Background
Biomarkers are biological indicators that can gauge the presence of a disease and can be used to support clinical diagnosis [
1]. One of the biomarkers used to support a diagnosis of Alzheimer’s Disease (AD) is based on brain amyloid, which can be visualized in situ by positron emission tomography (PET) and radiotracers that have a high affinity for binding to beta-amyloid plaques in vivo. PET imaging of beta-amyloid plaques in the brain has been approved for diagnostic purposes in cognitively impaired patients who are being evaluated for AD [
2,
3]. Besides being useful for clinical situations, PET imaging of beta-amyloid plaques is also being used in brain research on AD.
Little is known about patients’ motivations, opinions, and experiences in relation to their clinical trial participation and their IRR. In a research context, the degree to which individual research results (IRR) are shared with study participants varies, as the researcher is not obligated to do so [
3,
4]. Reasons to support the non-disclosure policy mostly relate to differences between the research and clinical context, the limited efficacy of current treatments, the experimental nature of certain types of data, and the limitations of lay participants to objectively and reasonably interpret the results [
5,
6]. On the other hand, some reasons to support IRR disclosure are the right of participants to know and be in control of information about their medical condition, it affords participants the opportunity to make informed lifestyle changes and practical personal and family arrangements, and it opens up the possibility for participants to enroll in (early) medical interventions [
5,
7,
8]. The previously mentioned reasons are based on the available theoretical literature and are mostly used by researchers to support or withhold IRR disclosure. Yet, what are the concrete reasons and motivations of research participants? Thus, the following questions arise. Why would participants opt for the possibility to be informed about their IRR? What do participants perceive as possible advantages and disadvantages of knowing their IRR? Clear answers to these questions will help guide researchers and experimental subjects in deciding which option—IRR disclosure or non-disclosure—to choose.
In this study, we focused on the reasons why patients with amnestic mild cognitive impairment (aMCI) were either in favor of or against the disclosure of their brain amyloid PET results. The experiences and opinions of these aMCI patients can differ from those of healthy individuals being screened for AD [
9]. We also focused on the disclosure of brain amyloid PET scan results, because these biomarker findings have been approved by the FDA (Food and Drug Administration) and EMA (European Medicines Agency) for clinical use in patients with a cognitive deficit [
10‐
12]. Our findings are based on interviews with aMCI patients before they received their IRR.
Methods
Recruitment
Recruitment took place between June 2015 and June 2016 after approval of the study by the Ethics Committee, University Hospitals Leuven. All participants provided written informed consent in accordance with the Declaration of Helsinki.
The study cohort consisted of a consecutive series of aMCI patients [
13] recruited via the memory clinic of the University Hospitals Leuven. The interview was part of a sub-study of the BioAdaptAD study, an investigator-driven longitudinal study of aMCI. The primary objective of the BioAdaptAD study (EUDRACT no. 2013–004671-12) was to evaluate the predictive value of baseline amyloid biomarker measurements for tracking longitudinal cognitive change over a two-year period. When candidate subjects met the inclusion criteria (see Appendix
1) of the BioAdaptAD study, they were given the option to participate in a sub-study investigating the opportunities to receive their IRR and the ethical challenges associated with it. More specifically, this sub-study provided the participants with the option of receiving their amyloid PET scan (binary visual read or positive/negative for brain amyloid) results that were obtained within the context of the BioAdaptAD study. We conducted semi-structured interviews with participants who agreed to take part in the sub-study in order to better understand their motivations, opinions, and experiences about the disclosure of their IRR.
Before the start of the sub-study, an informed consent brochure was given to candidate subjects. The content of this brochure was based on the E6 Guideline of Good Clinical Practice and contained background information about the study, study objectives, the interview process, and research subjects’ rights. Before the scheduled interview, the interviewer orally repeated the content of the informed consent brochure and asked the candidate whether they had any further questions. If they hesitated or had doubts about participating, the interview was re-scheduled to a later time to give the candidate sufficient time to decide.
Data Collection and Analysis
The interview guide was developed by GV, KD, and RV, and its content was based on findings in the literature on the topic of IRR. The first two interviews constituted a pilot study, which was used to evaluate the interview guide and make needed adjustments. The interview questions covered three content areas. The first part of the interview consisted of questions intended to help us better understand how patients describe and experience their current memory problems. Although patients’ description of their memory complaints were not intended to be part of the result section of this manuscript, these exploratory questions were used as opening questions for the interview and to provide the interviewer with some concrete patients’ experiences regarding their memory complaints. The second part consisted of open-ended questions about why they want to know their IRR result and what they perceive as possible (dis)advantages of their IRR disclosure. The third part consisted of hypothetical questions about how the patient thinks he would respond to a set of possible situations, such as: ‘What if the researcher informs you that you have a positive amyloid PET scan result? How do you think you will respond to this news? Several interview techniques were used, such as rephrasing part of the participant’s answer, asking yes or no questions, to briefly check whether the interviewer understood the participant’s answer correctly. Supplementary questions were also asked to get more in-depth information from the participant.
After a short introduction of what the research was all about and what to expect in the interviews, patients were invited to sign the informed consent form (IC) and then to complete the sociodemographic information form. Completion of the IC form indicated that the patient understood that his/her results and interview records would remain confidential, that participation in the interview was voluntary, and that it would have no impact on his/her participation in the general study or on any other medical intervention he/she might undergo at the hospital. Participants were informed that results of the study would be published in a scientific journal and that a lay description of results from the interviews would be provided to them after completion of the study.
The interviews were recorded on tape with the consent of the interviewee. A mixed-method approach was used to analyze the interviews. (1) Transcripts were analyzed using QSR International’s Nvivo 10 software and was performed according to qualitative conventional content analysis methodology [
14,
15]. In the first phase, the interviewer (GV) coded the interviews three separate times, with an interval of a few weeks between each coding session. Letting time pass between each session and re-coding allows one to check whether new interpretations might be attributed to the content that could have been missed in just one coding session. To protect against bias, in the second analysis phase, we had five interviews independently coded by a second researcher. In the final phase, we compared the codes assigned by the interviewer and those done by the independent researcher (KD), working to reach a consensus on the final codes to be used. (2) In addition, we used a quantitative approach to analyze the reasons, benefits, and disadvantages of receiving an IRR, as provided by the participants.
Interviews were conducted in Dutch, with the exception of one, which was conducted in English. For this interview, the patient preferred to speak in his native language, English. Quotations of patients presented here have been translated into English.
Discussion
In this study, we sought to document and better understand the reasons why aMCI patients participating in a research study were either in favor of or against the disclosure of their individual brain amyloid PET scan results and what advantages/disadvantages they perceived about IRR disclosure/non-disclosure. The two most frequently mentioned reasons for choosing IRR disclosure were to better understand their own brain health status and to be able to make informed decisions about future personal arrangements they might have to make if their PET scans were consistent with the development of AD. Emotional risk was mentioned as the primary disadvantage of knowing one’s IRR.
In the field of AD research, not a lot of research on brain amyloid IRR disclosure has been done. However, a number of studies focused in particular on the desire of participants to opt for genetic AD risk assessment [
16‐
18]. For example, a telephone survey in a community-recruited sample concluded that 79% of respondents expressed interest in predictive genetic testing for AD [
18]. Another recent survey in cognitively normal older adults by Gooblar et al. indicated that 97% of participants had a strong interest in obtaining their research results, including those related to the detection of the molecular pathology of AD, such as amyloid imaging [
19]. In all previously mentioned studies only a minority of participants preferred not to know their IRR. In our study, all 38 of the aMCI patients (i.e., 100% of participants) wanted to know their brain amyloid IRR. Our sample was quite diverse demographically, varying in age from 55 to 83 years (mean 71 years; 58% men), most being married with children, and most finishing secondary school or higher. An important limitation is that all 38 aMCI patients were recruited via the memory clinic and already took first initiative to receive medical advice.
Although all of our participants wanted to know their IRR, two caveats should be mentioned about these results. One participant initially did not want to know his IRR but later on changed his mind. Another participant stated that he did not want to know his IRR but said that his wife really wanted to know. In this case, we explained to him and his wife together what this would mean practically and how they might deal with the situation in which one spouse knew the results and the other did not. With additional time to reflect, this participant decided that he wanted to know the results. We carefully queried this participant’s reasoning to determine whether he was “pushed” into his decision for IRR disclosure by his wife, or whether he made the decision independently.
This high percentage of aMCI patients wanting to receive their IRR suggests that in the near future more participants might be open to participating in research if they have the possibility of receiving their IRR. However, researchers should be aware that if the release of IRR to subjects and patients were to become commonplace, it might attract people with subjective mental complaints or hypochondriacs to volunteer for research studies, which would then negatively affect not only recruiting criteria and the recruiting process but also research results. Thus, more research is needed on how participants might volunteer for a clinical trial, but opt out of receiving their IRR. Although in most studies, this group is smaller than the group of participants who do want to know, insight into the opinions and perspectives of the people who opt out of wanting to know is equally important for the IRR disclosure/non-disclosure debate.
The predictive value of brain amyloid PET scans of aMCI patients for progression to AD in the next years has certain limitations. This is an ethical issue being debated in the general literature about IRR disclosure [
20‐
22] and is a point highlighted in our information brochure. Our hypothesis was that the limitations of the predictive value of brain amyloid PET scans would be perceived as a disadvantage by participants for wanting to know this particular type of IRR. No participants mentioned this as a disadvantage, even though they presumably read about it in the brochure. To temper the possibility that participants were providing socially desirable answers, we asked more in-depth follow-up questions about this limited predictive value. Still, it was clear that participants did not perceive this as a disadvantage. Participants with aMCI still perceived that it was better to know if they had a high or low risk of developing AD than to remain completely “in the dark” about their current health situation. In addition, our results indicate how some participants perceived this uncertainty as a positive aspect, since it leaves open the possibility for hope. Nevertheless, accurate and clear information about IRR and its associated limited predictive value remains important in order to prevent patients from misunderstanding this information.
Another hypothesis we had was that participants would mention the limited efficacy of current treatment options as a disadvantage of knowing their IRR. This hypothesis was based on published reports suggesting that available treatment options are perceived as an ethical issue [
8,
20,
22]. This was not mentioned as a disadvantage by any of our participants. When asked, participants answered that they already knew before the start of the study that there is no cure available for AD, or that there is medication available but that it can only delay progression of the disease. The latter was perceived as a better option than receiving no medication at all. Some participants referred to the situation of cancer and its treatment, in which medicine can already do a lot, but is not always able to cure the disease. Most participants maintained strong hope that in the near future the situation will change for early treatment of AD, whereby researchers can rapidly make medical progress.
The concept of therapeutic misconception is important to consider when trying to understand participants’ decisions in choosing disclosure/non-disclosure of IRR. This concept, which is often mentioned by the general literature about IRR disclosure [
5,
6,
23,
24], refers to the situation in which patients often fail to distinguish between clinical care and the research setting. Participants believe the researcher is a physician who has two core ethical obligations: (1) endeavor to avoid non-maleficence (no harm), and (2) work toward beneficence (doing well). “Doing well” can be understood as taking care of the health and well-being of the patient by providing necessary therapy and drug treatments [
23,
25]. In this case, patients who become participants in clinical trials often expect or have high hopes that they will receive treatment. As described in our findings, participants do perceive the option for treatment as a reason to know their IRR. Thus, the therapeutic misconception also occurred in our study. Two nuances should be addressed.
First, described in the informed consent brochure as mandatory by the International Guidelines of Good Clinical Practice (ICH GCP) [
26] and explained by our researchers to the study participant, was that participants would receive cholinesterase inhibitors if they had a positive amyloid PET scan. Thus, this knowledge likely would affect a participant’s decision about IRR disclosure. For a negative amyloid PET scan, this medication would not be prescribed, because it is less effective when the underlying pathology is not related to AD. Second, the concept of treatment was interpreted widely by participants and was not limited to medication. Follow-up, disease monitoring, and imposition of health-related life style changes (e.g., taking food supplements, cessation of drinking, etc.) were also thought of as treatment. This broad understanding also likely affected participants’ decisions about IRR disclosure.
The majority of our participants reflected on planning for the future as an advantage of knowing their IRR. This finding is in line with other AD studies, such as the Risk Evaluation and Education for Alzheimer’s Disease study (REVEAL) who reported on the importance of making arrangements for the future [
16,
27,
28]. In our study, almost one-in-five participants also mentioned end-of-life decisions as playing a prominent role in their reasoning about IRR, more specifically the request for euthanasia, given a positive brain amyloid result. The participants who mentioned this possibility in our study might have been influenced by ongoing legal discussions in Belgium. First since 2002, euthanasia can be legally requested in Belgium under strict circumstances (Belgian Euthanasia Act, 28th of May 2002) [
29]. These circumstances, imply that: (1) the patient is legally competent and conscious at the moment of making the request; (2) the request is voluntary, well-considered and repeated, and is not the result of any external pressure; (3) the patient is in a medically futile condition of constant and unbearable physical or mental suffering that cannot be alleviated, resulting from a serious and incurable disorder caused by illness or accident (section 3, §1) [
29]. Second in past years, there has been a growing awareness in Belgium that healthcare institutions bear the responsibility for translating these legal regulations into optimal care for patients requesting euthanasia. For example, enhancing communication toward patients and their relatives about end-of-life decisions and early care planning [
30,
31]. To properly compare the situation in Belgium with that in other countries, research is needed to evaluate whether euthanasia might be considered as a reason for wanting to know one’s IRR, and if this would lead to an increase in euthanasia requests.
aMCI participants in our study mentioned euthanasia as a practical arrangement; two ethical concerns emerged from this. First, most participants explained how they do not want to become a burden to their family members. Thus, euthanasia was perceived as a solution, whereby family members would not be burdened with a beloved one who has AD. A person’s source of “burden awareness” is an important consideration: In a euthanasia request, did the patient independently arrive at the awareness that he may become a burden, or did he become aware of it by discussions with relatives? According to the Belgian Act on Euthanasia, the request needs to be voluntary, well considered, and not the result of any external pressure [
29].
Although there is a worldwide increase in the burden of caring for people with dementia, the solution for this problem should not be euthanasia but proper help throughout the care process from the perspective of both patients and caregivers. Government and healthcare regulations ought to (a) invest in damping down the perception that patients feel they are or will become a burden to their family members and in lifting the actual burden family members currently experience; and (b) maximize proper care that enhances the capabilities of patients and provide respite to ensure that the life of caregivers is not put “on hold” because they are providing care for their beloved one with dementia.
A second ethical concern that relates to aMCI participants in our study mentioning euthanasia as a practical arrangement is that participants feared living in a vegetative state, not wanting to “end up living like a plant.” This indicates how there is a negative perception or stigma about AD. Only one participant indicated that having AD does not necessarily mean that they will be living in an unhappy world.
Our findings also reveal how our participants are unclear about the legislation on end-of-life decisions in Belgium. This resulted in the use of the term euthanasia in general terms and showed that they were unaware of the different types of documents Belgian legislation provides. In Belgium, there are five different documents regarding advanced-care planning and end-of-life decisions [
32]. Three of these documents are important, and our participants appear to be confused about them. The first one addresses negative advance declaration about the refusal of certain treatment options; the second addresses the stipulation that advance declaration regarding euthanasia is a document that has to be prepared in advance and has to be renewed every five years. Euthanasia will only be performed in case of irreversible coma, which excludes dementia. The third addresses an active request for euthanasia [
32]. The above-mentioned criteria of the Belgian Euthanasia Act also need to be followed. It is clear that our participants confuse issues germane to these three documents. Many participants think that the advance declaration regarding euthanasia is sufficient if they receive an AD diagnosis. In reality, if a patient with an AD diagnosis wants euthanasia to be performed, he has to declare an active request for euthanasia at the earliest stage of the disease when he is still capable of expressing his wishes. This implies that the patient actively chooses to shorten his life by certain good months/years before cognitive deterioration becomes worse.
There was also confusion among some participants about the terminology used for positive and negative amyloid PET scans. Some patients understood a positive scan to be good news, a report with a favorable outcome for future brain health. Although the correct terminology was explained in the informed consent brochure and throughout the information session when the researcher explained the content of the trial face-to-face with the patient, we noticed that throughout the interviews, patients misused or misunderstood the terminology. If the interviewer noticed that a participant misunderstood the terminology, she elaborated on the point with the aim of avoiding the situation in which the interviewee and interviewer were talking about “different” results. Thus to clarify the terminology, we compared the amyloid test to an alcohol test: Having too much alcohol in the blood results in a positive alcohol test. In the same way, having too many amyloid plaques in your brain results in a positive amyloid test. Most of our participants understood this comparison. For patients to understand this type of research result, it is imperative to explain it clearly in a simple and correct way.