4
Risks of Ionizing Radiation in Medicine

The discussion of the regulation of radiation medicine in Chapter 3 concluded with an account of beliefs and attitudes about regulation that the committee found to be prevalent within the regulated community. One common theme was that the Nuclear Regulatory Commission (NRC) regulatory requirements are out of proportion to the risks involved. The present chapter brings to the forefront the issues of what the risks of ionizing radiation are estimated to be and how this issue is currently addressed in the exercise of regulatory authority.

The chapter opens with a general discussion introducing the concept of risk assessment and the conceptual model currently used by U.S. regulatory agencies as a matter of public policy to assess the risk of exposure to low-level radiation (the "linear, no-threshold" model). The next section addresses risk from a different standpoint: the likelihood that unintended exposures will occur as the result of error or accident in the medical use of ionizing radiation. Finally, the third section addresses the public's perception of risk.

RISK ASSESSMENT

For the purposes of this report, human health risk assessments include the evaluation of scientific information about (a) hazardous properties of radiation and radioactive materials and (b) the extent of human exposure to these agents. These risk assessments provide estimates of the probability that exposed populations will be harmed and to what degree. The probability may be expressed either quantitatively or qualitatively; it is typically arrived at through an analytic



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform 4 Risks of Ionizing Radiation in Medicine The discussion of the regulation of radiation medicine in Chapter 3 concluded with an account of beliefs and attitudes about regulation that the committee found to be prevalent within the regulated community. One common theme was that the Nuclear Regulatory Commission (NRC) regulatory requirements are out of proportion to the risks involved. The present chapter brings to the forefront the issues of what the risks of ionizing radiation are estimated to be and how this issue is currently addressed in the exercise of regulatory authority. The chapter opens with a general discussion introducing the concept of risk assessment and the conceptual model currently used by U.S. regulatory agencies as a matter of public policy to assess the risk of exposure to low-level radiation (the "linear, no-threshold" model). The next section addresses risk from a different standpoint: the likelihood that unintended exposures will occur as the result of error or accident in the medical use of ionizing radiation. Finally, the third section addresses the public's perception of risk. RISK ASSESSMENT For the purposes of this report, human health risk assessments include the evaluation of scientific information about (a) hazardous properties of radiation and radioactive materials and (b) the extent of human exposure to these agents. These risk assessments provide estimates of the probability that exposed populations will be harmed and to what degree. The probability may be expressed either quantitatively or qualitatively; it is typically arrived at through an analytic

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform process such as the four-step process described in Science and Judgment in Risk Assessment (NAS/NRC, 1994): Hazard Identification: Identification of the potentially hazardous agent and a description of the specific forms of toxicity that may be expressed in exposed populations. Dose-Response Assessment: Evaluation of the conditions under which an agent may be toxic and an evaluation of the quantitative relationship between the dose and the toxic response. Exposure Assessment: Specification of the population that might be exposed, identification of the routes by which exposure can occur, and estimates of the magnitude and duration of exposure that people are likely to receive. Risk Characterization: Development of a qualitative or quantitative estimate of the hazards associated with the agent that will be realized in exposed people. This includes a full discussion of the uncertainties associated with the estimates of risk. In the case of risks of exposure to radiation at low levels (less than 0.1 gray (Gy) effective dose equivalent delivered instantaneously or less than 0.2 Gy delivered at a low dose rate), the scientific uncertainty about negative side effects, such as long-term cancer, is considerable. Radiogenic cancer (i.e., neoplasms caused by exposure to ionizing radiation) has not been observed in humans at low levels of exposure, and it is unknown whether such negative sequelae to diagnosis actually exist. If they exist, it is beyond the range of current science and medicine to observe and measure them. Because of this scientific uncertainty, a public policy decision has had to be made regarding what approach to use in setting standards for protection of individuals against potential problems secondary to low-level exposures to radiation. U.S. regulatory agencies currently use a model that describes radiation injury as a linear function of radiation dose that has no lower threshold; this is called the linear, no-threshold model. Scientific consensus groups, including the International Commission on Radiological Protection (ICRP), the National Council on Radiation Protection and Measurements (NCRP), and a National Research Council Committee on the Biological Effects of Ionizing Radiation (BEIR), have also endorsed the use of this model for risk assessment (see NCRP, 1987; NAS/NRC, 1990; ICRP, 1991). This approach reflects an understandable tendency to be conservative in choosing analytical models that emphasize public safety. In the linear, no-threshold model, data from high levels of exposure (greater than 0.5 Gy effective dose equivalent), where some radiogenic cancers have been observed in humans, are extrapolated to low levels of exposure, where radiogenic cancer has not been observed. The model can then be used to predict the risks of cancer at various levels of exposure; it specifically predicts very

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform small but finite risks, at low-level exposure. One result of this predicted likelihood of very small but finite risk is that, if it is multiplied by a large number (for example, the population of the United States), it produces an estimate of a finite number of radiation-induced cancer deaths—even though whether such deaths actually occur is unknown. In reviewing the current approach to radiation risk assessment, the intention of this section is to focus on the relation between the linear, no-threshold model and the present status of empirical knowledge about the health effects of exposure to low-level radiation. The following subsections discuss kinds of radiation injury and the limitations of human studies and then recount the history of the adoption of the linear, no-threshold model; a more detailed history, including discussion of the scientific debates the model's widespread application has occasioned, appears in Appendix K. Kinds of Radiation Injury There are two distinct types of radiation injury—acute radiation injury and late radiation injury. Acute radiation injury, also called prompt injury, occurs in response to large doses of radiation delivered over relatively short periods of time. Acute injuries include erythema (skin reddening), epilation (hair loss), nausea, diarrhea, sterility, organ atrophy, tissue fibrosis, and even death. Some acute injuries may not appear symptomatically for several months; also, some may be irreversible. Acute injuries are said to be deterministic (nonstochastic ) because higher doses lead to more severe injuries. Late radiation injuries are limited to cancer and hereditary effects. These effects occur at lower doses than those that cause acute radiation injury. Furthermore, the dose may be spread over a longer period of time. Late radiation injuries are often assumed to be nondeterministic (stochastic) in nature, with a probability, but not severity, that depends on the dose. This assumption implies a model of radiation injury in which the likelihood of long-term radiation injury increases with dose. As noted earlier, this model is unproved at low levels of radiation exposure. Human Study Limitations Most of the human data related to long-term effects of radiation exposure have been obtained from epidemiological studies, including those of survivors of the atomic explosions in Hiroshima and Nagasaki. These Japanese survivors are by far the largest and most closely followed of all populations exposed to ionizing radiation. Many uncertainities remain, however, about the universal applicability of the results of the Japanese studies, as is the case with all human epidemiological studies. For example, Japanese survivors had been exposed to a single burst of mixed neutrons and gamma rays; their observed health effects cannot be directly applied or generalized to other groups exposed to different

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform types of radiation under different conditions, including protracted exposures to much lower levels of radiation. In other radiation studies, populations are frequently composed of individuals who are being treated for various diseases or who are exposed for diagnostic purposes because of suspected illnesses, including those later identified as possibly radiation induced. Dose estimates for exposed individuals in such studies may have to be reconstructed or inferred, often with considerable uncertainty. Although efforts are often made to compare injury in an irradiated population with that in a comparable control group, it usually is impossible to define a control population that is identical with the study group in all aspects except radiation exposure. Confounding factors frequently are simply ignored because they cannot be verified or quantified. In addition, because radiation effects such as cancer and hereditary injury are not different in expression from those due to "natural causes," studies must focus on increases in incidence and mortality rather than on the mere presence of disease in the exposed population. This difficulty in distinguishing radiation effects from natural causes introduces considerable uncertainty about the extent of health effects in populations exposed to low levels of radiation. In those studies in which health effects are suspected, the occurrence of no effect, or even a beneficial health effect, is within the range of statistical uncertainty (see, e.g., UNSCEAR, 1994). Models of Radiation Injury For the first decade or so following the discovery of x-rays in 1895 and radioactivity in 1896, many physicists and physicians experimented with ionizing radiation without concern for possible health effects. Soon, however, they identified effects such as skin burns, hair loss, and ulcerating sores, some of which eventually became skin cancers. Although a few radiation pioneers called for protective measures against radiation, not until the 1920s did physicians and professional organizations, primarily medical societies, acknowledge the need for control strategies and exposure limits for ionizing radiation. The Tolerance Dose, Threshold Model In 1925 the concept of the tolerance dose was introduced as an upper limit for the exposure of workers to radiation. This concept was based on the premise of a threshold dose, that is, a level of exposure below which ill effects do not occur in exposed persons. The tolerance dose model of radiation injury, with its implied threshold below which adverse effects of radiation exposure do not occur, was the preferred model of radiation injury until after World War II. For many years guidelines for radiation protection, as promulgated by advisory councils such as the ICRP, NCRP, and their predecessors, were based on the tolerance-dose, threshold model of radiation-induced injury.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform Introduction of the Linear, No-Threshold Model In the 1950s, during a period of new concerns about low-dose radiation from peacetime nuclear industries and from worldwide nuclear test fallout, a new model of radiation-induced injury began to be used for the purpose of establishing guidelines for radiation protection. This model assumed that late effects of radiation exposure (late radiation injuries, namely, cancer and hereditary effects) might increase linearly with increasing dose of radiation, and that a threshold dose might not exist below which late effects do not occur. This new model—called the linear, no-threshold model1 of radiation risk—assumed that the risk of late radiation injuries at low doses can be estimated by linear extrapolation from effects at high doses. Initially, this model did not dismiss the idea of a threshold dose. Instead, it included that possibility in the implication that the true risk of adverse health effects in the low-dose region lies somewhere along a range between zero and an upper limit defined by linear extrapolation. Over the years, however, the concept of a range of risk values from zero to an extrapolated upper limit has been largely forgotten. Instead, the extrapolated upper limit has assumed prominence as a quantified measure of health risk related to low-level radiation exposure. Numerous studies of large populations of humans, however, have failed to either demonstrate or refute with statistical significance any adverse health effects related to low-level radiation exposure. Despite this limitation, the linear, no-threshold model has become widely adopted for estimating radiation risk and for quantifying the number of potentially injured persons in a population exposed to ionizing radiation. The basis for this application appears to lie largely in a conservative approach to risk assessment that has been taken to compensate for a lack of reliable data in the relevant low-dose range (NCRP, 1993). Further discussion of the evolution of a linear, no-threshold model can be found in Appendix K. Summary Observations The contemporary practice of radiation protection is based on application of the linear, no-threshold model to estimate long-term risks to human health from exposure to low levels of ionizing radiation. This model is currently accepted by various advisory groups including those concerned with the BEIR committees, United Nations Scientific Committee on the Effects of Atomic Radiation, NCRP, and ICRP. It is often coupled with a concept that calls for maintenance of radiation exposures at as low as reasonably achievable levels. The committee assumes that this model is likely to remain as the underlying philosophy for the practice of radiation protection (see, e.g., UNSCEAR, 1962, 1964; ICRP, 1966). As a 1   Linear, no-threshold models are used for all toxic environmental exposures, not just for radiation. These issues are not unique to radiation or to its use in medicine.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform conservative and mathematically simple hypothesis, the linear, no-threshold model provides an upper limit of risk useful for setting protection standards. As a model of risk at low doses, however, it has not been verified by human epidemiological data. Acceptance of this model has discouraged efforts to establish minimum levels of radiation exposure that can be classed either as ''below regulatory concern" or as of "negligible individual risk levels," that is, levels that would imply that regulatory control of exposures is unnecessary (NCRP, 1993). Regulatory programs established to control radiation exposures at low levels should be tempered with a sense of practicality. The committee has two concerns: first, that the costs, both financial and administrative, of efforts to achieve increasingly lower limits of human exposure may compromise useful applications of ionizing radiation and, second, that this situation risks depriving the public of the medical and societal benefits of this medical source. A balance should be attained that reflects not only safe delivery of health care, but also a reasonable level of efficiency and cost-effectiveness in the regulatory process. RISKS OF IONIZING RADIATION IN MEDICAL TREATMENT Risk of Unintended Exposures in Radiation Medicine The preceding section has addressed the issue of risk from the standpoint of how models quantify the harm, if any, caused to humans who are exposed, under any circumstances, to low-dose radiation. This section addresses risk from a different standpoint: When patients are intentionally exposed to ionizing radiation for medical purposes, do they suffer unintentional exposures as a result of error or accident? No medical intervention, whether diagnostic or therapeutic, comes without risk. Acceptance of medical uncertainty is properly left to the discretion of patients in consultation with their doctors. Some risks, such as those relating to radiation overexposure, underexposure, or exposure of the wrong body part, can be minimized. Training and quality management programs may help to reduce these problems. Risk in the area of radiation medicine has several dimensions that are less common in other areas of medicine (although not nonexistent). First, there may be risks from overexposure that do not cause immediate injury. For example, the causal connection, if any, may be difficult or impossible to verify for a malignancy that surfaces several years after an inappropriate exposure. Second, the risks associated with the medical use of ionizing radiation extend beyond the patient and can affect health care workers and the public. In amplifying these and other aspects of the risks that attend medical uses of ionizing radiation, the discussion below addresses the following series of topics:

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform human error and unintended events; rates of misadministration in radiation medicine; misadministrations and adverse events in other medical modalities; inappropriate and unnecessary care; and efforts that reduce misadministrations and inappropriate care. Human Error and Unintended Events Errors occur throughout health care: A pharmacist fills a prescription with the wrong medicine; an x-ray technician takes a film of the wrong leg; a surgeon replaces the wrong hip. The advent of complex medical technology has increased the opportunity for error even as it has increased the opportunity for effecting cures. Injuries within the health care context, including those resulting from human error, are referred to as "iatrogenic." A landmark Harvard Medical Practice study reported that nearly 4 percent of patients hospitalized in New York in 1982 suffered an iatrogenic injury that resulted in a prolonged hospital stay or measurable disability (Brennan et al., 1991; Leape et al., 1991). The Harvard researchers conducted random samples of both acute care hospitals and patients, identifying some 1,133 adverse events from a total of 30,195 medical records reviewed. The investigators estimated that, of 2.7 million patients discharged from New York hospitals that year, 98,609 suffered an adverse event, for a rate of 3.7 percent of all hospital discharges. Based on the New York rate, the researchers estimated that in the United States more than 1.3 million people are injured annually by treatments intended to help them. By educating health care workers, and by circumscribing their actions, human error may be minimized. However, some number of mistakes will always, unavoidably, be made, and no amount of training or double-checking can erase that fact. Recent incidents at the Dana-Farber Institute and at the University of Chicago Hospital are cases in point. Both institutions enjoy strong reputations and have quality assurance programs in place; yet, chemotherapy overdoses escaped notice in both systems, killing two patients and leaving a third permanently disabled. Although well-crafted and well-implemented regulatory programs can prevent most safety problems, they cannot completely eliminate human error. Rates of Misadministration This report refers to errors and unintended events that occur in the course of administering ionizing radiation in medicine as "adverse events." "Misadministrations" and ''reportable events" refer to adverse events involving NRC-regulated byproduct material and are defined at 10 CFR Part 35 (see Appendix D). One task of the committee was to determine how often adverse events occur in the use of ionizing radiation in medicine.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform Misadministrations in Byproduct-Related Ionizing Radiation in Medicine In 1992, NRC data showed 7 diagnostic and 29 therapeutic misadministrations for the 2,228 licensees of the NRC. In that same year, the Agreement States reported 7 diagnostic misadministrations and 10 therapeutic "reportable events" for their 4,944 licensees. Combining the NRC and Agreement State data yields a total, for 1992, of the following numbers of misadministrations or reportable events for 7,172 licensees: 38 in therapeutic radiation oncology (teletherapy and brachytherapy); 1 in therapeutic nuclear medicine; and 14 in diagnostic nuclear medicine. Even if it is assumed that the numbers are underreported by a factor of 10, owing to confusion as to what is specifically reportable or to intentional non-filing, this level of misadministration is remarkably low. These figures can be used in conjunction with the total number of administrations in each category to estimate overall error rates. In 1992, about 11 million radiopharmaceutical administrations were given to patients in the United States; approximately 3.5 million of these were provided in NRC-regulated states. Using NRC data on misadministrations for diagnostic procedures (which are higher than those for the Agreement States), these numbers translate to an estimated diagnostic misadministration rate of 0.0002 percent (7 divided by 3.5 million) or about 2 per million administrations (not patients). If data from Agreement States and NRC-regulated states are combined, the estimate of diagnostic misadministrations becomes 0.00012 percent (14 divided by 11,000,000), or about 1.3 per million.2,3 2   The United States Pharmacopoeia (USP) runs a voluntary program within which users of radiopharmaceuticals report problems encountered in the administration of the drugs. Over a two-year period, from October 1, 1993, to September 30, 1995, only 42 voluntary reports were submitted. These "problems" are not misadministrations, but they do include incidents of incorrect biodistribution and other failures inherent in the patient reaction to the drug. Other USP problem reporting programs estimate that voluntarily submitted reports represent 10 percent of actual problems. Reports dealing with radiopharmaceuticals may represent an even higher percentage of actual problems, as physicians involved in the use of radiopharmaceuticals may be even more conscious of safety and adverse events. In any case, the tiny number of reported problems illustrates the low rate of adverse events associated with radiopharmaceuticals. 3   (ECRI) data reviewed by the committee showed an occurrence of only 168 total adverse events over a three-year period. These data are rough; there is no indication of the number of at-risk administrations, nor is the completeness of reporting defined. Nonetheless, the denominator must be huge. Additionally, of the total 168 events, only 43 caused actual injury.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform In the area of therapeutic radiation oncology, administrations in the United States in 1992 were estimated to be associated with treatments for a total of 545,600 patients (as cited by Pollycove, 1993). Of these, 100,600 had byproduct teletherapy and 30,000 brachytherapy. The remaining 415,000 patients were treated with linear accelerator radiation therapy. Thus, the NRC regulatory apparatus is directed at misadministrations involving only about 25 percent of the total radiation therapy patients treated, and only about 0.026 percent of the total of all medication administrations in the United States per year (12 million out of 3.75 billion hospital medication administrations). Patients receiving radiation therapy routinely receive multiple treatments. For purposes of calculating misadministration or error rates, the reasonable, but conservative, estimates of 20 treatments per teletherapy patient and 2 treatments per brachytherapy patient are used. This yields an annual administration rate of approximately 2 million. Using this figure, one can estimate the byproduct therapy misadministration rate at 0.002 percent (38 divided by 2,072,000, or about 2 per 100,000 administrations). NRC-regulated states gave about one-third of the treatments. For these states, the estimated misadministration rate for byproduct therapy is 0.004 percent (21 divided by 690,667). Nevertheless, lacking better data, the committee concluded that approximate rates of byproduct-related misadministrations for 1992 could be said to be: for diagnostic misadministrations (Agreement States and NRC-regulated states combined), 0.00012 percent of all such administrations; for diagnostic misadministrations (NRC-regulated states only), 0.0002 percent; for therapeutic misadministrations (Agreement States and NRC-regulated states combined), 0.002 percent; and for therapeutic misadministrations (NRC-regulated states only), 0.004 percent. All these figures must be regarded as very rough approximations, for several reasons. No information is at hand on the degree of undercounting; the counts themselves are necessarily subject to statistical variation (the standard error of each is roughly the square root of the count); and the denominators are very crude estimates.4 4   It should be noted that the NRC-regulated states have been reporting misadministrations, pursuant to the QM rule (10 CFR 35.32) since January 1992. The Agreement States have only been required to do so since January 1995. Other than this fact, the committee has no explanation for the apparent discrepancy in these rates.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform Gaps in Data Collection Only 10 percent of ionizing radiation used in medicine is subject to the NRC and Agreement State regulatory system. The remaining 90 percent is not federally regulated; in some instances, it is not state regulated either. Because no federal requirement exists for data collection pertaining to non-byproduct radiation sources (other than the Food and Drug Administration (FDA) reporting requirements concerning death, serious injury, or equipment malfunction), realistic, accurate data on the incidence and type of problems associated with non-byproduct radiation medicine remain elusive. The lack of data is a continuation of a problem noted by the NRC's Office of Policy and Planning in 1993 in its Task Force Report on Medical Radiation Protection. The report stated that sufficient data are not available to assess the level of protection for all sources of medical radiation. Although the report noted that acquisition of selected performance data could provide insights needed to consider appropriate regulatory changes, it concluded that "reliable data to assess risk and program effectiveness will be difficult to acquire, especially for non-byproduct sources of radiation" (NRC, 1993, p. 4). The General Accounting Office (GAO) has also considered the issue of inadequate data (GAO, 1993) and offered two recommendations. First, it advised the NRC to establish common performance indicators to obtain comparable data from all users of byproducts in radiation medicine practice, including both facilities regulated by the NRC and those regulated by Agreement States. The GAO believed that such information would facilitate an evaluation of the effectiveness of the NRC's program and those of the Agreement States. Second, the GAO recommended that the NRC establish specific criteria and procedures for suspending or revoking an Agreement State's program. The GAO report asserted that the NRC lacked both good criteria and data by which to evaluate the effectiveness of its radioactive materials programs, especially with respect to adequate protection of the public from radiation in either Agreement States or NRC-regulated states (GAO, 1993). It emphasized the need for a common set of performance indicators to evaluate Agreement State and NRC licensees and the need for all users to be required to collect and report information using the same definitions, procedures, and criteria. Comment As a general proposition, the committee subscribes to the view that performance indicators can serve the health care industry. Many areas of health care other than radiation medicine deserve equal attention, as suggested by the data cited below on adverse effects of medications in general. With regard to NRC data collection, the committee cautions that efforts directed at radiation medicine

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform should be justified on the basis of cost and benefit, with both risks of harm to patients and expenditures entailed taken into account. Today some data are collected on poor performance and adverse events. For example, the Joint Commission on Accreditation of Healthcare Organizations (JCAHO) has a performance standard that requires intensive assessment when hospital performance varies from recognized standards. The JCAHO specifies confirmed transfusion reactions and significant adverse drug reactions as cause for further assessment, but it does not specifically require hospitals to report medication errors except in accord with the hospital's written procedures (JCAHO, 1995). As another case in point, the medical centers of the Department of Veterans Affairs are required to report and track medication errors. Generally, however, this type of reporting to a central authority does not appear to be practiced in the private sector. In comparison, where radiation medicine is regulated by the NRC, either directly or through administration of NRC standards by the Agreement States, all administrations of radioactive materials must be documented and those that are classified as misadministrations must be reported. In 1992, the volume of services amounted to approximately 11 million NRC-regulated unsealed source radionuclide administrations in nuclear medicine and 130,600 NRC-regulated (brachytherapy and teletherapy) radiation therapy patients. Some members of the nuclear medicine and radiation therapy communities question whether the putative benefits of this level of regulatory oversight and reporting are commensurate with the costs (see Chapter 3), believing that relaxation of this recording and reporting system would not increase the risk of patient injury. Conclusion The committee was able to make rough approximations of the rate of misadministrations for medical procedures involving byproduct material. Because of the lack of data, it could not estimate rates for all ionizing radiation used in medicine. Misadministrations and Adverse Events in Other Medical Modalities The NRC had asked the Institute of Medicine to compare the errors in use of byproduct materials and the consequences of those errors with errors occurring in other medical interventions. Adverse events in administration of medications, including chemotherapy, blood transfusions, and surgical interventions, were specifically requested for this comparison. In statistical, clinical, and epidemiological terms, comparisons of the risks inherent in very different health care interventions can be problematic. However, the committee judged that such information

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform TABLE 4.1 Ordering of Perceived Risk for 30 Activities and Technologies   Experts League of Women Voters College Students Active Club Members Nuclear power 20 1 1 8 Motor vehicles 1 2 5 3 Handguns 4 3 2 1 Smoking 2 4 3 4 Motorcycles 6 5 6 2 Alcoholic beverages 3 6 7 5 General (private) aviation 12 7 15 11 Police work 17 8 8 7 Pesticides 8 9 4 15 Surgery 5 10 11 9 Fire fighting 18 11 10 6 Large construction 13 12 14 13 Hunting 23 13 18 10 Spray cans 26 14 13 23 Mountain climbing 29 15 22 12 Bicycles 15 16 24 14 Commercial aviation 16 17 16 18 Electric power (nonnuclear) 9 18 19 19 Swimming 10 19 30 17 Contraceptives 11 20 9 22 Skiing 30 21 25 16 X-rays 7 22 17 24 High school and college football 27 23 26 21 Railroads 19 24 23 20 Food preservatives 14 25 12 28 Food coloring 21 26 20 30 Power mowers 28 27 28 25 Prescription antibiotics 24 28 21 26 Home appliances 29 22 27 27 Vaccinations 25 30 29 29 Note: The ordering is based on the geometric mean risk ratings within each group. Rank 1 represents the most risky activity or technology. Source: Slovic, 1987, Table 1, p. 281; reprinted with permission. Copyright 1987 by American Association for the Advancement of Science. radiation in medicine. Inasmuch as it is reasonable to presume that the common adjective "nuclear," in both nuclear power and nuclear medicine, or the word "radiation" itself, may trigger public concern with respect to risk, this section explores the public perception of radiation and the risks it may pose to health and well-being of patients, health care personnel, and the public. Evidently, people perceive that different sources of radiation exposure pose different levels of possible harm (Kunreuther et al., 1988; Slovic et al., 1991b) (see Table 4.2). For instance, experts recommend that action be taken to reduce

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform FIGURE 4.1 Mean perceived risk and perceived benefit for medical and nonmedical sources of exposure to radiation and chemicals. Each item was rated on a scale of perceived risk ranging from 1 (very low risk) to 7 (very high risk) and a scale of perceived benefit ranging from 1 (very low benefit) to 7 (very high benefit). Note that medical sources for exposure have more favorable benefit/risk rating than do the nonmedical sources. SOURCE: Slovic, 1993. the moderate risk associated with radon, but the public is apathetic toward the problem. Conversely, consumers question the acceptability of food irradiation, while experts dismiss such worries. Perhaps the difference is clearest when it comes to nuclear power. Experts consider this a moderate, but acceptable, risk, whereas the public has serious questions about the safety of nuclear power. The association of radiation, nuclear power, and nuclear waste with catastrophe has a long history (Weart, 1988). The bombings of Hiroshima and Nagasaki are forever a part of the public's collective memory: "Nuclear energy was conceived in secrecy, born in war, and first revealed to the world in horror. No matter how much proponents try to separate the peaceful from the weapons atom, the connection is firmly embedded in the minds of the public" (Smith, 1988). In addition to immediate destruction, public fear relates to contamination,

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform TABLE 4.2 Summary of Perception and Acceptance of Risks from Diverse Sources of Radiation Exposure   Perceived Risk   Technical Experts Public Nuclear power/nuclear waste Moderate risk Acceptable Extreme risk Unacceptable X-rays Low/moderate risk Acceptable Very low risk Acceptable Radon Moderate risk Needs action Very low risk Apathy Nuclear weapons Moderate to extreme risk Tolerance Extreme risk Tolerance Food irradiation Low risk Acceptable Moderate to high risk Acceptability questioned Electric and magnetic fields Low risk Significant concerns beginning to develop   Acceptable Acceptability questioned   SOURCE: Slovic, 1990, Table 2, p. 79; reprinted with permission. Copyright 1990 by National Council on Radiation Protection and Measurements. whereby radiation affects the landscape, body tissues, and the genetic makeup of future generations (Erickson, 1990, 1991). In contrast is the public's lack of concern regarding radon. Although some experts view radon as a moderate risk requiring action (Bord and O'Connor, 1990), the public, influenced by radon's natural origin, its occurrence in a familiar setting, and the absence of someone to blame, sees radon as extremely low risk. Similarly, experts rate nuclear power as a lower risk than medical x-rays, whereas lay persons think just the opposite (see Table 4.1). Impact of Perceptions When people perceive that something is hazardous or poses threats to life, health, or well-being, they want that risk reduced and they are willing to employ regulation to do so. Accusations by experts that public reactions are "irrational" or "phobic" notwithstanding, perceptions are real and must be dealt with. Public assessment of risks has ripple effects that can result in substantial social, political, and economic impact. The accident at Three Mile Island, for instance, not only affected that specific nuclear plant but also had enormous implications for the nuclear industry and for society; these included stricter regulation, increased costs of reactor construction and operation, fewer reactors worldwide, greater public opposition to nuclear power, and reliance on more expensive energy sources (Evans and Hope, 1984; Heising and George, 1986).

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform A theory as to how various factors in society "amplify risks" and produce ripple effects has been presented by Kasperson et al. (1988). The long-range impact of an event depends upon its "signal value," which reflects new information about the likelihood of similar future events. This signal value relates to the context in which the event takes place. For example, a train accident with major loss of life may produce little social disturbance if it occurs within a familiar system. In contrast, a smaller accident within an unfamiliar system, such as a brachytherapy misadministration, may have immense social impact if it is viewed as a harbinger of future major mishaps. Society understandably reacts negatively to poorly understood and seemingly catastrophic events. Risk Communication and Trust The variety of risk perceptions and the heterogeneity of assessment among experts and lay persons make risk communication extremely important. Despite the substantial difficulty involved, the public must be better informed about risks so that they can put them into perspective, facilitate decisionmaking, and diffuse unnecessary anxiety. However, even well-thought-out approaches to risk communication have the potential for achieving just the opposite result; namely, they can enhance public distrust. In fact, the very discussion of potential risk might be perceived as revealing a real risk. Several approaches have been used to communicate risk. One such approach has been "risk comparisons" (Wilson, 1979). Such comparisons may be more meaningful than the presentation of simple probabilities, especially when these numbers are quite small. Risk comparisons may provide some clarity, especially for persons comfortable with quantitative analysis, but they do not always educate effectively among groups less "numeric" in training or orientation. The way information is presented is also important in influencing risk perceptions. Subtle changes in the way that risks are "framed" can have a major impact on decisions. This was amply demonstrated in a landmark project in medicine (McNeil et al., 1982). In it, the investigators demonstrated that when individuals are offered the options of surgery or radiation therapy as treatment for lung cancer, the percentage of patients choosing a specific therapy dropped dramatically when success rates were stated in terms of dying rather than surviving. Indeed, the very discussion of risk can fuel a perception of potential hazards. Even assurances of low risk fail when the public focuses on the word "risk" and not on its minimal ("low") nature. In this sense, regulations first developed to provide safe nuclear energy and then superimposed on the medical use of byproduct material communicate to the public a continuum of hazard. Within that communication, the concept of absolute risk, rather than degree of risk, predominates.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform Perceptions also relate strongly to trust. Greater public acceptance of medical, as opposed to industrial, technology grows from a relatively high degree of trust in physicians.7 The limited effectiveness of risk communication efforts in many circumstances may be attributable to lack of trust. Trust, in fact, may be more fundamental to changing public perception of risk than is clarity of communication. Trust is a fragile phenomenon. It tends to be created slowly, may be destroyed by a single mishap, and takes a long time to rebuild. This asymmetry is believed to be a mechanism of human psychology (Slovic, 1995). Trust-destroying events are more visible to the public, and they carry much greater weight than do trust-building events. As a case in point: college students who rated the impact on trust of a series of 45 hypothetical events pertaining to the management of a large nuclear power plant demonstrated this asymmetry dramatically (see Figure 4.2). For example, although on-site inspectors and responsiveness to the first signs of problems inspire some trust, the discovery of poor recordkeeping decreases trust by an even greater percentage. The importance of an event also relates in part to its rarity. An accident in a nuclear plant affects trust far more than does a large number of accident-free days. Another aspect of the asymmetry principle is the phenomenon whereby trust-destroying news is seen as more credible than good news. For instance, one study demonstrating potential carcinogenicity in animals carries more weight in the public mind than several studies that disprove such an effect (Efron, 1984). Another important psychological tendency is that distrust, once initiated, tends to reinforce and perpetuate future distrust. The news media also give greater weight to negative than to positive events, thriving, as they do, on trust-destroying news. On March 20, 1991, the Journal of the American Medical Association carried two studies, both of which evaluated potential links between radiation exposure and cancer. One study indicated an increased risk in leukemia for white men working at the Oak Ridge National Laboratory. The other suggested no increased risk of cancer in people residing near specific nuclear facilities. The subsequent newspaper coverage focused in far greater detail on the study showing increased risk than on the other article (Koren and Klein, 1991). Implications for Radiation Medicine The past 20 years of research into the perception of risk have seen little attention paid to medical uses of radiation, in contrast to multiple studies evaluating the perception of risks associated with nuclear power and waste. Analogous to x-rays, other radiation in medicine is likely to be viewed more 7   Such trust, however, may be waning in the face of higher costs of medicine, the epidemic of malpractice claims, and the general public decline in deference to authority.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform FIGURE 4.2 Differential impact of trust-increasing and trustdecreasing events. NOTE: Only percentages of Category 7 ratings (very powerful impact) are shown here. SOURCE: Slovic, 1993.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform favorably than nuclear power because of the perceived benefits associated with medicine and the relatively strong trust in the medical profession. Public perception, however, is not free of anxieties associated with radiation's risk of subsequent cancer. In view of the psychological profiles discussed above, strong public reaction to incidents of overexposure is not surprising. Small incidents can cause major ripple effects, and these in turn may prompt calls for stricter regulation. Further studies are clearly necessary to understand some of the perceptions of radiation medicine risk. There is a distinct need to develop appropriate strategies for dealing with these perceptions. Effort should be placed on trust-building strategies that ease the "dread" and "unknown" aspects of radiation risk. The major medical benefits of radiation medicine must be emphasized. Furthermore, the perceptual linkages among nuclear power, Three Mile Island, and radiation medicine must be uncoupled. CHAPTER SUMMARY This chapter has explored several aspects of the risks involved in the use of ionizing radiation in medicine. Having set out basic concepts necessary to understanding the regulation of these risks, including the linear, no-threshold model currently used by U.S. regulatory agencies, the chapter then goes on to look at what is known about the actual incidence of adverse events in radiation medicine. Although comparisons between misadministrations involving NRC-regulated materials and adverse events in other medical modalities are imperfect, the committee felt that such a broad contextual view helped it in its assessment of the risks arising from use of byproduct materials. Finally, recognizing that regulation is often as much a response to public pressure as it is to scientific opinion, the committee has included a look into what is known about the public's perception of ionizing radiation. REFERENCES Allan, E.L., and Barker, K.N. Fundamentals of Medication Error Research. American Journal of Hospital Pharmacy 47:555–571, 1990. Anonymous. Deaths During General Anesthesia: Technology-Related, Due to Human Error or Unavoidable? An ECRI Technology Assessment. Journal of Health Care Technology 1(3):155–175, 1985. Anonymous. Thousands of Nursing Homes Do Not Follow Drug Orders. American Journal of Hospital Pharmacy 46:426–434, 1989. Bates, D.W., Cullen, D.J., Laird, N., et al. Incidence of Adverse Drug Events and Potential Adverse Drug Events. Journal of the American Medical Association 274:29–34, 1995.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform Bord, R.J., and O'Connor, R.E. Risk Communication, Knowledge, and Attitudes: Explaining Reactions to a Technology Perceived as Risky. Risk Analysis 20:499–506, 1990. Brennan, T.A., Leape, L.L., Laird, N., et al. Incidence of Adverse Events in Hospitalized Patients: Results from the Harvard Medical Practice Study II. New England Journal of Medicine 324:370–376, 1991. Classen, D.C., Pestotnik, S.L., Evans, R.S., Burke, J.P. Computerized Surveillance of Adverse Drug Events in Hospital Patients. Journal of the American Medical Association 266(20):2847–2851, 1991. de Planque, E. Gail. The Regulation of Radiation Medicine: Back to Basics. Statement before IOM Committee for Review and Evaluation of the Medical Use Program of the Nuclear Regulatory Commission, October 13, 1994. Derrington, M.C., and Smith, G. A Review of Studies of Anesthetic Risk, Morbidity and Mortality. British Journal of Anesthesiology 59:815–833, 1987. Efron, E. The Apocalyptics. New York: Simon & Schuster, 1984. Erikson, K. Toxic Reckoning: Business Faces a New Kind of Fear. Harvard Business Review January–February:118–126, 1990. Erikson, K. Radiation's Lingering Dread. The Bulletin of the Atomic Scientists March:34–39, 1991. Evans, N., and Hope, C. Nuclear Power: Futures, Costs, and Benefits . Cambridge, England: Cambridge University Press, 1984. Fischhoff, B., Slovic, P., Lichtenstein, S., Read, S., and Combs, B. How Safe Is Safe Enough? A Psychometric Study of Attitudes Towards Technological Risks and Benefits. Policy Sciences 9:127–152, 1978. GAO (General Accounting Office). Nuclear Regulation: Better Criteria and Data Would Help Ensure Safety of Nuclear Materials. RCED-93-90. Washington, DC: General Accounting Office, April 1993. Guernsey, B.G., Ingram, N.B., Hokanson, J.A., et al. Pharmacists Dispensing Accuracy in a High Volume Out Patient Pharmacy Service: Focus on Risk Management . Drug Intelligence and Clinical Pharmacy 17:742–746, 1983. Heising, C.D., and George, V.P. Nuclear Financial Risk: Economy-Wide Costs of Reactor Accidents. Energy Policy 14:45–52, 1986. ICRP (International Commission on Radiological Protection). The Evaluation of Risks from Radiation. ICRP Publication 8. Oxford, England: Pergamon Press, 1966. ICRP. 1990 Recommendations of the International Commission on Radiological Protection. Annals of the ICRP 21(1–3):1–201, 1991. JCAHO (Joint Commission on Accreditation of Healthcare Organizations). Manual for Hospitals. Chicago, IL: Joint Commission on Accreditation of Healthcare Organizations, 1995. Kasperson, R.E., Renn, O., Slovic, P., et al. The Social Amplification of Risk: A Conceptual Framework. Risk Analysis 8:177–187, 1988. Koren, G., and Klein, N. Bias Against Negative Studies in Newspaper Reports of Medical Research. Journal of the American Medical Association 266:1824–1826, 1991. Kunreuther, H., Desvousges, W.H., and Slovic, P. Nevada's Predicament: Public Perceptions of Risk from the Proposed Nuclear Waste Repository. Environment 30(8):16–20, 30–33, 1988. Leape, L.L., Brennan, T.A., Laird, N., et al. The Nature of Adverse Events in Hospitalized Patients: Results from the Harvard Medical Practice Study II. New England Journal of Medicine 324:377–384, 1991.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform Linden, J.V., Paul, B., and Dressler, K.P. A Report of 104 Transfusion Errors in New York State. Transfusion 32:601–606, 1992. Manesse, H.R. Medication Use in an Imperfect World: Drug Misadventuring as an Issue of Public Policy: Part I. American Journal of Hospital Pharmacy 47:555–571, 1990. McNeil, B.J., Pauker, S.G., Sox, H.C., Jr., and Tversky, A. On the Elicitation of Preferences for Alternative Therapies. New England Journal of Medicine 306:1259–1262, 1982. NAS/NRC (National Academy of Sciences/National Research Council). Health Effects of Exposure to Low Levels of Ionizing Radiation: BEIR V. Washington, DC: National Academy Press, 1990. NAS/NRC. Science and Judgment in Risk Assessment. Washington, DC: National Academy Press, 1994. NCRP (National Council on Radiation Protection and Measurements). Recommendations on Limits for Exposure to Ionizing Radiation. NCRP Report No. 93. Bethesda, MD: National Council on Radiation Protection and Measurements, 1987. NCRP. Limitations for Exposure to Ionizing Radiation. NCRP Report No. 116. Bethesda, MD: National Council on Radiation Protection and Measurements, 1993. NRC (U.S. Nuclear Regulatory Commission). Task Force Report on Medical Radiation Protection. Rockville, MD: Office of Policy Planning, September 15, 1993. Perry, Z.A., and Knapp, D.E. Annual Adverse Drug Reaction Report: 1987. Rockville, MD: U.S. Food and Drug Administration, Office of Epidemiology and Biostatistics, Center for Drug Evaluations and Research, 1987. Pollycove, Myron. Memorandum to John E. Glenn: Rate of Radiotherapy Misadministration, March 8, 1993. Sazama, K. Reports of 355 Transfusion-Associated Deaths: 1976 Through 1985. Transfusion 30:583–590, 1990. Slovic, P. Perception of Risk. Science 236:280–285, 1987. Slovic, P. Perceived Risk, Trust, and Democracy. Risk Analysis 13(6):678, 1993. Slovic, P. Perception of Risk from Radiation. Paper commissioned by the Institute of Medicine Committee for Review and Evaluation of the Nuclear Regulatory Commission Medical Use Program. March 7, 1995. Slovic, P., Fischhoff, B., and Lichtenstein, S. Rating the Risks. Environment 21(3):14–20, 36–39, 1979. Slovic, P. Perception of Radiation Risk. In: W. K. Sinclair, ed. Proceedings of the 25th Annual Meeting of the National Council on Radiation Protection and Measurements. Bethesda, MD: National Council on Radiation Protection and Measurements, p. 79, 1990. Slovic, P., Flynn, J., and Layman, M. Perceived Risk, Trust, and the Politics of Nuclear Waste. Science 254:1603–1607, 1991a. Slovic, P., Layman, M., Kraus, N., Flynn, J., Chalmers, J., and Gesell, G. Perceived Risk, Stigma, and Potential Economic Impacts of a High-Level Nuclear Waste Repository in Nevada. Risk Analysis 11:683–696, 1991b. Smith, K. Perception of Risks Associated with Nuclear Power. Energy Environment Monitor 4(1):61–70, 1988. Surgenor, D.M., Wallace, E.L., Hao, S.H., et al. Collection and Transfusion of Blood in the United States, 1982–1988. New England Journal of Medicine 322(23):1646–1651, 1990.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform UNSCEAR (United Nations Scientific Committee on the Effects of Atomic Radiation). Report of the United Nations Scientific Committee on the Effects of Atomic Radiation. General Assembly Official Records: Seventeenth Session Supplement No. 16 (A/5216). New York, NY: United Nations, 1962. UNSCEAR. Report of the United Nations Scientific Committee on the Effects of Atomic Radiation. General Assembly Official Records: Nineteenth Session Supplement No. 14 (A/5814). New York, NY: United Nations, 1964. UNSCEAR. Sources and Effects of Ionizing Radiation: UNSCEAR 1994 Report to the General Assembly, with Scientific Annexes. New York, NY: United Nations, 1994. Weart, S. Nuclear Fear: A History of Images. Cambridge, MA: Harvard University Press, 1988. Welborn, J.L., and Hersch, J. Blood Transfusion Reactions: Which Are Life Threatening and Which Are Not? Postgraduate Medicine 90:125–138, 1991. Wilson, R. Analyzing the Daily Risks of Life. Technological Review , 81:40–46, 1979. Winslow, C.M., Solomon, D.H., Chassin, M.R., et al. The Appropriateness of Carotid Endarterectomy. New England Journal of Medicine 318(12):721–722, 1988a. Winslow, C.M., Kosecoff, J.B., Chassin, M.R., et al. The Appropriateness of Performing Coronary Artery Bypass Surgery. Journal of the American Medical Association 260(4):505–509, 1988b.

OCR for page 111
Radiation in Medicine: A Need for Regulatory Reform This page in the original is blank.