Our results suggest several reasons why participants decline optional brain MRI scans. While claustrophobia, MRI safety, busy MRI scanner, and discomfort were the main reasons for not completing the MRI scan, MRI being too far, sickness, and no perceived benefit from the scan were among the other reasons. Participants reported wanting to help future patients, contributing to research, and general curiosity as reasons to complete the scans. Younger and more educated participants were more likely to decline at least one scan, whereas older participants were more likely to complete at least one scan. To our knowledge, this is the first study to discuss barriers and facilitators that patients with cancer face when choosing to complete optional brain MRI scans and to offer suggestions that may help improve study participation.
Our results align with previous findings about barriers faced by patients undergoing MRI more generally. While few studies were looking at barriers to MRI study participation in cancer research, some studies discussed ways to improve the MRI scan experience. Similar to our studies, key challenges were enclosed space and scan duration; but unlike our studies, participants in prior studies also reported scanner noise as a barrier; all of which were related to claustrophobia, pulmonary symptoms, and other existing comorbidities [23]. Other non-cancer related studies in neuropathic pain [24] and autism [25] reporting on participants’ acceptability of MRI scans reported barriers such as noise and scan duration, and also mentioned the importance of accessible communication and pre-scan familiarization of the MRI scanner and what to expect during the scan.
Although mandatory MRI scans are feasible in cancer research, it may create research limitations. Our results suggest that when the MRI scanning is mandatory the study is still feasible, given that only 24% of required scans were not completed. This 24% missing data is similar to the amount of missing data for other outcomes measured in supportive care in cancer such as daily diaries (20% missing) [26], actigraphy (22% missing), and overall study retention (26% attrition across 18 studies) [27]. However, for behavioral interventional studies that may benefit patients, such as all three studies described here, mandating MRI scanning may inadvertently introduce research inequities by excluding patients who are claustrophobic or not safe for MRI. For example, given that cancer and its treatment cause anxiety [28], patients with cancer may be more likely to experience claustrophobia [29] preventing those patients from doing a brain MRI scan. It is important that behavioral interventional studies targeted at improving supportive care include these patients, as these interventions may show beneficial effects in reducing symptoms such as anxiety [30]. Moreover, additional MRI-related study exclusions can lead to slower progress in study recruitment, especially if such studies are conducted in rural areas, which tend to be further from MRI scanners [31].
The participant's perceptions and understandings of research studies may influence study participation. Studies have suggested that willingness to undergo MRI can be improved by an individual’s level of health literacy, thoroughness of education provided regarding the procedure and its purpose, and the interpersonal relationships developed with study coordinators [32]. These subjective insights align closely with existing literature on research participation more broadly (not just MRI). Factors such as higher levels of health literacy, being younger, female, or having more education are positively associated with interest and participation in clinical research [33, 34]. Conversely, low health literacy is a barrier to participation in research and optional aspects of a study such as an MRI scan. However, our data suggest that younger and more educated participants were more likely to decline scanning, perhaps due to busier lifestyles and career responsibilities, whereas older participants were more likely to be retired.
Our study suggests ways to improve participation in brain imaging research in cancer. As our results suggest, most patients do not complete the MRI scans because of anxiety, discomfort, a busy MRI schedule, inability to travel to the MRI scanner, and MRI contraindications. Here, we offer possible solutions to these barriers (Table 4). 1) Improve communication about the study, the MRI, the procedures, what to expect during the scan, and how these measures contribute to providing better supportive care. Educating patients regarding MRI procedures and their purpose may improve study participation by reducing anxiety, promoting acceptability, and increasing willingness to help other patients. Study participants who feel that they have been informed about MRI report higher levels of acceptability, satisfaction, and utility of MRI [33]. Indeed, educational materials and other communication, presented at the 6th-grade reading level, have been found to improve comprehension, anxiety, and levels of satisfaction [34–36]. Furthermore, providing patients with information about MRI data’s purpose to research itself further encourages participation by appealing to participants’ curiosity and desire to contribute to high-quality outcomes. Although the risk associated with participating in an MRI study is often well-communicated to participants, it is also important to clarify the benefits of the MRI to the participant (if any), as well as how it contributes to helping future patients. 2) Provide assistance and incentives to travel to the MRI scanner. 3) Ensure that the population and study time points do not exclude people unnecessarily. For example, patients with breast cancer often undergo breast reconstruction surgeries that require implanting a chest expander, which is not MRI-safe. 4) In instances described above, explore other comparable but better-suited alternative and/or portable options instead of MRI, such as functional near-infrared spectroscopy (fNIRS) or EEG.
Even though MRI is the gold-standard for brain imaging in research, when possible, considering other brain imaging modalities such as EEG or fNIRS can help improve feasibility, research inclusivity, and accessibility in cancer studies. The relatively mobile setup of these techniques may alleviate the anxiety and claustrophobia experienced during MRI scanning. Moreover, unlike in MRI, the use of fNIRS and EEG does not have any known safety concerns. fNIRS is an optical brain imaging technique that non-invasively quantifies brain activity via changes in blood oxygenation (similar to fMRI’s BOLD signal) [35, 36]. It is a widely used alternative to fMRI in various domains, including pain [37], pediatrics [38], and cognitive research [39] due to its superior cost, acceptability, and portability. fNIRS has been brought to patients in clinics [40, 41] and rural areas [42], and used during walking [43]. These types of imaging modalities may have a future alongside routine clinical assessment (like echocardiogram or ultrasound) to formulate brain-based treatments, risk prediction of symptoms, or their chronicity and inform treatment decisions (e.g., whether to dose-reduce chemotherapy to avoid chronic CIPN). Although fNIRS or EEG may be a good alternative to fMRI in cancer supportive care research, depending on the research question, a pilot MRI study may be necessary to determine the brain areas of interest, as these imaging techniques may be most useful in investigating the superficial cortical areas accessible by fNIRS and EEG.
Strengths. Our analysis provides a better understanding of the barriers and facilitators patients with cancer face when participating in an MRI study. These findings are noteworthy as our study had a moderate sample size (N = 72) and used both descriptive and quantitative methods to understand the impact of patient-related barriers on MRI research studies. We have explored these effects based on study design, mandatory MRI vs. optional MRI, and have asked participants reasons for or for not completing the MRI scan. In addition, our manuscript provides concrete action items to address barriers reported by patients, that may be helpful for other researchers interested in conducting an MRI study in patients with cancer.
Limitations. First, there are missing data specifying the reason for not completing the MRI scan because these questions were only added later into data collection. However, our moderate sample size was sufficient to gain a reasonable understanding of the common barriers to completing an MRI scan. Moreover, although a larger sample size might have shown additional barriers, they are likely to be less common (e.g., scanner noise, which we did not observe here, or other personal responsibilities). Second, the barriers discussed in this paper are more applicable to patients in urban areas in the US, with better access to an MRI scanning facility and potentially more likely to have better research and/or medical literacy. Conducting similar MRI studies in more rural areas may present researchers with more or different barriers in terms of study participation as well as available MRI scanner technological capabilities.
Future work. We believe that although having an MRI as an optional component of our study leads to more participants declining the scan, studies can be designed to include MRI as long as the sample size accounts for an estimate of what percentage of individuals will complete optional MRI. While we suggest strategies for future studies (Table 4), we will also explore quantitative analysis that will be important to better predict the completion of the MRI scans based on factors such as pre-existing conditions, level of education or research literacy, age, other participant obligations, geographic location (rural vs. suburban), and psychological wellbeing.