All medical decision making is influenced by cognitive and affective responses. The complex interaction between the two has been referred to as a “dance of affect and reason” [44], whereby the cognitive aspects bring logic, reason, and scientific deliberation to bear, while the affective responses result in fast, instinctive, and intuitive reactions to perceived dangers. Indeed, facts may not be the most important aspect of patients' assessments of risk, which have been shown to be primarily determined by emotions [45, 46].

Affect influences virtually every aspect of human functioning, such as perception, attention, inference, learning, memory, goal choice, physiology, reflexes, and self-concept [47]. Many medical decisions involve intense affective states such as fear, anxiety, pain, discomfort, and diverse other moods and emotions [47]. These very real, rapid, and automatically experienced feelings are often used as information in the decision process [42, 44] and subsequently guide information processing and judgment. As such, they may skew an understanding of the facts, impairing the ability to make objective decisions about medical radiation.

Medical decision making would typically involve conscious decisions taken after much thought and discussion [48]. However, it is important to recognize that when we make decisions, we often choose an option as soon as we find one that is satisfactory, just as patients often extract the gist of any information rather than the details [49]. In high-stress situations, such as learning that one may need a medical procedure, especially one involving radiation, people may experience “mental noise,” an emotional block that can make it difficult to hear, understand, or remember information [50]. Such an affectively laden situation creates insensitivity to the often relatively low risks associated with the imaging [33, 44]. Anxieties can result in an inability to understand the information being conveyed and may result in a reduction of a patient's ability to attend to and retain information; in contrast, a risk taken voluntarily is less frightening than an imposed risk, and the more patients feel they can control unfamiliar situations, the less afraid they may be of the situation [42].

Feelings of dread and the unknown are major determiners of public perception and acceptance of risk for a wide range of hazards [44, 51] and may also be associated with factors such as voluntariness, controllability, lethality, and fairness [52]. Some researchers have suggested [52–54] that the perception of risks should be viewed as being based on hazard (danger) plus outrage (affective emotions about actions). Therefore, a mere presentation of technical facts alone will not necessarily give patients what they need or want.

Anticipated regret (based on a patient's predictions of option, outcome, or process regret before the choice and before outcomes occur) has been shown to influence medical decision making [48]. Cancer is such a feared event that anticipated regret may be influential in medical radiation imaging, even though risks are generally low. A patient under the influence of anticipated regret might think, “If I decide to have this series of imaging scans, I will surely regret getting cancer 20 years from now, even though I realize the chances are very low from the procedure. Therefore I better not have the scans, because I don't want to seriously regret this decision later.” Some patients may actually assume that all cancers in the future are attributed to the imaging and discount the large natural risk of cancer [33]. Many patients also have a general side effect aversion or an excessive sensitivity to negative side effects of medical treatments, which may lead them to refuse beneficial therapies [55], especially when those side effects are more difficult to evaluate, as in medical radiation imaging. Similarly, time perceptions while differentiating between short-term costs and benefits versus long-term benefits and risks play a role in medical radiation decision making [56].

The source of the information itself will often carry an innate bias [57]. Physicians typically know much about medical benefits and express benefits and risks in terms of medical results. Physicists typically know much about imaging and radiation delivery, focusing on keeping doses as low as reasonably achievable. Epidemiologists know much about risks and effects, often expressing risk in terms of effects in a population (not to an individual). Patients themselves may have individual bias because they may rely on nontraditional sources of information, particularly Internet sources of varying validity [58]. Although it is often acknowledged that health professionals typically provide the technical knowledge and patients provide the values, beliefs, and emotions, it is also true that medical professionals also have values, beliefs, and emotions that subtly affect how risks are communicated, and patients may also have some level of information, especially in this Internet age [59–61].

Covello has aptly stated that “people want to know you care before they care what you know” [43]. A medical professional should strive to display both a competent and caring approach, citing basic risk data in general terms, explaining the justification logic, addressing probabilities for benefit and risk outcomes, and expressing encouragement and hope [45, 62]. When evaluating risk perceptions, the most consistent dramatic finding appears to be the impact of trust and confidence on acceptability [43, 63]. The higher the level of trust, the lower the estimate of risk and the higher the estimate of benefits [64]. Indeed, if feelings toward medical radiation imaging are favorable, patients are likely moved toward judging the risks as low and the benefits as high; if their feelings toward it are unfavorable, they tend to judge the opposite—high risk and low benefit [44].

In an earlier era, the physician advised the patient what procedures and treatments were recommended, and the patient was expected to unquestionably follow such advice [58]. Such an approach can be summarized as the “you should have this procedure because I'm your doctor” method. Although it may be true that, in some situations where the patient is in an affectively aroused state, the more dispassionate perspective of a physician may provide a more stable basis for decision making and lead to decisions that are more consistent with the long-term interest of the patient [47], such an approach is no longer considered to be the standard of care [39, 40].

Attempts to overcome the public's total lack of familiarity with radiation have included risk comparisons using the concept of effective dose, which was introduced by the International Commission on Radiologic Protection in an attempt to provide a relative risk of radiation exposure [65–67]. Several comparison approaches are used, including comparisons with the background equivalent radiation time [68] in the United States or in Denver, Colorado; comparisons with the number of cross-country airplane trips; expressing the dose as some multiple of a common examination, such as a chest radiograph; and by comparing an examination's dose to the safety levels prescribed for occupational exposures. Table 1 gives examples of such approaches for a chest radiograph and a CT scan of the abdomen.

There are obvious problems with each comparison. For example, background radiation dose rates can be highly variable; elevation changes cosmic radiation exposure during air travel; and the effects of a onetime exposure can be expected to differ from the effects associated with long-term low-level exposures such as the occupational dose limits. Although such risk comparisons may provide illustrations to assist in comprehension of dose magnitude, they do not educate effectively because, from the standpoint of the individual, averages do not adequately capture the essence of such risks [20].

Many research institutions have developed lengthy “informed consent” forms that include a fair amount of what can be termed “risk numerology” (e.g., relative risks, excess cancer rates, increased rates over background levels, and log-based hazard comparisons). Many investigators have used a factor for the increased risk of cancer on the order of about 5% per sievert, a value initially based on overall population averages and prudence in safety guideline setting [33, 69]. In addition to offering an incomplete picture, there are difficulties with strict risk numerology. Even highly educated patients sometimes find it difficult to understand basic probability concepts or to perform even simple mathematic operations, such as comparing the magnitudes of two probabilities [55], and some patients neglect or are insensitive to probabilities altogether [44], perhaps basing decisions solely on the perceived numbers of listed benefits and harms [55]. In addition, comprehension becomes increasingly difficult as risk probability increases. For example, a probability of 1 in 2 with no background seems simple enough to understand, whereas the additional risk of 1 in 2000 on a background risk of 3 in 5 is difficult, if not impossible, to grasp.

Framing effects, which arise depending on the perspective from which data are presented, can influence the perception of risk. Wagner uses the following example [57]: After a pelvic CT scan of a pregnant woman, which statement delivers the most appropriate message about risk? “This study could perhaps double the risk that your child will develop cancer before age 19 (0.6% vs 0.3%),” or “The risk of adverse outcome is very small and the likelihood of normal development is nearly the same as it is for any child (96.7% vs 96.4%).” Both statements are technically accurate. The first focuses on the risk magnification that could result from the procedure. The numeric risk estimates attached to the statement support the assertion that risk is doubled but require interpretation by the patient to appreciate the relatively low level, both before and after, of risk magnification. In the second statement, the patient is assured that the overall risk of an adverse outcome is low, and the likelihood of a normal outcome is emphasized. Any deliberative framing of information raises ethical questions about the right to manipulate a patient's preferences in such a way [44].

Some institutions rely on what can be called “quality assurances” while communicating benefit and risk in medical radiation. Quality assurance statements such as, “We have the newest state-of-art imaging models,” “We test our machines in accordance with the American Association of Physicists in Medicine guidelines,” “We meet the American College of Radiology guidelines,” “The doses you will receive in our clinics are less than the typical average levels in the industry,” or “Our protocols are designed to deliver doses as low as reasonably achievable so you don't really need to worry about it” will not convey the proper overall perspective. Although these statements may indeed be true and may represent a healthy continuous improvement culture in the institution, they do not address benefits or risks in a quantifiable manner. As aids to understanding the active risk management posture of a radiology department, however, they may be reassuring.

It is important to develop simple clear messages at a sixth-grade reading level, devoid of jargon, technical terms, and acronyms [43, 70, 71]. In general, there should be no more than three key messages. Key points should be repeated frequently and should explicitly convey the conclusion [72]. The aim should be to simplify language and presentation, not content. Patients do not need to understand benefits and risks at the same level as the medical professional, but they do need to understand them well enough to assist them in making an informed decision [61].

Rather than avoiding the use of numbers, numbers and visual aids should be used extensively. Several communication alternatives have been shown to significantly increase the ability to evaluate a risk probability tradeoff: requiring less cognitive effort, using a graphical display [73], and expressing numbers in percentages rather than frequencies [72]. When communicating risks with patients, Paling [45] suggests that it is important to elaborate and provide estimated numbers, to avoid using descriptive terms only, to use standardized risk vocabulary, to use a consistent denominator, to offer both positive and negative outcome information [55], to use absolute numbers (rather than relative risks), and to use visual aids for probabilities. Using visual aids also helps to foster good physician-patient partnerships and can help the viewer to see the risk numbers in context, thus providing information and not simply data.

A visual aid may be very helpful in improving the dialogue about such risks, especially those based on dose proportions (ratio to a reference) [74], or what some have termed a “Richter Scale of Risk” [75], using a factor of 10 approach. Indeed, when discussing radiation risks, these should be described using broad categories such as those shown in Table 2 (e.g., negligible, < 0.1 mSv; minimal, 0.1–1 mSv; and minor, 1–10 mSv) [76–78].

Medical professionals should strive for an interactive dialogue between the medical professional and the patient. In addition to being credible and knowledgeable about the topic area, medical professionals should ideally be skilled in interpersonal communication, be able to convey empathy, use active and effective listening strategies, and be respectful of patient's concerns [71].

What the public or a patient may wish to know, and often has trouble absorbing, is the answer to specific questions such as, “Should I get this CT scan?” or “Should I agree to this radioactive stress test?” or “Is it OK for my 3-year-old to have another CT scan to rule out appendicitis?” [20, 79]. The answers to such questions are far more complex than estimates of the radiation detriment associated with the examination in question and need dialogue. The patient needs to understand the medical indication for the procedure, and referring clinicians should participate in the explanations as part of medical imaging decisions. The potential benefits of modern medical imaging procedures, which almost always far outweigh the associated risks, also need to be clearly discussed.

One approach is to explain what information might be obtained from the study and how not having the study could affect the patient. For example, to the patient who asks, “Should I agree to this cardiac stress test, which uses radioactivity?” one could answer that certain patient symptoms or risk factors have suggested that cardiac ischemia might be present. Modern imaging techniques permit us to obtain images of the heart at rest and after stress, and these images can be used to diagnose cardiac ischemia. If cardiac ischemia is diagnosed, doctors can plan further treatments aimed to treat the abnormality, including potentially life-preserving interventions. If cardiac ischemia is not present, doctors may need to pursue other diagnostic considerations to help the patient. A potential benefit of having the procedure is, therefore, obtaining a diagnosis for which a treatment strategy can be devised, possibly prolonging life. Conversely, the risk of not having the procedure might be failure to obtain this diagnosis.

It is important to capture and maintain the patient's attention. This is facilitated by using the strongest points at the beginning of the message and repeating them at the end of the message. During the dialogue, it is important to evaluate the patient's understanding. The medical professional should check the patient's eye contact to ensure that the person is still listening, be aware of postural changes in the patient, and obtain periodic feedback from the patient to ensure comprehension. When actively evaluating patient understanding, the medical professional is better able to hear and address emotions behind specific patient concerns.