|
MARKETWATCHTechnology And The Boundaries Of The Hospital: Three Emerging Technologies
U.S. hospitals have proved remarkably adept at altering their service offerings to incorporate new technologies. New technologies threatened to undercut hospitals’ central role in health care delivery in the 1980s. An array of new technologies promise yet again to alter the boundaries of hospitals’ franchise. These technologies will not only continue the shift away from acute, inpatient care that we have seen for the past thirty years but will also challenge hospitals to collaborate more effectively with physicians and technology developers. How hospitals and policymakers respond to these emerging technologies will help determine whether hospitals remain at the center of the U.S. health system.
As Rosemary Stevens explored in her marvelous history, In Sickness and in Wealth, U.S. hospitals in the early and mid-twentieth centuries successfully exploited new technologies to expand and strengthen their franchise.1 The advent of anesthesia, surgery, infection control, and radiological diagnosis enabled hospitals to morph from a largely custodial role first to diagnostic and then to curative institutions. These technologies broadened the hospital’s influence and its economic footprint. Physicians who wanted access to these technologies gravitated to hospitals as practice sites. During the 1980s a wave of new technologies posed a major threat to hospitals.2 Powerful new imaging tools, such as magnetic resonance imaging (MRI), ultrasound, and fiberoptic scopes, enabled physicians to detach large amounts of diagnostic and therapeutic work from hospitals and move them to sites they controlled. Combined with important payment changes such as diagnosis-related groups (DRGs), these technologies helped reduce hospitals’ lengths-of-stay and, less expectedly, hospital admissions. Hospitals responded to these new technologies as they had to earlier technological innovations: by incorporating new ambulatory services into their product offerings. They developed ambulatory surgery and imaging facilities both on and off campus, seeking to stanch the outflow of these profitable services. The result was a pronounced shift in hospitals’ service mix. Ambulatory services, which accounted for only 13 percent of hospital spending in 1980, represented more than 37 percent in 2002. Inpatient census fell more than 20 percent, and ambulatory services became the hospital’s most rapidly growing service.3 However, hospitals were only partially successful in defending their franchises in the face of less invasive surgical and imaging technologies. For example, hospitals controlled barely half of MRI procedures at the beginning of the 1990s and were able to grow their share during the decade only modestly.4 In computed tomography (CT) scanning, hospitals lost market share steadily to nonhospital sites during the same decade.5 Finally, hospitals’ ambulatory surgical volume leveled off in the late 1990s, despite continued rapid growth in overall ambulatory surgical volume.6
Looking forward twenty years requires more than a modest suspension in disbelief. However, it is possible today to glimpse emerging technologies that may further reshape the service offerings of the nation’s hospitals and challenge our health care payment system. If the technologies mentioned earlier emerged from surgery, imaging, and subspecialty internal medicine during the 1980s, the new technologies will emerge from the clinical laboratory and pharmacy. The three technologies discussed in this paper were chosen either because they contain the most important potential for reducing hospitalization or because they create new service opportunities for hospitals. Personalized medicine. Genetic profiling for targeting drugs. The Human Genome Project has inundated pharmaceutical and biotechnology companies with promising new therapeutic targets.7 However, it has also provided new knowledge about how better to target existing drug treatments. Almost two million people suffer from adverse drug reactions each year, tens of thousands of which are fatal. There is encouraging circumstantial evidence of a genetic basis for adverse drug reactions.8 Scientists and diagnostic companies are focusing increasing attention on the subset of human genes that influence how drugs are metabolized in the liver. The first genetic assay for these key genes, a microchip-based assay for CYP 450 (AmpliChip), was released for experimental use in mid-2003 by Roche Laboratories and subsequently recalled for further testing by the U.S. Food and Drug Administration (FDA).9 It may be possible to identify through genetic testing not only people who may react poorly to drugs but also those for whom the drugs have no therapeutic benefit.10 The former information would be of interest to drug companies, which could limit their liability and preserve their research investments by limiting their exposure to people with genetic risks of poor reactions. The latter information would be of interest to health plans, government payers, and hospitals eager to contain drug spending, because it will help avoid wasting drugs on people who will derive no therapeutic benefit from them. Genetic diagnosis, testing, and therapy. Treatment of infectious diseases and cancer is being transformed by genetic diagnosis.11 It is possible today to identify specific genetic types of HIV present in a patient with AIDS, which predict the virus’s resistance to the various antiviral agents used in cocktail therapy. By targeting drugs to viruses based on their resistance profile, it is possible to clear the virus from the body far more efficiently and quickly.12 Similarly, with cancer, genetic profiling can help identify specific genetic variants of cancer cells, which may either resist or succumb to various chemotherapy options.13 As a consequence, nucleic acid testing is the most rapidly growing segment of the in vitro diagnostic laboratory business. Spending for DNA testing exceeded $1 billion in 2001 and is estimated to be growing 30–35 percent a year.14 Genetic testing will prove to be a powerful lever in increasing patient safety and clinical effectiveness in hospitals. Payers, including Medicare and Medicaid, have an incentive to encourage its use to reduce costly adverse drug reactions and drug-related "outlier" cases. Looking further ahead, as more becomes known about the genetic variability of human disease, it may be possible by 2015 or after to craft customized immune therapies or vaccines for a person’s specific infection or cancer.15 Whether personalized medicine will become literally "individualized" medicine is still highly speculative. Drug company executives dismiss the potential for "individualized" medicine, citing safety and, therefore, regulatory concerns. Role for hospitals. If genetic selectivity enables individualized therapy, there would be a large hospital service component focused on genetic variation in the pathogen at the root of a person’s disease, as well as a person’s immune competence. Hospitals and large, regional multispecialty groups such as the Mayo Clinic and Memorial Sloan Kettering Cancer Center could play a growing role in crafting custom therapies for infectious agents and genetically based disease. Depending on how this technology evolves, there could be important boundary issues between the highly subspecialized hospital/multispecialty clinic and drug company economic spheres. If "home brewing" of therapeutic agents becomes both technically possible and affordable, large pharmaceutical companies may find that their distance from the care process handicaps them in competing with care providers in personalized medicine. It is not beyond reason that as radiology morphed from a diagnostic to an interventional discipline, so too might the clinical laboratory, in collaboration with hospital pharmacies, someday morph from a purely diagnostic resource to the source of new therapeutic tools to address the genetic variation in disease. Because they can focus capital and program spending on emerging service opportunities, hospitals seem uniquely positioned to shape this emerging service. Although large pieces of laboratory analysis have migrated to high-throughput national laboratory companies, surgical pathology has remained the province of hospitals because of its tight temporal linkage to direct patient care. If there is strong linkage between genotyping of pathogens and direct service provision, large national clinical laboratories will face major constraints in controlling the growth of these applications, and hospitals may find themselves at the leading edge of a high-leverage new business. Regenerative medicine. Culturing and grafting human cells. Another important service opportunity for hospitals will emerge from the ability to replace or repair tissues damaged by trauma or illness. Remarkable advances have already been made in culturing and grafting human cells to repair burn damage.16 Cosmetic and maxillofacial surgeons are already experimenting with prosthetic ears, noses, and other organs, and the ability to replace ligaments, tendons, cartilage, and bone seems within reach for orthopedic surgeons as well.17 The next few years will see a growing complement of specialized tissues, both artificial and cultured, for musculoskeletal restoration. Stem-cell research. However, the Holy Grail in restorative medicine is the ability to culture a person’s own cells to create replacement tissues for repairing internal organs damaged by trauma, disease, or aging. Understanding the biochemical basis for tissue growth and repair may emerge from research into the growth and maturation of human stem cells. Stem cells contain the genetic instructions for growing into mature cell lines that make up more than two hundred types of human tissues. As we have all come to learn, the embryonic stem cell contains the capability to differentiate into all of these tissue types. Understanding how to grow stem cells into replacement tissues could lead to restorative therapy for spinal cord injuries, strokes and myocardial infarctions, Type I diabetes, and degenerative diseases of the nervous system such as Alzheimer’s and Parkinson’s diseases, as well as a host of other now irreversible clinical conditions.18 As biologists gain a better understanding of the chemical signals that guide stem-cell differentiation, clinicians will eventually be given the means to culture a person’s own cells into replacement cells that can be used for restorative therapy. Stem cells also contain, in thousands of genes in some cases, the "assembly instructions" that turn those tissues into functioning organ systems. It is not inconceivable that in twenty years, clinicians may be able to direct the creation of whole, functioning, biocompatible human organs from a culture of one’s own cells, enabling transplantation without resorting to immune suppression. Costly immune-suppressing drugs, and managing the inevitable infections that occur when the immune system is suppressed, add dramatically to the lifetime cost of transplantation. Role for hospitals. Regardless of which pathway is followed, hospitals are likely to play a decisive role. Flow cytometry, a remarkable tool that can sort and capture individual cells in a blood sample based on patterns of receptors on the cell surface, or even specific DNA sequences present in the cell nucleus, will be the key tool clinicians use to find and isolate human cells for cell culture and tissue development. Cell culture is now done in hospital clinical laboratories for diagnostic purposes. It seems logical that if it proves technically feasible, therapeutic culturing of human cells through nuclear transplantation will also take place in the hospital’s clinical laboratory, although it could conceivably be outsourced to firms created specifically to do this type of work. Although stem cell–related research is still early in the discovery process, progress in tissue engineering appears promising enough to suggest that some hospitals will offer restorative medicine as part of their service offerings, perhaps within the decade.19 It is reasonable to speculate that restorative medicine may eventually become the marker for "tertiary" medical care in hospitals a decade or two hence, as open-heart surgery did in the 1970s. Remote patient monitoring. A third rapidly emerging area of service opportunity is in remote monitoring of patients with some degree of unstable clinical risk. An early forerunner was remote cardiac monitoring, where electrocardiogram (EKG) readings were transmitted through dedicated telephone lines to cardiac intensivists from remote locations such as small rural hospitals. This monitoring extended the reach of tertiary medical facilities. When wedded with helicopter transport, cardiac telemedicine enabled cardiac patients in rural areas to enjoy a greater measure of safety from heart attacks. These technologies have been refined so that it is possible to manage large numbers of intensive care unit (ICU) patients through remote locations using broadband Internet connectivity. These technologies are already in use in both teaching and community hospitals around the country. Remote ICU systems integrate voice, visual images, telemetry data, and summary patient history information, all in digital form, and enable intensivists to monitor dozens of patients—perhaps more than a hundred—in remote ICUs, by coordinating the activities of on-site ICU nursing teams.20 These same technologies will not only enable patient management in a wide range of hospital inpatient units, including postsurgical recovery, coronary care, and twenty-three-hour observation beds attached to emergency rooms. They will also enable large groups of patients who are now hospitalized to be monitored in real time in the home and community. Sensor monitoring. Several technological advances will dramatically expand the capacity to monitor at-risk patients from a distance in the next three to five years. One is the rapid growth of sensor technology. A remarkable range of physiological conditions can be monitored in diverse, nonhospital settings. These include heart rate, temperature, respiration, blood pressure, blood oxygen, perspiration, physical location (for example, using global positioning system, or GPS, technology), physical orientation (standing, lying down, and so forth), and movement. These sensors can be imbedded in wristbands or articles of clothing or in the living surroundings of the person at risk. They can also be implanted in intelligent implantable devices such as the pacemaker, defibrillator, or insulin pump. Wireless broadband. A second vital technology is the growing availability of wireless broadband connectivity. Wireless broadband enables signaling from wherever the at-risk person may be to a monitoring station. The bandwidth also enables multiple channels of communication, including physiological data, voice, and video images, which can be integrated into a "dashboard" for the monitoring health professional. Clinical information systems. A final technological enabler will be the intelligent clinical information system that integrates all of these diverse channels of input and connects the patient to the clinical care system. These systems will contain not only decision-support tools to help organize the health system’s response to the patient’s unique risk, but also screen digital signals from the patient that will prompt a response from the care team when a monitoring "threshold" (such as unstable heart rhythm, cessation of breathing, or falling) is reached. Clinical personnel cannot be expected to be watching screens constantly. Intelligent clinical software will help them focus on the subset of patients needing some type of intervention. Role for hospitals. Different monitoring companies are building their own software and monitoring stations and are connecting patients to their "power user"—for example, pacemakers to cardiologists, breathing monitors to pulmonologists, and so forth. Eventually, these devices will proliferate so that centralized monitoring is the most sensible economic solution. At this point of transition, hospitals could play a major role. Eventually, hospitals will find themselves connected in real time to hundreds or perhaps thousands of potential or former patients through wireless networks. For the next few years, monitoring-system vendors will be selling their products to hospitals to attempt to remove unstable but not acutely ill patients from the inpatient setting. The economic justification will be the lowering of lengths-of-stay in ICUs, coronary care units (CCUs), telemetry units, and emergency department observation beds and the avoidance of capital expenditures to construct additional bed capacity. These devices will help extend scarce critical care nurses and intensivists and enable more precise and thoughtful matching of a patient’s acuity level to the level of care or service provided. Many patients who are hospitalized today will find themselves returned to their lives and work at the end of a friendly and unobtrusive electronic tether. Policy And Payment Issues For personalized medicine. If the broad-scale application of pharmacogenomics to identifying risks of drug therapy proves fruitful, clinicians and hospitals may be required by conditions of drug approval to screen potential patients prior to commencing therapy for certain drugs. Accreditation standards for hospitals, physician groups, or health plans could require surveillance of genetic risk factors as part of the prescribing process. This will become easier with computerized physician order entry (CPOE), since much of the checking could be automated and the decision-support guidelines referenced above could be hard-wired into the prescribing process. For this reason, genetic information will likely become a vital component of a person’s medical record. The tightening linkage of genetic information to patient safety will affect, first, medical practice norms and then medical liability standards. Practitioners or institutions that make prescribing or other therapeutic decisions without taking genetic risk factors into account will be exposed to potential liability. The use of genetic information in clinical care will heighten consumers’ concerns about the privacy and confidentiality of their medical information and reinforce the importance of the Health Insurance Portability and Accountability Act’s (HIPAA’s) medical privacy provisions. Consumers will be unlikely to submit to genetic testing if they do not feel that the resultant information is safeguarded by hospitals and clinicians and used only by those entrusted with their care. Following this logic, tighter licensure and competency evaluation of hospitals as well as clinical laboratories and pharmacies involved in setting the therapeutic mix or potentially crafting personalized therapies may be the appropriate mechanism for assuring quality and patient safety. Hospitals, multispecialty clinics, and other providers could be specially certified to provide certain forms of personalized therapies, broadening the locus of regulation from the manufacturers of drugs to those who create the therapies. How payers would pay for personalized medicine is still highly speculative. Genetic testing relies increasingly on the use of costly microarrays—gene probes wedded to micro-chips, which can generate hundreds of thousands or even millions of data points. As manufacturing technologies improve, the cost of genetic information is likely to fall, perhaps dramatically, as the cost of micro-chips has done historically. As costs fall, health insurers will probably apply pressure to providers to perform safety-related genetic testing for high-risk drug therapies. At the same time, they will face political restrictions that will narrow how the information is used, to assure that people with genetic risks are not subjected to job discrimination or loss of benefits. Combination therapies are the standard method of treating HIV and some forms of cancer today. Much of personalized therapy will likely be administered through infusions or combinations of infusions and pills, most likely in a clinic or other outpatient setting, although some small-molecule therapies may be administered through patches or inhalers. The ability to set therapeutic dosages more precisely and exclude ineffective or damaging drugs from the therapeutic cocktail could save money and shorten the course of treatment, heightening payers’ interest in the process. Minimizing hospitalization, avoiding complications, and avoiding wasting drugs and other therapeutic tools would be rewarded by higher margins of cost versus payment. These opportunities suggest that paying hospitals or physicians for personalized medicine through some type of age- and severity-adjusted case For regenerative medicine. Hospital clinical laboratories could be pivotal not only in isolating stem cells and other human cells but also in stimulating them to grow in mass and functional characteristics, to enable them to supplement tissues or repair organs damaged by disease. From a payment standpoint, there will be advantages to encouraging this type of therapy as an alternative for organ transplantation where possible, particularly if the need for lifetime immune suppression is reduced or eliminated. Conceptually, tissue replacement would be treated similarly to a surgical procedure or infusion therapy, depending on the mechanism of access to the site of a lesion or damaged organ. Regenerative medicine is likely to evolve as a service, custom-built around the point of care, which gives hospitals and regional or national multispecialty clinics an inside track in developing it. It is worth noting, however, that many of the same therapies that apply to disease or trauma could also be available to restore damage from aging. This will probably create a huge elective-care market similar to today’s cosmetic surgery market. How payers choose to limit application of regenerative therapies, and how to split costs with patients, will be an interesting policy and business challenge. For remote monitoring. How the health care system pays for remote monitoring will also be an interesting challenge. Clearly, family members whose time is freed up from twenty-four-hour caregiving will have an economic motivation to pay for monitoring devices and services for unstable relatives with definable health risks. There is a large hidden cost of "informal" care systems for chronically ill or unstable patients that does not get recorded in our National Health Accounts. However, if vendors and care providers can demonstrate reduced hospital admissions or emergency room visits for patients with unstable cardiac rhythms, breathing problems such as asthma, diabetic shock, and other manageable conditions, there will be a strong case for both public- and private-payer coverage. For monitoring that takes place inside hospitals, this technology may make intensive care and postintervention observation of patients less expensive and more systematic. This would be achieved by spreading the fixed cost of monitoring stations, software, and operators over a larger number of patients. Reducing variations in resource use and in patient safety represents major opportunities for savings in the health system. With larger numbers of frail, chronically ill patients in the health system, it will make sense for those who pay for care to encourage the substitution of systems and machine intelligence for chronically scarce, trained clinical person power. Real-time monitoring could also lead to continuous and highly effective disease management for these patients, creating major health and economic benefits for society.
Hospitals will probably experience continued erosion of profitable ambulatory volume and continued shortening of lengths-of-stay over the next decade. Two decades ago the idea of operating on a beating heart without using a heart/lung machine seemed utterly incomprehensible. The continued development of less invasive imaging and surgical modalities will further reduce lengths-of-stay, complication rates, and recovery times for ever more complex services. Joint replacement and spinal fusion are two emerging candidates for near-single-day hospital stays, heralding a further shrinkage of traditional acute care patient volume.21 The new technologies discussed in this paper have several things in common: large, fixed-cost expenditures to acquire the technologies; a need for highly trained clinical operators; and the need to integrate a clinical response across specialties and geographies. These costs make sense only if they can be spread over a large base of potential patients/ users. These technologies continue a pattern of shifting the focus of the health system away from expensive acute medicine toward a medicine more attuned to managing clinical risks. As such, they seem likely to reduce, rather than increase, the demand for inpatient services, the hospital’s traditional core product. Depending on how hospitals assess the capital risks, however, these new technologies could nonetheless strengthen the hospital’s importance in the care system, not diminish it. If they display a defensive, turtle-like aversion to technological innovation, however, hospitals could see major pieces of these new franchises disappear into new investor-owned or physician-controlled service companies. Hospitals have historically responded with alacrity to new technologies that enhance their service capabilities and enable them to attract and retain physicians who use the technologies. Physician entrepreneurship and the availability of equity capital have complicated this process by enabling physicians to isolate and capture new technologies in settings they control. The emerging technologies may reverse this dynamic somewhat because they rely on costly equipment that will require high clinical volumes to amortize acquisition costs. These technologies cannot be fully exploited without a clinical team to shape the therapeutic response, something hospitals have historically been able to do with their multidisciplinary clinical workforce. Hospitals that can collaborate effectively with their physicians could leverage these technologies not only to grow their franchises but also to generate real health and cost benefits for society. Alternatively, hospitals could see a new wave of defections of physicians and patients from their systems by failing to invest aggressively and collaborate effectively with their physicians. That so much of the outcome may depend on the quality of hospital leadership makes predicting the ultimate impact on the hospital itself exceptionally difficult. Policy-makers should seek to encourage responsible, site-neutral experimentation with these new technologies and devise responsible payment incentives to encourage the use of technologies that reduce risk to patients and save lives.
Jeff Goldsmith (hfutures{at}healthfutures.net) is president of Health Futures Inc. and associate professor of medical education in the School of Medicine, University of Virginia, in Charlottesville. The author acknowledges Anita Gupta for her valuable research assistance.
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Three New Technologies
Impact On Hospitals
