Development of a Standardized Approach to Disclosing Amyloid Imaging Research Results in Mild Cognitive Impairment

Protocol development

The process of protocol development began with an examination of consensus body recommendations for genetic counseling in AD and Huntington’s disease [3, 4]. The rationale for examining guidelines for genetic counseling was twofold. First, many of the issues associated with testing for genetically based risk for neurodegenerative disease, like the potential for psychological distress, also apply to testing for non-genetic biomarkers of AD, especially amyloid imaging. Second, consensus body recommendations for genetic testing in AD are informed by decades of empirical research on the most effective approaches to pre-test counseling and results disclosure, and reflect a process of rigorous discussion among a diverse group of stakeholders.

Upon review of these recommendations we identified elements of genetic counseling (e.g., information about testing, impact of testing, reflection) that would potentially apply to counseling sessions focused on amyloid imaging. We created a table summarizing our findings and highlighting areas of concern, which we circulated among a panel of experts from the fields of neuroimaging, bioethics, risk communication, neuropsychology, geriatric psychiatry, and regulatory affairs. The 9-member multidisciplinary panel engaged in an iterative discussion of the applicability of each element recommended for use in genetic counseling to the context amyloid imaging, and worked to identify additional elements that may be relevant to amyloid imaging but are not present in existing protocols for genetic counseling in AD. Content regarding reproductive decision making, for example, is standard in genetic counseling for AD, but was not deemed by the panel to represent key information to be conveyed during pre-test counseling for PET amyloid imaging. In contrast, pre-test counseling for amyloid imaging in symptomatic populations such as MCI requires a clear discussion of limitations of the testing, as is similarly necessary in genetic counseling. Specifically, there is a need for a clear explanation of the test’s specificity for AD as opposed to non-AD forms of dementia, which could represent alternative underlying causes of MCI symptoms for which amyloid imaging would not be informative. This process of analysis continued until a final set of core elements for amyloid imaging counseling was agreed upon by panel members.

Once the core elements for amyloid imaging counseling were identified, we created text to provide a basic description of PET amyloid imaging, potential results of amyloid imaging (positive, negative, and inconclusive), interpretation of such results in the context of MCI, and general recommendations for participants receiving each type of result. A set of visual aids to accompany this text was developed based on previous work by the Risk Evaluation and Education for Alzheimer’s Disease (REVEAL) team [5]. The same 9-member expert panel engaged in a second round of iterative review to reach consensus on the relevance and clarity of the text and visual aids that were developed or adapted.

Pilot testing

Although materials were developed for use in both pre- and post-test counseling sessions, pilot testing focused on the text and visual aids that would be used during amyloid imaging results disclosure. To assess the clarity and comprehensibility of the text and visual aids, 10 MCI care dyads (patient + family member) were recruited from the Alzheimer Disease Research Center (ADRC) at the University of Pittsburgh. All patient participants met ADRC enrollment criteria [6]. Those eligible for this pilot testing also had to: 1) carry a current ADRC consensus diagnosis of MCI (isolated impairment in memory; isolated deficit in non-memory domain; or, mild deficits in multiple cognitive domains) [7]; and 2) provide written informed consent to participate. The exclusion criteria were: 1) familial Alzheimer’s disease genetic mutation carriers (this group already has biomarker-derived knowledge of their dementia risk); and 2) previous participation in an amyloid imaging research study (to avoid any confusion over whether the hypothetical results are actually real). Family member participants had to: 1) be at least 18 years of age; 2) be English speaking; and 3) provide written informed consent to participate.

After providing informed consent, dyads were separated for the administration of a health literacy measure, the Newest Vital Sign [8]. Dyads then came together to undergo a simulated amyloid imaging results disclosure session using the test and visual aids developed through the above-described process. Simulation sessions began by instructing participants to imagine that the dyad member with MCI had recently undergone a new type of brain scan designed to detect the presence of AD brain pathology, and to imagine that they have come in to learn of the scan results. Participants were instructed to ask questions during the simulation and to make every attempt to conduct themselves as though the information being conveyed was real. Four of the dyads were randomized to receive hypothetically positive amyloid imaging scan results, four were randomized to hypothetically negative results, and two received hypothetically inconclusive results.

Following each simulation, patients with MCI and their family members were interviewed separately to assess comprehension of the hypothetical information conveyed and satisfaction with the interaction. Comprehension assessments were adapted from those used in the REVEAL studies and consisted of two open ended questions: 1) “Please tell me the mock results of your PET scan in your own words,” and 2) “Please explain what these results mean in terms of your (or your loved one’s) chances of developing Alzheimer’s disease or another form of dementia.” Responses to these questions were audiorecorded and transcribed verbatim. Blinded to the type of respondent (patient with MCI versus family member), an interdisciplinary team of four faculty and Master’s level staff from nursing, public health, neuropsychology, and bioethics, rated responses as demonstrating adequate or inadequate evidence of comprehension. Ratings were conducted using a consensus meeting format to reach agreement following a scoring rubric that the team developed as part of this process. According to the scoring rubric developed by this team, features of adequate responses included: Paraphrasing results and/or the explanation of their implications for AD risk; applying an appropriate value or direction (e.g., “high”) to AD risk or acknowledging the uncertainty of the information’s implications for one’s personal trajectory, often in combination with naming one’s result. A rating of inadequate evidence of comprehension was applied when a participant expressed frank confusion or offered a vague response to the comprehension questions.

Finally, with participants’ explicit consent, basic sociodemographic information and other sample characteristics were abstracted from each participant’s most recent ADRC visit record.

Protocol refinement

After completion of the pilot testing, we conducted a focus group with a subset of pilot study participants to engage these individuals in a broader discussion of the process of pre- and post- scan education and counseling. The focus group included four persons with MCI and four family members of persons with MCI, and included representation from individuals with a range of responses to the satisfaction and comprehension questions posed during the pilot testing phase of the study. Focus group attendees were not required to participate as dyads, although two families did so.

The focus group interview guide was designed and implemented following standard guidelines, including the use of semi-structured questions with cues and prompts to facilitate a rich exchange of viewpoints [10, 11]. Specifically, the focus group guide consisted of a brief recap of amyloid in MCI and 14 open-ended questions and 9 follow up probes designed to augment the satisfaction data collected in the pilot study and to determine what, if any, additional information and support might be beneficial to those contemplating amyloid imaging. The audiorecording of the focus group was transcribed verbatim, yielding 32 single spaced pages of discussion, the content of which was analyzed using qualitative description [12]. Consistent with this methodological approach, three investigators performed line-by-line coding of the transcript, iteratively generating and refining a set of descriptive categories to group codes based on their similarities. After individually coding the data, this team convened to review and compare categories, and to identify emerging themes within the data.

Pilot test findings

Characteristics of participants in the results disclosure simulation are described in Table 1. Of note, most (n = 7) of the family member participants, but only half (n = 5) of the patients with MCI had Newest Vital Sign scores indicative of adequate health literacy. Mean ratings of satisfaction with the simulated results disclosure session are presented in Table 2. All 20 participants agreed or strongly agreed that information in the session was “clearly presented.” Nineteen of 20 (95%) agreed or strongly agreed with statements indicating that the session was “easy to follow,” and “just about right” in length. Agreement with the statement “After the …  .session, I still had questions about my (or my loved one’s) mock results” varied widely among both patient and family member participants.

Based on the above-described scoring rubric for assessing comprehension of the information disclosed, 16 participants displayed adequate comprehension of their, or their loved one’s, hypothetical PET amyloid imaging results. Two patients with MCI and two family members provided responses that did not meet the rubric’s criteria for adequate comprehension. In each case that was rated as “inadequate,” the other member of the dyad did display adequate comprehension, underscoring the need to have at least two individuals present during results disclosure. Stated another way, there were no instances in which both members of a given dyad failed to meet our criteria for adequate comprehension. Scrutiny of the participant profiles of those displaying inadequate comprehension did not any reveal clear patterns with respect to characteristics of the participants (e.g., health literacy level or years of education) or the type of result presented (positive versus negative).

Focus group findings

Upon qualitative analysis of the focus group data, the theme “satisfaction with the disclosure process,” emerged quickly during the analysis, serving to validate the findings from our initial post-simulation satisfaction surveys, with both patient and care partner participants consistently expressing the view that the information presented during the simulated disclosure sessions was clear and comprehensible. In addition to this finding, two more major themes emerged during data analysis; the first was “recommendations for best practice” and the second was “knowledge is power.”

Within the first theme were six specific suggestions for best practice: 1) offer pre-test counseling, 2) use clear graphics, 3) review patients’ own brain images during disclosure, 4) offer take-home materials describing follow-up options, 5) call patients post-disclosure to address emerging questions,and 6) communicate seamlessly with primary care providers. Exemplar quotes for each of these recommendations is provided in Table 3.

The second theme, “knowledge is power,” reflects participants’ widely shared view that there was inherent value in learning one’s brain amyloid status, irrespective of the type of result that one might receive or whether that result would lead to a change in the treatment approach. Participants expressed this view with statements such as “Knowledge is strength. It helps you to do the best for the people you love and yourself” made by the husband of a woman with MCI. Similarly, a widowed woman with MCI described how she would use the information as a basis for future medical decision making “[using results] to write down what I would want to have happen [in the future]” and adding, “the unknown is sometimes the hardest.” While the majority of focus group participants expressed support for this perspective, the following quote from the wife of a participant with MCI shows the potential for uncertainty regarding the value of advanced diagnostic testing for AD pathology, “At least [with breast cancer testing] you can, they do a double mastectomy and reduce your risks, but so far with Alzheimer’s they have no idea what’s causing it, well very little idea. And there’s nothing they can do. So I’m not sure, I’ve never quite figured out if it’s worth it, to know. To know what’s happening because you can’t do anythingwith it.”

The notion of information’s inherent value, as distinct from and despite the lack of pragmatic use of the information, is exemplified by the following quote from a woman with MCI who participated in the study with her husband as her care partner, “[some people think] but nothing can be done about it yet [so] why do we want to burden ourselves knowing this is the direction we are going in. And our attitude is very different. We want to know as much as we can.” In another case, the value of an amyloid scan, if positive, was considered would be to serve as validation of the symptoms that one spousal care partner and his wife were witnessing: “ …  [if other family members had this information] they could no longer stick their head in the sand.” However, it was not clear whether a negative scan would be viewed as equally important.