The Two-edged Sword: How Technology Shapes Medical Practice
Gregory A. Patton
Technology is assuming an increasingly important role in medical practice and health care delivery, fueled by forces such as uncertainty; variability, error, and quality problems. While the benefits of technology are obvious, there are insidious costs that are harder to discern. Technology has a significant, but less appreciated, role in imposing standards and constraints upon medicine. These ancillary effects account For some of the physician reluctance to embrace technological innovations perceived as controlling. This article explores technology’s wide-ranging effects in shaping medical care delivery. Technology is not a passive servant of the health care delivery system, but rather acts as a catalyst and shaper of that system. In the process of becoming more technological, medicine has been transformed from a profession with unmatched sovereignty into an industry shaped by technology amidst a context of social and political forces.
* Health Information Systems
* Medical Informatics
* Practice Guidelines
* Technology Shaping Medical Practice
* Physician Reluctance to Embrace Technology
“TECHNOLOGY… IS A queer thing. It brings you great gifts with one hand, and it stabs you in the back with the other.”
The New York Times, March 15, 971.
Technological advances have undoubtedly improved medical practice and health care delivery. Perhaps less noted is that technology plays a role in imposing standards upon the practice of medicine. Physician reluctance to embrace technology may be due to perceived constraints on autonomy as a necessary byproduct of embedding technical systems in medical care.
What constitutes technology? The dictionary provides a staffing point: “The application of scientific knowledge to practical purposes, as in a particular field,”  or, in the context of medical practice, the application of knowledge to preserve and restore health. Willoughby defines technology as “an ensemble of artifacts intended to function as relatively efficient means.”  From this perspective, technology embodies a collection of skills and specialized equipment focused on the delivery of health care in an effective and efficient manner.
Rapp provides an alternate view: “Technology… aims at the domination of nature through transformation of the outside material world.”  In this context, the essence of technology is control: mastery of scientific principles, manufacturing processes, and/or human behavior.
Technology exists as a complex interwoven network built one layer upon another as new developments depend upon previous ones. The overall matrix of innovations, rather than any single technology, defines health care delivery. Modem medicine rests upon a confluence of technological systems: the operating room, clinical laboratory, radiology department, and radiation oncology facility each incorporate an interrelated network of technologies.
Technology as shaper of medical care delivery
The health care delivery system has been promoted and shaped by technology. News reports trumpet medical innovations and treatments daily. Technology also creates new medical diagnoses by expanding the range of possibilities. Electrocardiography and echocardiography allowed the characterization and classification of multiple cardiac conditions, including atrial fibrillation, heart blocks, and mitral valve prolapse. Before these diagnostic tools were available, life-threatening coronary artery compromise and silent myocardial infarctions often went unappreciated. Mammography led to recognizing ductal carcinoma in situ of the breast. Prostate cancer commonly lurked unsuspected only to be noted at autopsy (if at all) before the prostate specific antigen (PSA) test and transrectal ultrasound-guided biopsies allowed early diagnosis.
Just as it has expanded the diagnostic spectrum, technology has also given rise to a host of new therapies, including transplants, renal dialysis, minimally invasive endoscopic surgeries, and prosthetic joints.
Such remarkable advances come at a significant expense. Technology is the major cost driver and accounts for over half of the increase in medical expenditures over the last 50 years.  Technology has increased reliance upon formal organizations and equipment, rendering the profession answerable to whoever provides the capital investment. Few physicians are able to practice without the technological support provided by hospitals, laboratories, and imaging centers and by other specialized practitioners. True physician autonomy is rare.
Given the magnitude of health care costs, few patients are able to pay their medical bills. In 1995, only 21 percent of health care expenditures were paid by patients, with the remainder covered by third party payers, such as government or private insurers.  Fiscal intermediaries influence which medical services are reimbursed and to what degree. The reimbursement system has an affinity for medical procedures, which tend to be more technological, rather than cognitive services.
This bias creates a feedback loop (please see Figure 1): technology facilitates medical advances and especially procedures that are reimbursed, which in turn stimulates further innovation. Technology is not a passive servant, but rather acts as a catalyst and shaper of the health care system. In the process of becoming more technological, medicine has been transformed from a profession with unmatched sovereignty to an industry shaped by technology amidst a context of social and political forces. 
Forces promoting technology’s application to medical care
Four phenomena demonstrate inefficiencies that promote the increasing use of technology in medicine: uncertainty, variability, error, and quality problems.
1. Uncertainty–Despite the large body of scientific information and evidence upon which modem medicine is based, much remains uncertain or unknown. Physicians routinely process limited information and deal with uncertainty as they diagnose and treat. As observed by Eddy, “Uncertainty, biases, errors, and differences in opinions, motives, and values weaken every link in the chain that connects a patient’s actual condition to the selection of a diagnostic test or treatment.” 
The sources of randomness are multiple:
* For many diseases, debate exists about what constitutes a confirmed diagnosis
* Even when precise diagnostic criteria exist, physicians vary in their application of these criteria
* The clues upon which we base many diagnoses can be subtle
* Many of the symptoms, signs, and conditions defining a given disease are extremely common
* The division between “normal” and “abnormal” is blurred
* The natural history of disease is largely unpredictable in a given patient.
Uncertainty pervades medical practice, leaving ample room for differences of opinion: “The physician labors in a sea of uncertainty.”  Such underlying uncertainties require that physicians use subjective judgment. [12,13,14] Unsurprisingly, the result is variability in medical care.
2. Variability–The practice of medicine involves a complex web of different disease pathways and outcomes, amidst which physicians show dramatically variable practice patterns. [12,13,14] Substantial therapeutic differences exist for seemingly similar patients with a given diagnosis and comparable seventies of disease. Treatment of Medicare beneficiaries varies considerably across the United States:
* Coronary artery bypass surgery varies by a factor of 4
* Coronary angioplasty varies by a factor of 7.8
* Radical prostatectomy for prostate cancer varies by a factor of 11
* Breast conserving surgery for cancer varies by a factor of 34 
The recognition of variations in medical care may be attributed to two phenomena. The emergence of third-party payers (and particularly government programs such as Medicare) spurred the development of claims processing systems. Coupled with the appearance of robust computer systems, the capability to not just pay medical bills but to examine and characterize the details attendant with such claims allowed the identification of patterns and variations in care. Such variations translate into large differences in expenditures for third-party payers. Having played a seminal role in identifying treatment variability, computer systems are being incorporated in the ancillary role of controlling physician behavior and reshaping it into desirable channels to curtail health care expenditures.
The root cause of variability is the lack of consensus on the “correct” way to practice medicine.  Differences arise from one geographic area to another as standards within a single area narrow the range of acceptable therapies, with local physician choice accounting for the selection between alternatives.  The problem is not that one community harbors “better” or “worse” doctors compared to another, but that we lack a scientific basis for many of the therapeutic choices made in the course of caring for patients. Evidence-based medicine (EBM) is an attempt to integrate best evidence with clinical expertise, pathophysiological knowledge, and patient preferences.  EBM is an incomplete solution to the problem of variability given uncertainty: the evidence of which treatments are best is often fragmentary, contradictory, or lacking altogether.
3. Error–Humans are inherently imperfect, and error is a frequent companion of medical care, a fact highlighted by the recent Institute of Medicine Report. Although most medical errors do not cause injury, treatment-related injuries are common among hospital inpatients.  The Harvard Medical Practice Study found an injury rate of 3.7 percent among hospitalized patients, of which 13.6 percent proved fatal.  This contrasts with other industries where lower failure rates are the norm.  A paradox exists: “although the standard of medical practice is perfection–error-free patient care–physicians recognize that mistakes are inevitable.”  Tellingly, a term exists to describe injuries resulting from physician error: iatrogenic.
The prevailing school of thought dealing with medical error has been the perfectibility model: if physicians and nurses were properly trained and motivated, they would make no mistakes. Training and punishment have been the methods used to achieve this goal. However, Eddy notes that “the complexity of modern medicine exceeds the inherent limitations of the unaided human mind.”  Leape stresses that instead of the reactive approach to errors, a more productive preventive approach focuses on root causes of system errors. “Physicians and nurses need to accept the notion that error is an inevitable accompaniment of the human condition, even among conscientious professionals with high standards. Errors must be accepted as evidence of system flaws, not character flaws. ” 
Systems should make it difficult for individuals to err, but recognize that errors will inevitably occur and plan for their recovery. Design strategies that harmonize with this theme incorporate reduced reliance on memory, improved information access, and standardization. The salient point is the potential for health information systems to support these strategies.
4. Quality problems–Health care quality is intimately connected to problems of uncertainty, variability, and error. Initially a major focus in industry, quality assurance and improvement have assumed increasing importance in health care as lapses in quality have been demonstrated. [22,25] Given the complexity of modern medicine, it is unlikely that quality can be meaningfully improved without creatively applying the powerful capabilities of technological systems. [26,27,28]
These four forces are eloquent indicators of inherent problems in medical practice. Such inefficiencies are catalyzing an intellectual revolution in medical practice away from the narrow application of raw clinical judgment and qualitative reasoning to a more scientific and quantitative mode of practice. [28,29,30] This new approach will further cement health care delivery’s dependence upon technology through information systems.
Standards represent one step in the direction of shaping physician behavior. As the complexity of technology increases, there must be agreement in terms of how components relate to each other. Standards undergird our entire existence in the technical milieu. Standards define which side of the road we drive on (at least within a given country); what significance red, yellow, and green traffic lights have; how telephones hook into the wiring network; and how a credit card number issued by a firm in New York is accepted across the country or even abroad. Standards ensure compatibility of technical systems– without them, chaos reigns.
Standards pervade and facilitate medicine. Needles, catheters, sutures, and other medical equipment employ common systems of nomenclature and sizing. Laboratory tests, pulmonary function tests, radiation doses, and medications are each measured against common standards. Such medical standards synchronize and harmonize medical care, promoting efficiency and making high quality medical care possible.
Technology’s role in imposing standards in medicine
Medicine has undergone a fundamental shift since World War II– once a profession founded in art, medicine is now based on science, while preserving an important element of art. The idea of a “scientific medical practice” became intertwined with the idea of standardization; indeed, standardization emerged as the sine qua non for the development of this new science. 
Medical practice standardization was not warmly received by physicians, given the conflict with their autonomous self-image. Standardization was suspect and considered more beneficial to hospital administrations, insurance companies, and government agencies, threatening the physicians’ valued and fiercely defended professional sovereignty.
Yet, despite strong reservations, standardization’s intimate connection to reimbursement systems has facilitated its imposition. By accepting the benefits of third-party reimbursement, physicians have, in the bargain, been unable to avoid the attendant standardization typified by CPT and ICD-9 coding, DRGs, and Evaluation and Management (E&M) definitions. Compliance has burgeoned as a major theme of the health care infrastructure resulting from these increasingly rigid rules.
With the advent of complex standardization and coding schemes and with the transformation of medicine into a discipline measured by scientific yardsticks, it has become increasingly difficult for the unaided practitioner to keep abreast of both technological developments and regulatory guidelines. Computer-based tools were designed to aid physician decision-making processes. [26,27,32,33,34]
Augmenting such technological tools were structured methods specifying optimal paths of action. Variously termed protocols, clinical pathways, care paths, practice policies, or algorithms, the intent was the same: to guide–and control– physician behavior. The protocol and its brethren share a common thrust in shaping medical practitioner action in specific situations. These measures find powerful application in disease management initiatives within computerized health information systems  and in expert systems that emulate human problem-solving abilities. Through such vehicles, technology is cast in the role of instigator, definer, and enforcer of standards in medicine. Technology’s transformation has been substantial– from its humble origins as a servant to medicine, to a benign but active role as a shaper of medicine, to a less benign or even malignant role as an enforcer of medical practice.
Figure 2 illustrates Willoughby’s definition of technology from the medical perspective. Medical ends are of two types, intended and collateral ends incorporating unintended and/or unanticipated consequences, such as iatrogenic injury. Medical technology has achieved great advances in patient care, although at the expense of initially unanticipated effects, including greater control over medical practice. An important collateral end of technology in the medical domain– a goal usually unstated–is to control physician behavior.
Technology’s impact on decision-making
Medical decision-making may be illustrated graphically by plotting medical decisions along a continuum of relative entropy (please see Figure 3). At the origin on the left is optimal medical care, for example, the minimum amount of entropy or disorder possible, a point where a given treatment conforms exactly with best practice (colloquially known as the “gold standard”). As treatment migrates to the right away from this origin, entropy increases. Historically, prior to the era of scientific medicine, concepts such as uniformity and standardization were seldom considered: virtually any medical care was deemed acceptable.
Only with the advent of scientific medicine did a normative framework emerge allowing judgment of the quality of medical care. Figure 4 divides the medical decision-making continuum into specific zones. Near the origin, considerable conformity exists between treatments: this region is termed the “Zone of Standardization.” Extending farther to the right, entropy increases as treatments show greater variance from the ideal at the origin.
This region incorporates the variations endemic in medical care: here lies the “Zone of Variability.” Acceptable treatment is defined by the sum of these two zones. Extending more to the right to greater entropy, we transgress into the realm of unacceptable care termed the “Zone of Error,” where medical care has deviated so far from the ideal as to no longer be acceptable.
Figure 5 builds further upon this construct and illustrates the impact of technology upon medical decision-making. Compared to Figure 4, technology has compressed the range of acceptable care and reduced the degree of freedom of the medical practitioner. The “Zone of Standardization” has broadened as decision support tools, such as protocols, care paths, guidelines, and algorithms have been implemented to orchestrate and systematize medical care; the “Zone of Variability” has narrowed as a result.
The “Zone of Error” has expanded at the expense of the narrowed range of acceptable care; some treatments at a given point on the entropy scale that would have previously been considered acceptable are interpreted as errors within the technology-driven regime.
The implication is that physician liability increases when practice lies outside the technologically-narrowed range of acceptable care. Such liability concerns are one factor accounting for physicians’ hesitancy to embrace technology they view as nonessential to delivering care. Physician reluctance to use decision support tools exposes their concern of being judged deficient within this normative framework.
Given the constraints on their mode of practice, it is no surprise that medical practitioners harbor reservations about the imposition of medical technology. Physicians do not want to abdicate their heretofore sacrosanct professional autonomy. Technology is not the only culprit; fiscal and political forces also constrain their autonomy. However, technology as embodied by health information systems reminds the physician on a daily basis of their changing status.
Does this characterization suggest that technology is perverse and misdirected? Certainly not: technology remains an essential component of modern health care without which the tremendous advances we have come to take for granted could not exist. However, accompanying the visible benefits of the technology iceberg are less-appreciated, but nonetheless important, ancillary effects hiding below the waterline. Technology offers powerful assistance to address the problems of uncertainty, variability, error, and quality, but technology is a two-edged sword, conveying benefits to medical practice simultaneously with significant costs. Robert Heinlein stated it best: “There ain’t no such thing as a free lunch.”
Implications for physician executives
Control and shaping of physician behavior are unavoidable byproducts of medical technology. Physician autonomy will increasingly be limited by technology. Physician acceptance of technology will vary directly with the particular technology’s utility and its ease of use but inversely to perceived constraints on autonomy. In basic terms, the benefits of technology must outweigh the costs. Technological innovations that are both effective and efficient (i.e., user friendly) will more successfully overcome physician resistance to their adoption.
The cries of anguish-labeled “infobia” by Rose-from practitioners living through the technological and information evolution of medical care are temporary. As older physicians fade from the scene, their ranks will be filled with those gestated in a system shaped by technology’s benefits and by its collateral effects. The health care system will transform itself with new recruits embracing health care technology and all of its ramifications.
“Technology, while adding daily to our physical ease, throws daily another loop of fine wire around our souls.”
My Faith in Democratic Capitalism Fortune, October 1995
Gregory A. Patton, MD, MS, CPE, is an Assistant Professor and Medical Director of the Department of Radiation Oncology at the University of Utah Health Sciences Center in Salt Lake City.
(1.) Gove, P.B. Webster’s Third New International Dictionary. Springfield, MA: Merriam-Webster, Inc., 1981, 2348.
(2.) Willoughby, K.W. Does Technology Need to be Managed? Syllabus: Management of Technology (B), Technological Contexts. David Eccles School of Business, University of Utah, 1995.
(3.) Willoughby, K.W. Technology Choice: A Critique of the Appropriate Technology Movement. Boulder, CO: Westview Press, 1990, 30.
(4.) Newhouse, J.P. An Iconoclastic View of Health Care Cost Containment. Health Affairs 1993; 12 Suppl:152-171.
(5.) Thorpe, K.E., Knickman, J.R. Financing for Health Care. In Kovner, A.R., Jonas, S. (eds). Jonas and Kovner’s Health Care Delivery in the United States. Sixth Edition, New York: Springer Publishing, 1999, 32-63.
(6.) Starr, P. The Social Transformation of American Medicine. New York: Basic Books, 1982.
(7.) Eddy, D.M. Variations in Clinical Practice: The Role of Uncertainty. Health Affairs. 1984; 3(2):74-89.
(8.) Goodwin, J.S. Culture in Chaos: The Need for Leadership and Followership in Medicine. Journal of the American Medical Association. 1997; 278(17):1399-1400.
(9.) Wennberg, J.A. Professional Uncertainty and the Problem of Supplier-Induced Demand. Social Sciences Medicine. 1982; 16:811-824.
(10.) Logan, R.L., Scott, P.J. Uncertainty in Clinical Practice: Implications for Quality and Costs of Health Care. The Lancet. 1996; 347:595-598.
(11.) Eddy, D.M. The Challenge. Journal of the American Medical Association. 1990; 263(2):287-290.
(12.) Chassin, M.R., Kosecoff, J., Park, R.E. et al. Does Inappropriate Use Explain Geographic Variations in the Use of Health Care Services? A Study of Three Procedures. Journal of the American Medical Association. 1987; 258(18):2533-2537.
(13.) Ashton, C.M., Peterson, N.J., Souchek, J., et al. Geographic Variations in Utilization Rates in Veterans Affairs Hospitals and Clinics. New England Journal of Medicine. 1999; 340(1):32-39.
(14.) Wennberg, J.A. Understanding Geographic Variations in Health Care Delivery. New England Journal of Medicine. 1999; 340(1):52-53.
(15.) The Center for the Evaluative Clinical Sciences, Dartmouth Medical School. Dartmouth Atlas of Health Care. Chicago: American Hospital Association, 1996.
(16.) Wennberg, J.A. The Paradox of Appropriate Care. Journal of the American Medical Association. 1987; 258:2568-2569.
(17.) Wennberg, J.A. Dealing with Medical Practice Variations: A Proposal for Action. Health Affairs. 1984; 3(2): 6-32.
(18.) Ellrodt, G., Cook, D.J., Lee, J. et al. Evidence-based Disease Management. Journal of the American Medical Association. 1997; 278(20): 1687-1692.
(19.) Kohn, L.T., Corrigan, J.M., Donaldson, M.S. To Err Is Human: Building a Safer Health System. Washington, D.C.: Institute of Medicine Press, 2000.
(20.) Leape, L.L., Brennan, T.A., Laird, N., et al. Incidence of Adverse Events in Hospitalized Patients: Results of the Harvard Medical Practice Study II. New England Journal of Medicine. 1991; 324(6)370-376.
(21.) Brennan, T.A., Leape, L.L., Laird, N., et al. Incidence of Adverse Events and Negligence in Hospitalized Patients: Results of the Harvard Medical Practice Study I. New England Journal of Medicine. 1991; 324(6):370-376.
(22.) Berwick, D.M., Godfrey, A.B., Roessner. J. Curing Health Care: New Strategies for Quality Improvement. San Francisco: Jossey-Bass, 1990.
(23.) Leape, L.L. Error in Medicine. Journal of the American Medical Association. 1994; 272(23): 1851-1857.
(24.) Eddy, D.M. Practice Policies–Where do They Come From? Journal of the American Medical Association. 1990; 263(9):1265-1275.
(25.) Chassin, M.R., Galvin, R.W. The Urgent Need to Improve Health Care Quality. Journal of the American Medical Association. 1998; 280(11):1000-1005.
(26.) Classen, D.C., Pestotnik, S.L., Evans, R.S., Lloyd, J.F., Burke, J.P. Adverse Drug Events in Hospitalized Patients: Excess Length of Stay, Extra Costs, and Attributable Mortality. Journal of the American Medical Association, 1997; 277(4):301-306.
(27.) Bates, D.W., Leape, L.L., Cullen, D.J., et al. Effect of a Computerized Order Entry and a Team Intervention on Prevention of Serious Medication Errors. Journal of the American Medical Association. 1998; 280(15):1311-1316.
(28.) Slack, W. Cybermedicine: How Computing Empowers Doctors and Patients for Better Health Care. San Francisco: Jossey-Bass, 1997.
(29.) Eddy, D.M. Money, Medicine and Mathematics. Bulletin of the American College of Surgeons. 1992; 77(6):36-49.
(30.) Millenson, M.L. Demanding Medical Excellence: Doctors and Accountability in the information Age. Chicago: University of Chicago Press, 1997.
(31.) Berg, M. Rationalizing Medical Work: Decision Support Techniques and Medical Practices. Cambridge, Massachusetts: MIT Press, 1997.
(32.) McDonald, C.J. Protocol-based Computer Reminders, the Quality of Care, and the Non-Perfectibility of Man. New England Journal of Medicine. 1976; 295:1351-1355.
(33.) Evans, R.S., Pestotnik, S.L., Classen, D.C., et al. A Computer-Assisted Management Program for Antibiotics and other Anti-infective Agents. New England Journal of Medicine. 1998; 338:232-238.
(34.) Berner, E.S. Clinical Decision Support Systems: Theory and Practice. New York: Springer-Verlag. 1999.
(35.) Metzger, J. Cross-Continuum Care Management: Information Management Challenges. In Drazen, E., Metzger, J. (ed). Strategies for Integrated Health Care: Emerging Practices in Information Management and Cross-Continuum Care. San Francisco: Jossey-Bass, 1999, 160-181.
(36.) Miller, R.A. Medical Diagnostic Decision Support Systems–Past, Present and Future: A Threaded Bibliography & Commentary. Journal of the American Medical Informatics Association. 1994; 1(1):8-27.
(37.) Treister, N.W. Physician Acceptance of New Medical Information Systems: The Field of Dreams. The Physician Executive. 1998; 24(3): 20-24.
(38.) Heinlein, R.A. The Moon is a Harsh Mistress. New York: Tor Books, 1967.
(39.) Rose, J.S. Medicine & the Information Age. Tampa, FL: American College of Physician Executives, 1998.
COPYRIGHT 2001 American College of Physician Executives
COPYRIGHT 2002 Gale Group