| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Perspectives |
1 From the Department of Radiology, Massachusetts General Hospital, MZ-FND 216, Box 9657, 14 Fruit St, Boston, MA 02114. Received August 2, 2005; accepted August 9. Address correspondence to the author (e-mail: thrall.james{at}mgh.harvard.edu).
Advances stemming from the digital age of imaging have positioned radiology on a critical path for the diagnosis of disease and/or treatment of most patients admitted to short-term care hospitals in the United States. This new position for radiology has brought with it a higher expectation from clinicians, who are increasingly dependent on imaging in their practices, and a higher profile with the public. It has also required hospitals and other health care organizations to markedly increase investment in their radiology departments in order to install innovative new digital systems and increase the total capacity for imaging. Hospitals have high expectations for both appropriate returns on their investments and the productive management of the increased institutional resources devoted to radiology.
It is important to look critically at the implications of these changing stakeholder expectations and to respond effectively. The management challenge for leaders in radiology is to match the spectacular capabilities of the various new digital and interventional technologies that are transforming radiology with equally creative new ideas and solutions for facility design, work process reengineering, and deployment of a changing radiology "job family" (ie, all the different kinds of jobs in radiology). These new approaches are vitally needed for the success of the next generation of radiology practices and to help assimilate advances in technology in pursuit of increased productivity and quality.
New Jobs in the Digital Department
The effective management and support of radiology information systems, picture archiving and communications systems (PACS), and voice recognition systems are central to success in the digital age. Radiology departments must either provide support for these systems by creating new positions or receive support through institutional information technology departments. Either pathway is challenging; people with computer backgrounds typically do not understand the interactions between these systems and the radiology work process and therefore need substantial training on radiology-specific issues. People with radiology backgrounds typically do not have the computer expertise and likewise need additional training. Overall, hospitals have almost undoubtedly underestimated the number and caliber of personnel required, especially at larger institutions. The price of underinvestment is excessive system downtime, disillusionment of physicians, and failure to reap the full benefits of the technology.
Having the appropriate support personnel immediately available in the workplace to troubleshoot and respond to problems is crucial to the successful management of clinical computer systems. There is little tolerance among hard-pressed radiologists or clinicians for anything that slows them down in accomplishing their work. Physicians who previously waited tens of minutes or hours for images to reach them have recast their own work processes and patient interactions in the expectation of near instant access to images. Simply stated, the relentless pace of clinical practice has increased in the digital age; people expect instant access to data and information. Once the digital Rubicon is crossed, there is no quarter given for lapses in technical support.
Another growing and unmet need in the modern digital department of radiology is for personnel with advanced training in image processing. Radiologists and clinicians increasingly rely on three-dimensional renderings and other postprocessing methods to extract the maximum amount of information from image data sets and to consolidate the hundreds of individual sections from computed tomographic (CT) or magnetic resonance (MR) imaging examinations into more manageable numbers. Some image processing applications can take an hour or more of hands-on time, thereby making it unrealistic for either the radiologist or the technologist responsible for data acquisition to perform the work.
Some radiology departments have had to forego the advantages of image processing owing to a lack of either available personnel, institutional support for three-dimensional processing workstations, or the expertise necessary to set up a processing laboratory. Applications that are being underused range from those that are simple (eg, CT urography) to those that are more complex (eg, functional MR imaging). If a large fraction of radiology departments remains without the ability to perform image processing efficiently, then a huge percentage of the potential value of working in the digital age will be squandered.
Optimistically, image processing programs will become increasingly automated in the future, thereby simplifying the various applications and obviating many support personnel. The magnitude of clinical and commercial opportunity that is created by applications like CT coronary angiography (1–4) and CT colonography (5–7) is stimulating intense work in the direction of automation in both academia and industry.
Although only partly related to the digital age of radiology, there is increasing interest in the concept of the radiologist assistant (8,9). This is a controversial subject, with some radiologists pushing to upgrade the skills of support staff and others expressing concern that overly empowered allied health workers, to the detriment of radiologists' own practices, might want to practice independently. These divergent opinions will not be easily reconciled, but the practical reality is that the entire radiology job family is rapidly being redefined in the digital age and more responsibility is being placed on advanced practice technologists and other personnel. Radiologists are expecting technologists to take more responsibility and are therefore turning to technologists and those who have been formally trained as radiologist assistants for help with image processing and selected procedures, as well as for first-line help with image interpretation. The key concept is not historic job titles or descriptions but the training and demonstrated skills to accomplish the job functions. After all, radiologists would still be performing radiography if job descriptions remained unchanged from 100 years ago.
Work Process Reengineering and New Job Responsibilities for Existing Personnel
Digital imaging devices that are coupled with PACS are underwriting new work processes and associated job responsibilities for the incumbent workforce in most areas of radiology. The elimination of cassette-based radiography and hard-copy production are allowing a redefinition of both how technologists perform their work and the level of responsibility. In the case of radiography, instead of having to leave patients unattended to process films and perform quality control functions or consult with radiologists outside the imaging room, technologists need only wait for each image to appear on a screen to assess its suitability without leaving the patient. This transformation is successful in every respect—technologists shed the onerous chore of carrying cassettes, patient safety is improved through better observation, productivity goes up (10), and control of image quality is greatly facilitated. Work process reengineering is almost implicit in the new technology and is probably intuitive to most technologists.
Eliminating the film cassette and film printing has liberated radiology technologists from performing time-consuming menial tasks, which have now been replaced by job functions that require new training in computer skills and that entail more independent decision making and responsibility. The ability to perform quality control in the imaging or control room yields the greatest benefits to productivity if the technologist is empowered to make the decision to repeat examinations as necessary and to sign off on the completion of an examination.
In some areas, the best strategies for reengineering the work process may be less obvious than those provided by the radiography example and may require more analysis (11). A proved approach is to make a flowchart of the existing process and then to create an additional flowchart of the potential alternative processes that take advantage of the new capabilities of improved technology. At Massachusetts General Hospital, flowcharts are used both to identify who is involved in each step of the process and to design the necessary training that is required to successfully complete each step. The aggregate output of this exercise is the creation of a blueprint for managing change that includes an outline of the new work process, a definition of new job responsibilities, and a set of competencies that are required for each participant. The latter is then used to design a curriculum for competency-based training and testing. The time course for training and competency is recorded for each staff member. The flowchart is also useful for assessing bottlenecks and failure points.
Facility Design
Most departments of radiology will not have the luxury of designing brand new facilities to accommodate the needs of the digital age and will face the daunting task of remodeling the existing space to install new digital imaging systems and to create facility designs and infrastructures that optimize the capabilities of these new systems. Little has been written about how to undertake this task; therefore, each institution almost becomes its own experiment, whether it is involved in retrofitting old facilities or starting fresh.
Limitations with regard to the space or configuration of current imaging and procedure rooms are not always obvious until detailed planning is undertaken. For example, rooms designed for CT or positron emission tomography (PET) are tempting to use for PET/CT but are often not suitable because of the longer distance of table travel that is required with the latter device. A critical goal in building the facilities that house expensive high-technology equipment is to design the space and workflow so that high-cost procedure rooms are not used for patient preparation and postprocedure recovery steps, as is common in current practice. Architectural design must support work process efficiency and productivity.
Interpretation areas that were designed for view boxes or alternators typically require modification (12,13) to create the optimum environment for workstation-based interpretation; the lower luminosity of workstation screens compared with that of view boxes requires special attention to background lighting. Radiology practices have also learned that it is desirable to have more workstations than the number of alternators they replace; therefore, the total space needed for interpretation may need to be expanded. Background noise mitigation is highly desirable in areas where voice recognition systems are used. The special computer rooms that are needed to house computer servers for PACS and radiology information systems and the secondary off-site locations that are needed to serve as back-up systems for these functions simply do not exist in older departments.
The sharp increase in interventional procedures is also creating major changes in radiologic facility design that will continue for years to come. As the complexity and number of image-guided interventional procedures increase, the design features of the imaging and intervention rooms in which these procedures are performed must keep pace. Indicative of this challenge, 8%–10% of services performed at Massachusetts General Hospital involving the administration of general anesthetics are now done in the department of radiology. When designing new imaging rooms, the radiology department routinely includes a member of the anesthesiology department on every design team in order to make sure that the new rooms meet the standards for anesthesia and that the rooms allow anesthesiologists enough space and access for safe practice. New rooms are built with operating room surface materials, such as tile walls and terrazzo floors, to facilitate cleaning.
We are now designing a new building at Massachusetts General Hospital that includes substantial space to construct new rooms for image-guided interventional procedures. The space is located with vertical adjacency to two floors of new operating rooms. It is apparent that the border between operating rooms with imaging capability and imaging rooms with interventional capability is rapidly blurring. It may become more relevant to think in terms of level of isolation (ie, clean rooms vs sterile rooms) rather than trying to distinguish between operating rooms and imaging rooms.
Important crossovers are occurring between the design concepts for high-level imaging rooms of the future and those for operating rooms of the future. On the radiology side, we are adopting operating room–level patient monitoring systems and using operating room–style designs to mount equipment from the ceilings or walls, thereby reducing clutter, simplifying room cleaning, and shortening cycle time between examinations. These rooms may be thought of as high technology imaging-based operating rooms.
On the operating room side (14–17), designers are installing large scale high-resolution flat panel displays that are linked to hospital PACS and image processing laboratories so that personnel can view, acquire, and process images during surgical procedures. Both types of rooms will have robust information technology and communications capabilities, including two-way audio and video links and common displays for many kinds of images (eg, radiologic, endoscopic, and conventional photographic images) and other alphanumeric data. Touch panel and voice-activated controls will support improved work processes. The information and communications links will also facilitate new robotics applications. Wireless communications will be used to further reduce clutter and to improve space efficiency.
CT as Exemplar: Facility Design, Work Process Reengineering, and New Job Responsibilities
CT was initially applied for examination of the head, and data acquisition times approached 1 hour or more. The number of examinations per day was limited by the lengthy data acquisition times and by the fact that image reconstruction was performed on the computer overnight. CT was so valuable and scarce that, among other institutions, the University of Michigan Hospital, where I worked in the mid-1970s, by policy did not allow a patient to be imaged unless a neurologist or neurosurgeon had examined the patient first and agreed that a CT examination was indicated. Times have certainly changed! Now, some specialists will not consult on a patient's care until CT or some other high-value imaging examination has been performed and the results are available.
The length of data acquisition and processing in the early days of CT dwarfed the time required to move patients onto the scanning table and to insert the intravenous line or whatever other preparation was required. Because personnel were able to perform only a limited number of examinations per day, the efficiency of the workflow used for handling the patient was not a major focus of attention. The scanning room was used in many departments for whatever pre- and postimaging functions were required.
Even when CT data acquisition became progressively faster, the culture of use, on balance, did not change accordingly. The scanning room was and still is used in many institutions to evaluate and counsel the patient and to insert the intravenous line. CT devices were routinely left idle while technologists printed images and performed data archiving or other administrative tasks. Patients were left on scanning tables while radiologists reviewed the images.
The relative amount of time required for data acquisition versus the logistics of getting the patients fully prepared, moved in and out of rooms, and on and off of tables changed progressively and dramatically as data acquisition became faster. Today with multi–detector row CT devices, the time needed for patient handling and the turnaround time between examinations dwarfs the amount of time required to obtain the image data—a complete reversal of the original situation. Total data acquisition time for an unenhanced head scan obtained with a 64-section scanner is only 2 seconds, and the majority of single body part protocols are less than 20 seconds.
How then do we better use the CT resource? There are undoubtedly dozens of ways of approaching system optimization, but we have adopted a few guiding principles that have helped shape our approach. The first revelation is that there is only one major attribute that distinguishes CT scanners from anything else: CT scanners produce CT scans. Therefore, anything that prevents the CT scanners from being used in their singular role detracts from productivity and should be designed out of the system. An idle scanner might as well be a million dollar boat anchor!
The second revelation is that high quality and productivity should go hand in hand and are best achieved prospectively, not retrospectively. Upgrading the training and responsibilities of technologists, empowering technologists to be the first-line arbiters of scan quality, and providing well-designed work processes and protocols that match CT system capabilities trump quality control by inspection after the images are obtained.
Two different approaches have proved to be the keys to maximizing "beam-on-target" time (ie, exposure time) for CT scanners. First, only the functions that absolutely must be performed in the scanner room are performed there. To achieve this, we have designed each of our primary CT rooms to have a small anteroom equipped with a sink and headwall functions, such as oxygen and suction systems. Patients are evaluated and prepared in this area, which does not block access to the scanning room. A high-intensity light and special "intravenous chair" are available in each preparation area to aid in the insertion of the intravenous line. By the time the patient enters the room, the only steps that are left to perform are connecting the power injector to the intravenous line and transferring the patient to the scanning table.
This design can be applied in other areas, such as in MR imaging and interventional rooms, to reduce the amount of time taken up by noncritical functions. The imaging room of the future will be just that; it will host as little else which is required for the total care process as possible. Surgeons have recognized the same imperatives and potential benefits. The operating room of the future project (18) at Massachusetts General Hospital employs a similar design, with a special preparation area located immediately adjacent to the operating room that allows a surgeon to rapidly stage a series of shorter cases without unnecessarily taking up valuable time in the operating room.
The second part of the strategy for keeping the scanners in use required a redesigning of the work process. A nurse or technologist with skills in intravenous line insertion is assigned to the CT suite to accomplish the preparation steps outside of the scanning room so that the primary technologist is always available to be with a patient inside the scanning room; no scanning time is lost unnecessarily to patient preparation. A further extension of this is having two technologists assigned to each room so that one can be printing films or archiving data while the other is interacting with the patient. Some radiology departments have come to this intuitively, but it is useful to document the scanning process with a flowchart (11,19) to better understand which steps can be performed in parallel and which steps are necessarily sequential. This process, including how work can be divided between two technologists who are working as a team, has been described in some detail by Rhea et al (19).
When this work process and staffing strategy is employed, it may look at first as though it is wasteful because not everyone is busy every minute. While that is true, and even desirable, the benefit is that the process will accomplish far more total work. Personnel costs per shift are higher, but the revenue from increased productivity more than compensates for it; a single additional scan per shift offsets the cost of a technologist or nurse at Medicare fee schedule reimbursement rates (20).
Team strategy and parallel processing obviously make the most sense if an unmet demand is present, whether this demand is in the form of reducing cycle times for inpatients to help decrease the length of hospital stay or reducing waiting times and backlogs for outpatients preparing to undergo CT examinations. If an institution has excess capacity, adding personnel may not make sense but, by the same token, should generate an assessment of whether there are more imaging devices than would be required if CT were more productive.
Redefining the scope of job responsibilities is another important part of the productivity equation. Because of their training and demonstrated ability, technologists at Massachusetts General Hospital are empowered to judge examinations complete, unless alerted otherwise, and are also free to check with the attending radiologist about any cases where they are unsure of image quality. This is a marked change from our historic practice of years ago of having a physician check every image from the examination before the patient was taken off the table. The old practice of quality control by physician inspection added substantial time to the work process, devalued the technologists' expertise and overall role, and did not force radiologists to think as critically about the best protocol because they knew they could always get additional images after their initial review.
Unless institutions in the digital age recognize the interplay between the improved performance of contemporary digital imaging devices and how these advanced devices are deployed, they risk overinvesting in equipment and facilities. Survey data from the University Hospital Consortium (21) appear to indicate that such overinvestment is common. According to the 2001 University Hospital Consortium benchmarking project, 46 institutions reported relative value units delivered per CT device (21) that ranged from 6964 to 99586, with a mean of 44979. This astonishing finding indicates that lower productivity institutions have capital and space costs for equipment that are many fold higher than those of productivity leaders, thereby taking these resources away from their other institutional needs.
Although CT scanning has been used to illustrate a number of concepts, the underlying principles apply across the entire practice of radiology. The magnitude of opportunity will be different in each area and will be influenced in large part by how much performance change there is between different generations of equipment and what training opportunities are available for technologists to play upgraded roles.
Conclusion
The rapidity of change in the digital age of radiology has overtaken the ability of some—probably many—institutions and practitioners to keep pace, thereby leading to a widening gap between those who readily adopt new digital technology, new practice models, and new methods of operation and those who do not. Leaders in the digital age are finding ways to assimilate the costs and complexities of new technology and to harness the benefits of this technology to deliver services of higher medical value and higher quality while also achieving higher efficiency to meet stakeholder expectations. Given the regional nature of the health care system in the United States, late adopters undoubtedly have more latitude to embrace digital technology than is afforded to companies competing in national or global markets, but everyone must also realize that the pace of change is unlikely to slow down.
 |
References |
|---|
 TOPINTRODUCTION References |
|---|
This article has been cited by other articles:
|
|
 |
|
  E. A. Krupinski and M. Kallergi Choosing a Radiology Workstation: Technical and Clinical Considerations Radiology, March 1, 2007; 242(3): 671 - 682. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G W L Boland Government reform of the National Health Service: implications for radiologists and diagnostic services Br. J. Radiol., November 1, 2006; 79(947): 861 - 865. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
