Commodifiable Me: A First-person Account of the Virtues of Imaging Informatics

The Argonauts (and Odysseus, after them) had to sail past rocky islands housing the enchanting Sirens. Their wonderful songs made sailors hurl themselves overboard and swim toward them, even as they died upon the jagged rocks. After my fifth birthday, however, I accepted that plugging my ears never makes bad news go away for long. Can I stop myself from being successfully and wholly commodified? Commodification has pierced our daily workflow: Requests for imaging studies are automated through computerized provider order entry, yet scrutinized at different levels for appropriateness. Protocols are standardized for problem sets, but customized for specific patients. Postprocessing offers an array of presentation displays, yet not all displays are used by all radiologists completing final reports. With standardized reporting, streamlined workflow, and cloud technology, we risk losing value differentials between our work products. Equivalent work products necessarily imply equivalent work producers—and the lowest bidder wins. The injection of informatics into diagnostic imaging implies our willingness to accept a matrix of threats that might ultimately make us expendable. With ubiquitous connectivity, we see empowered consumers taking ownership of their health care, changing the market into one that is both patient centered and patient driven. The Plain Truth Many imaging clinics and health-care enterprises have successfully integrated data-intensive and technology-driven informatics into their workflow in the past decade. Examples include PACS, the electronic medical record (EMR), the RIS, hospital information systems, and telemedicine (including teleradiology) systems. The ubiquity of teleradiology has accelerated the commodification of the radiological consultant—while improving health-care access and expediting patient care. Initially touted as a time-shifting solution for consultant radiologists, teleradiology metamorphosed into a burgeoning market boasting connectivity between specialist providers and specialist consultant radiologists. Increased competition (with reduced reimbursement and the rise of radiology benefit managers) sparked price wars that led to the rise of predatory teleradiology companies. As consultant radiologists, we need to take concrete steps to correct the total commodification of our role as consultants. This is not an insuperable goal: We need to innovate and to favor process over results, at least initially. Innovative changes might or might not affect existing market conditions. Many disruptive innovations in informatics are a combination of evolutionary (made in response to customer needs) and radical (unexpected and extreme). An evolutionary change, for example, would be diagnostic decision support for image interpretation. This requires active engagement of the interpreting consultant radiologist in the EMR. At our facilities, we have embarked on offering further guidance (for studies requested by nonspecialist providers) by recommending surveillance studies at the evidence-based intervals recommended by central bodies. An example of a radical change would be the codification and distribution of appropriateness criteria for studies, especially high-ticket exams. This necessarily mandates interspecialty collaboration. We have successfully implemented diagnostic pathways for stroke and for right upper-quadrant disease (with corresponding correction of the use of multiple studies). Work also is ongoing to reduce the use of studies that employ ionizing radiation in children. This project uses collaboration with providers to choose alternative (or no) tests, along with collaboration with vendors to apply dose-reducing strategies. Here to Serve The RSNA has initiated a comprehensive response to the challenge of patient-centered practice by launching the Radiology Cares campaign.¹ Its mission is to facilitate our meaningful engagement in the patient experience. Disruptive informatics: Use Web-based technology to permit patients online access to schedule, cancel, reschedule, and check in for imaging encounters. Enable patients to receive (on their authenticated smartphones) impressions of final reports—within 24 hours of the appointment—for any study. Continuous enhancement of online offerings keeps the customer base excited. Tip: Several patients with chronic illness, or otherwise in need of recurring appointments for similar imaging encounters, will be early adopters. They will be particularly adept at passing acquired skills to their peers. Trap: High touch always trumps high tech. Machines will fail and need service calls, and will always warrant human redundancy. Elderly patients, those with visual impairment, and those who have low familiarity with touchscreen technology will need staff assistance before fully migrating to automation. Visible Radiology Unless engaged by an interventional radiologist or breast imager, many patients never see the consultant radiologist. Our need to see the patient behind the image should align with the patient’s need to see the author behind every final imaging report. It is impractical and disruptive (in the negative sense) to inform every patient of preliminary results immediately. Anti-informatics: In a study conducted at the McWhorter School of Pharmacy,² patients indicated an overwhelming preference for a pharmacist with a white coat. Appropriate attire is important in creating an image consistent with the public’s perception of health-care personnel, even though it does not correlate with overall satisfaction with care. Integrate the product and the product offering by amplifying every interaction to a professional, personalized level. Front-desk staff and technologists should receive ongoing training in discussing radiation safety and procedural details with patients. They represent us, as our agents. Tip: An uncluttered check-in desk initiates a positive imaging experience. We strive for our check-in area to be not too sterile, but equal parts welcoming and brand promoting. Consultant radiologists’ smiles beam from large, framed photographs, alongside clear posters informing those waiting in line of the other services that we offer. These posters are regularly rotated. We need to tell everyone what we do; there is an opportunity to do so wherever there is a bored captive audience. Trap: Overpersonalization of the check-in desk might be off-putting. Staff members should have no food, beverages, or personal belongings visible at the desk. Farmers’ Market Cost will be the driving force as consumers shuttle between providers to get what they want—at a value adjudged by them as feasible. Disruptive informatics: Price containment for individual imaging encounters is beyond my pay grade, but improving the quality of my work to align closely with the requirements of both providers and administrators is not. Subspecialty distributed reading is de rigueur. The difference between what we do and our actual capabilities (balanced against more training, superior education, and creative scheduling) will magnify our commitment to excellence. Tip: Customized reporting requires effort and ongoing education if it is to engender a high level of customer satisfaction. We conduct regular quality-management conferences to address patients’ and providers’ grievances anonymously: Specific examples can teach us so much, and we suddenly understand something that we’ve understood all along, but in a new way. Learning is a lifelong process—and it’s what we think we already know that stops us from learning more. Trap: Hospital administrators assess concrete factors, such as waiting times for specific encounter streams (for example, diagnostic mammography) and report-turnaround times. Consultant radiologists focus on final reports that advance the patient’s position on a clinical spectrum, on early diagnosis of treatable disease states, and on preclinical detection of iatrogenic problems. It is rather challenging to align processes and results. Repeat Customers Results of the CMS standardized Hospital Consumer Assessment of Healthcare Providers and Systems survey have been publicly reported since 2008 to reflect patients’ perspectives on hospital care. It takes the average respondent about seven minutes to respond to the 27 items in this survey. Medicare is expected to reduce payments to hospitals by nearly $2 billion in the next fiscal year. These withholdings will be redistributed to hospitals with high scores in assessments where satisfaction surveys make up 30% of the total score. Disruptive informatics: Follow a fixed percentage of imaging encounters with meaningful single-page surveys with binary response choices. The more complex the survey is, the less it will be worth. Use the results of the survey to generate key performance indicators. Tip: We employ the Kaiser Member Patient Satisfaction Survey, which returns extremely granular results on an average of 100 surveys per year, per physician. Two lessons learned from this are the need to treat a patient not only the way that you would want to be treated, but also the way that he or she wants to be treated. Prior to every fluoroscopy-guided procedure, I briefly discuss the small risk of radiation and obtain verbal consent to proceed, thus engaging and involving my patients in obtaining a better-quality product. At the start of our workday, we huddle with relevant stakeholders to review the schedule, prepare for special-needs cases, and plan bridge coverage or access during scheduled meetings or off-campus travel. Our concierge service quickly directs providers to the most appropriate consultant radiologist. Trap: Scoring for imaging encounters typically reflects interactions at the front desk and with technologists. We have an interest in keeping our staff members educated, content, and empowered to obtain access to any consultant radiologist. Being accessible via multiple technologies, as needed, is a mixed blessing. Product-development Cycles The final report is our product, and we are only as good as it appears to be. The report’s utility is compounded by the need to satisfy three users: the provider who requests the study, the services broker (third-party insurance carrier and/or health-care enterprise), and the patient. In addition to satisfying regulatory, billing, legislative, and coding requirements, the product has to be all things to all people. Disruptive informatics: As a patient, I want simple answers. As a consultant radiologist, it is beyond indefensible to couch my work in hedges, waffles, and more disclaimers than definitive conclusions. While evidence is lacking to support the superiority of structured, itemized reporting over the historical practice of free-form narrative, the former has the advantage of letting the reader instantly find what he or she is looking for (without a struggle). Impressions, in the final report, should be a précis (not a recital of observations) in language that is lucid and readily grasped by the patient. Tip: Using robust voice-recognition software and reporting templates, we deliver a high-end product with a level of uniformity that aids in the detection of specific problem areas. Impressions (only) are automatically made available to patients using secure access. Trap: Report generation is not static; today, my reports look nothing like they did three years ago. Reporting templates need curating (at least quarterly) to incorporate ongoing changes in our understanding of disease processes, as well as to address specific itemization changes requested by providers. The quality-management service keeps abreast of the inclusion of Physician Quality Reporting System measures, which are tracked on a monthly basis as part of a larger performance-improvement process. The Trust Relationship Local data storage (film jackets, CDs, and magnetic optical disks); dedicated workstations; and paper charts have incrementally progressed to cloud technology, with the notion of putting stakeholders first. This enables the consumer to access imaging data from any point. This ethereal location can be securely accessed by multiple stakeholders, including the buyer of services; this has the inadvertent effect of dislodging the provider–consultant trust relationship from the acquisition-to-interpretation segment of the product cycle. Disruptive informatics: Cloud access allows the integration of value-added services (such as off-site reading, vendor-neutral archiving and transfer, and interspecialty consultation within the enterprise) into the core offering. For specialists working from multiple locations, with variable scheduling and availability, this serves as an elegant way to eliminate delays in information transfer. Tip: Within a closed network such as ours, we can use the cloud to our advantage through interspecialty consultation within the enterprise. When a particularly vexing imaging finding drops me between Scylla and Charybdis, I can always count on my talented coworkers to fish me out of the water. Trap: With an open network, the importance of trust relationships among end users, consumers, and producers cannot be overemphasized. For example, given a choice between airlines with hard products that are equal in every way (including price), consumers are likely to let the soft products (courtesy, accessibility, on-time service, and follow-up) drive the decision. Radiation Overexposure Media attention to skyrocketing volumes for studies using ionizing radiation has percolated into the collective consciousness of our patients, who demand to know more. Fear of radiation poisoning is amplified because radiation is insidious (being colorless, odorless, and tasteless). Disruptive informatics: In California, it is mandatory to document radiation dose in every final report generated for a study using ionizing radiation. Notwithstanding the controversy pertaining to the choice of dose-monitoring indices, it is time for the unification of exposure indices across different vendors and conditions of calibration, with automatic uplinks to update a record of the cumulative doses incurred. Tip: Whenever CT technology is used, structured reporting (using templates) force fills fields with documentation of the dose–length product and volumetric CT dose index by default. These data are also screen captured as part of the permanent medical record, in accordance with regulation. Trap: More systematic work needs to be done by governing agencies. Much controversy exists as to the meaningful use of dose–length product and volumetric CT dose index, which use phantom estimates and are not truly representative. External calculators must be used to determine actual effective dose per patient. Vendors are actively pursuing automated solutions that not only generate these data, but permit algorithms to autopopulate final reports. There is a need for a widely accepted, easy-to-use national dose registry similar to the Multidisciplinary European Low Dose Initiative (MELODI), implemented in the European Union in 2010. The ACR^® Dose Index Registry, which lets organizations compare their anonymized dose information (by study type and/or body area) with that of other facilities, is not intended as a patient-specific repository of dose data. California has mandated dose reporting, but this information is not necessarily available in other states. There is no uniformity, and more legislation is needed. The European Union is not yet ready for universal dose reporting, but MELODI member states are leading the effort. We can’t extrapolate their standards to our patient population, however, because we have been using higher doses for our larger patients, and we also use CT more than many Europeans (who often prefer sonography or MRI) typically do. Instead of despairing over informatics having disrupted the lives that we lead, we should repurpose informatics to achieve our own targets. This is a domain-specific challenge, and we should equip ourselves with tools that help us master it. When Orpheus heard the songs of the Sirens, he simply used his lyre to make music that was louder and more beautiful; his music drowned out the bewitching songs, and the Argonauts sailed past the craggy isles unharmed. Let the music play.

Sundeep Nayak, MD, is a consultant radiologist practicing nuclear medicine and neuroradiology in Hayward, California, with the Permanente Medical Group, Inc. He oversees radiation safety and imaging quality management in the Greater Southern Alameda area for the Kaiser Foundation Hospital, Inc, in Northern California. The views stated here are those of the author and not necessarily those of the Permanente Medical Group, its parent, its affiliates, or its subsidiaries.