 2). Furthermore, in the U.S., a major factor in the growth of PACS storage has been the requirement to comply with regulatory demands of the Health Insurance Portability and Accountability Act of 1996 (HIPAA). According to Ortiz, "These regulations mandate strong patient privacy protections and other requirements regarding the preservation and use of medical records, and they are driving demand for cost-effective mirroring and archiving solutions" (3).

A recent study by Bui et al. (2002) examines the cost-effectiveness and other factors associated with the use of PACS at a tertiary care hospital. According to these researchers, "All radiologists performed image interpretation by using the departmental PACS. Each radiologist had access to the written requisition for the imaging examination, and PACS prefetched the most recent (and other) previous images for comparison; these images were made available on a dedicated imaging workstation. In addition to the PACS workstation in each radiology reading room, separate terminals provided access to the RIS (all radiology reports), the HIS (laboratory data, electrocardiograms, operational reports, and medical consults), and the Internet (Bui et al. 664).

A simplified use case model for the radiologist work flow is shown in Figure 1 below; this figure illustrates the interaction between multiple actors and use cases.

Figure 1. Simplified Use Case Model for Radiologists Using the Integrated PACS and RIS Application.

Source: Bui et al. 665.

In Figure 1 above, the actors are shown on the left, and the system is shown in the large box. Individual use cases are shown in ovals; an alternate flow is shown with dashed lines. This is a high-level view of the radiologist reading process, because as many of these "steps" are use cases involving finer levels of detail; however, the diagram serves to highlight the fundamental flow and end goal of the radiologist and the interaction of multiple use cases.

The authors report that observations of the radiologists in the clinical environment showed that they relied predominantly on interpretation of images from PACS to render their conclusions, with only occasional RIS terminal use and even less frequent HIS access. The principal explanation for this behavior most likely relates to the inconvenience of accessing such data: Three different computers are required to access three separate clinical systems (ie, PACS, RIS, HIS). The Internet was not used by attending radiologists to search for external medical evidence; instead, they relied on their training and experience to reach conclusions; however, residents frequently used the Internet and reference textbooks as a source of external information while they waited for the attending radiologist's review (Bui et al. 664).

Based on their findings, Bui and his colleagues suggest that the first step toward a more effective radiology practice is for radiologic consult reports to be as comprehensive as possible, similar to the medical reports from other clinical fields (e.g., pulmonology, urology). A comprehensive consult report by radiologists that incorporates all relevant data -- "as opposed to just a description of the findings -- "may assist in the transition of radiology from the area of an ancillary service toward the mainstream practice of medicine. Bui and his associates also point out that streamlining the process in the future will require improved access to patient information and access to current literature. "In the former," they say, "imaging with any relevant patient data (e.g., laboratory tests, histopathology) are used to reach conclusions and to render a consultation report. In the latter case, explicit external evidence from outside sources of knowledge (e.g., peer-reviewed medical articles) is employed to help validate conclusions" (Bui et al. 665).

The ideal method for the practice of radiology can be equated to known scientific methods as shown in Figure 2 below.

Figure 2. The ideal process of care consists of four major stages; this process can be compared to known scientific methods whereby a problem is identified, causes are hypothesized, experiments are conducted to test hypotheses, results are analyzed, and a conclusion is then drawn.

Source: Bui et al. 667.

The authors believe that this approach is "perhaps the idyllic" (in other words, high-level) use case model for radiology today. According to their summation:

The first step is patient assessment and formulation of a sound hypothesis; for a radiologist, this requires access to information relating to the patient's presentation, pertinent past medical history (e.g., surgery before the imaging examination), and other medical assessments. This first step is equivalent to the evidence-based medicine practice of defining the problem and then formulating a question. On the basis of this initial information, the most effective imaging examination is then selected to test the hypothesis (i.e., experiment). The examination selected, however, must be the most effective, not the most efficient, evaluation procedure (emphasis added). (Bui et al. 667).

The key to improvements in the RIS/PACS approach to delivering more efficient and effective imaging services is the objectification of a patient's subjective clinical presentations, thereby facilitating a more "scientific" treatment of the data obtained by the radiologists. Because objective information is used to reach a conclusion, all applicable patient information must be accessible so that an informed decision can be made (Bui et al. 667).

In the case model of radiologists by Bui and his colleagues, physicians had access to patient information; however, this information was not fully integrated. In their concomitant case studies of the pulmonary and urologic departments, the authors found that clinician access to patient information was more rapid, and that only relevant information was presented. As a result, clinic staff are trained to effectively preobtain specific patient data from the RIS and HIS -- " and to filter out nonrelated information -- " for the practitioner; for example, pulmonary clinic staff typically did not retrieve an RIS report on a patient's broken leg but they did retrieve all chest radiograph consults (Bui et al. 667).

The needs for radiologists in this regard are significantly different, though, because typical support staff in this instance are responsible only for retrieving past film images from the film storage area; in sharp contrast to their peers in other disciplines, radiologists themselves are responsible for gathering past PACS images (usually prefetched by the system), for accessing the RIS and HIS, for integrating data, and for filtering out nonpertinent patient information (Bui et al. 667). The current electronic environment, however, means that this type of information is not routinely used simply because of the inconvenience of accessing it; furthermore, because many radiologists face increasing demands on their time and performance is measured in terms of the number of studies read (e.g., relative value units), radiology consults are constrained to image interpretation, with only limited incorporation of other relevant patient data. "Ultimately," they say, "the constraints of quick access to integrated and relevant patient data can be overcome effectively in an appropriate electronic information environment" (Bui et al. 667). Therefore, it is increasingly important for the supporting electronic infrastructure to expedite retrieval of patient information by integrating multiple data sources (Osada & Nishihara 103). In fact, an integrated electronic medical record was the focus of several ongoing research projects cited by the authors specifically targeted at radiologists. "Indeed, all radiologists involved in our study indicated their desire for a single workstation with access to HIS, RIS, PACS, and the Internet so they could better access and utilize patient data" (Bui et al. 667). While these researchers note that formal testing to determine if use would actually increase under these circumstances was not performed, a pilot study that evaluated an integrated PACS-RIS-HIS interface demonstrated that, on average, radiologists did use more RIS and HIS information to guide their final conclusions if such information was directly available (Bui et al. 667).