We conducted several literature reviewsusing the STARLITE method, to examine patient and non-radiologist physician knowledge regarding the risk of low-dose radiation, the patient physician dialogue and the sources of information they use, as well as the frequency and topics of the discussions. PubMed and MEDLINE databases were searched to include articles from 2007 to 2017. A wide array of search terms was employed to identify articles based on the specific question to be answered. Peer-reviewed, original research articles were the focus of the review.
Three literature reviews were conducted to acknowledge each aspect and served to answer the following questions:
1) How well do patients understand low-dose radiation and its risks?
2) How well do physicians understand low-dose radiation and its risks?
3) Do patients and physicians discuss the risks of low-dose radiation?
I. PATIENT KNOWLEDGE OF LOW-DOSE RADIATION AND ITS RISKS
The first component to of the study was to examine is patient knowledge and understanding of radiation risk. Several key indicators were evaluated. These indicators identified patient awareness of carcinogenic risk of radiation, and which modalities are ionizing between XR, CT, MR and US.In the current investigation, 10 articles were found to be relevant to this component using the search keywords “patient knowledge radiation” and “patient understanding radiation” and a summary of these articles can be seen in Table 1.
Lam et al. discovered a weighted average of 88% (230/260) of patients that underestimated doses or believed that ionizing radiation did not cause cancer [3]. Some studies were restricted only to a sole modality [21], [22], [23], [24], typically CT. While some patients are aware that XR or CT use ionizing radiation [25], [26], [22], [15], [27], the studies expressed discontentment with the totals. Those that are aware often underestimate doses [25], [21]. Patients fail to recognize that MRI does not utilize low-dose ionizing radiation [25], [26], [28], [23], [15], [34] and small percentages even believe U/S is ionizing as well [23], [33], [15], [36]. Patients are also unable to accurately compare and estimate doses between modalities [35].
In Takakuwa et al.’s study [21], 79% (/383) of patients estimated the risk of cancer from a chest x-ray scan to be none, very small, or small, compared to 83% of patients who estimated the same for a chest CT. Overall, the data suggest that patients are not aware of the potential for cancer and this gap should be addressed. This is echoed in other work [15]. One study clearly identified that there has been a significant improvement in patient knowledge from 2002 to 2010 [22]. In 2002, 3% (2/76) of patients believed that CT can increase the overall risk of cancer, but the current study found that 25% (50/200) believed it to be associated with cancer [22].
Patients should be informed of which modalities do and do not use ionizing radiation and that ionizing radiation does carry a carcinogenic potential. Furthermore, patients should be educated about the doses associated with their examinations and the “risk versus benefit” concept explained to them.
II. PHSICIAN KNOWLEDGE OF LOW-DOSE RADIATION AND ITS RISKS
The same key indicators were evaluated as was done with the question, “Patient Knowledge of Low-Dose Radiation and its Risks.”. An indicator was added to determine if the physicians understandood if they underestimate the dose, and if it could lead them to prescribe more ionizing imaging studies. The search terms used to reveal this section were “physician knowledge radiation” and “physician understanding radiation” and a summary of the articles found can be seen in Table 2.
The majority of studies indicated that non-radiologist physicians underestimate the doses associated with ionizing radiation examinations [29], [30], [31],[32], [33], [34], [35], [36]. Only two studies featured physicians who were able to successfully identify that U/S does not use any ionizing radiation [36], [32] and only one for MRI [36]. It is interesting to note that emergency residents were the only non-radiologist physicians who were successful at both [36]. Some studies did ask the physicians if there is a cancer risk associated with the imaging studies [28], [29], [30], [31], [37], [34]. However, in most instances, physicians are unable to complete correctly stating that any amount of radiation dose does indeed increase the risk of ionizing radiation as per the LNT model [37], [33], [34], [39], [40], [36].
Madrigano et al. asked, “Is there a dose threshold below which the exposure is safe?” to which 67.5% (81/120) of non-radiologist physicians incorrectly answered, “Yes”. This idea of a safe dose threshold is likely why in some instances, despite the fact physicians are able to fairly successfully identify that XR and CT use ionizing radiation, they are unable to successfully say that radiation is carcinogenic at any level, resulting in discrepancies between the “Radiation is carcinogenic” and “XR-” or “CT is ionizing” columns.
These surveys had the limitation of featuring quite small sample sizes. Brown & Jones’ study [34] had the largest population of 603 respondents. Also, because some studies did not ensure that physicians were aware of the cancer risk associated with XR and CT, those results may call into question physician’s understanding of risks associated with these modalities. Moreover, when analyzing the fact physicians are to inform patients of these risks, it is easy to wonder if these conversations are occurring if the knowledge they need is lacking.
III. PATIENT-PHYSICIAN DIALOGUE OF RISKS ASSOCIATED WITH LOW-DOSE RADIATION
This questionwas divided into patient and physician perspectives. The indicators used here provide insight on the frequency, content, and medium of the discussions. These indicators were chosen to determine if the discussion is occurring every time, whether the risks are being divulged, and if the discussions were in fact oral discussions or perhaps facilitated by digital tools. The search terms used were “patient radiation discussion” and “physician radiation discussion” and a summary of the articles found can be seen in Table 3.
1) Patient Perspective: Nine relevant articles were identified for the analysis of the patient physician dialogue from the patient perspective.
Bohl et al.’s study [24] involved patients that were referred to spinal surgery and required imaging. Although the risk of XR is lower than that of CT, interestingly enough 34% (34/100) of XR patients were informed compared to 26% (26/100) of CT patients. The lowest value was identified in Ukkola et al.’s survey [340], where only 3.4% (5/147) patients were informed of the dose and/or risks or the consequences if not done. A sole study mentioned that the physician explicitly stated the benefits of having the study [26]. None of the studies explicitly stated that cancer risks were discussed.
Five studies [14], [15], [21], [26] determined that patients have a fairly strong desire to have a discussion with their physician. Some patients want the discussion to allow for some mutual decision making with their physician [15]. Physicians should explain the risk every time patients are about to undergo an imaging exam. Exceptions to this could perhaps be in an emergency situation where the time of the imaging is crucial. If time allows, 52.9% (90/170) of patients did indicate that they were interested in speaking to a radiologist prior to their examination [28].
For many patients, the referring physician is the major source of information and if an ineffective dialogue takes place, then there will remain a massive gap in the patient’s comprehension of the imaging examination. This can prevent patients from being able to request an alternate imaging modality, refuse an examination that they are uncomfortable with, or just asking general questions to improve their understanding about the examination.
Patients also obtained information from nurses [37], [340], [15], radiographers [37], [40], [15], the Internet [37], [27], [15], [22], direct mailing of letters [39], books [15], pamphlets [37], [15], or family and friends [27], [15]. This shows a very diverse amount of information sources, but ultimately the patient-physician dialogue should prevail.
2) Physician Perspective: Barbic et al. [39] evaluated emergency physicians’ experiences with the radiation risk discussion. 27.5% recalled being asked, “Will this give me cancer?” by a patient. Communication with patients about CT risks is also subpar, as 32.7% reported having a discussion most or all of the time, 60.8% occasionally or rarely, and 2.0% for special populations or if explicitly asked. Since these results are quite similar to patient reported percentages, there is no response bias evident here.
Brown & Jones [34] found that for CT imaging, 12.7% of physicians reported to have discussed the risks every time a study is ordered. 56.9% of physicians will have a discussion when they feel it is appropriate, 8.4% when asked, and 22% reported to never have a discussion.
Horowitz et al. [41] studied if an educational intervention can improve residents’ discussion with patients regarding radiation risk. A radiologist provided a one-hour lecture on risk, carcinogenesis, LNT, and discourse methods to disseminate the risk when a study is ordered. Prior to the intervention, 48% (10/21) of residents stated that patients had asked about the radiation issues. Post intervention, 29% (6/21) of physicians stated that patients had asked about the risks. As follows, 38% (8/21) of residents stated that they used information from the radiologist’s lecture, including: “risk of malignancy”, “amount of radiation in a CT scan”, and, “more informal discussion of risks and radiation.” So while there was a lower number of questions asked than use of data, it suggests that physicians may be instigating the conversation in 10% (2/21) of the cases. Overall, Horowitz et al.’s study appears to suggest that the patient should be instigating the conversation, but the physician should possess sufficient knowledge to answer the questions.
Ditkofsky et al. [35] directly examined provider comfort level when discussing radiation risks. Their study, akin to Kruger et al.’s [42], included physicians, residents, nurse practitioners, and physician assistants, but only the initial two are most relevant. First, Ditkofsky et al. examined how comfortable the providers are at explaining the risks of radiation exposure. 17.07% of physicians felt extremely comfortable, 68.29% somewhat uncomfortable, and 14.63% are not very comfortable. 6.25% residents indicated feeling uncomfortable, 31.25% not very comfortable, 53.13% somewhat comfortable, and 9.38% extremely comfortable.
Stickrath et al. [43] surveyed fourth-year medical students, attending physicians, and house-staff from internal medicine, emergency medicine, radiology, cardiology, and pulmonary services. Stickrath et al. identified that 71% (212/300) of providers indicated that they discussed the risks of a CT scan 25% of the time or less. 27% (81/300) respondents felt comfortable discussing the risks associated with the imaging exams. Time was indicated as the largest pressure to prevent a discussion regarding risks in 42% (125/298) of providers. In addition, the physicians did not feel that informed consent should be obtained prior to ordering a CT scan (72%, 215/299). Stickrath et al. identify that physician education could improve the physician’s comfortability to discuss the risks with their patients and improve the potential dialogue.
Overall, discussions between physicians and patients regarding radiation risks are not occurring at an acceptable rate, with majority of studies indicating either never, usually not, or some discussion. Physicians and any other clinicians should have these conversations, but there is a gap in the ability to commence these dialogues due to their discomfort with the topic [44], [40]. While some patients had indicated an interest in gaining information from a radiologist [25] or a radiographer [26], the majority of patients wished to be informed by their referring physician.
