Abstract
Recommendations for communicating Alzheimer’s disease (AD) biomarkers include pre-disclosure participant education and counseling, to allow individuals to make an informed decision. In a cohort of largely non-Hispanic White, cognitively unimpaired older adults from the Wisconsin Registry for Alzheimer’s Prevention, we conducted a structured amyloid PET disclosure process that included knowledge assessment and education. Baseline participant knowledge about AD biomarkers and research was high, but information needs existed around dementia causes, early AD symptoms, genetic information, and psychosocial consequences of disclosure. Knowledge scores increased after education, highlighting the potential of brief educational interventions to improve informed decision-making about biomarker disclosure.
Keywords
INTRODUCTION
Alzheimer’s disease (AD) biomarkers are routinely collected in longitudinal cohort studies and clinical trials. Increasingly, studies are returning, or “disclosing,” biomarker test results to research participants [1]. With advancements towards accurate, noninvasive blood-based biomarker testing and recently available disease-modifying therapies, disclosure is expected to become more widespread in research and clinical settings in the coming years [2, 3].
In light of the potentially significant impacts of biomarker testing on individuals, such as psychological distress, stigmatization, or discrimination (e.g., inability to access care insurance), current guidelines for AD biomarker disclosure recommend offering education before testing and returning results [4 –7]. Evidence-based approaches to education and communication are essential to effectively disclose biomarker results, minimize negative psychosocial outcomes, and maximize potential benefits [8]. However, there is very little evidence about how much knowledge individuals have prior to making a decision whether to learn their results, or about the efficacy of education on participant knowledge. We present a novel study with an observational cohort of cognitively unimpaired research participants examining the efficacy of an educational intervention on AD and biomarker knowledge as part of a structured amyloid positron emission tomography (PET) disclosure process.
METHODS
Participants
Participants were recruited into the Amyloid Disclosure Study (ADS) from the Wisconsin Registry for Alzheimer’s Prevention (WRAP), a longitudinal observational study enriched for parental family history of AD clinical syndrome [9]. Eligibility criteria included being aged 65–89, cognitively unimpaired (no diagnosis of mild cognitive impairment or dementia), no active DSM-5 psychiatric disorder, and PET scan completion within prior 18 months. Participants were excluded from study participation if screening questionnaires indicated moderate to severe depression or anxiety, suicidal ideation within the past month, or history of suicide attempt. The University of Wisconsin-Madison Institutional Review Board approved the study protocol, and all participants provided informed consent to participate in the study.
Demographics
Demographics
Percentage of correct responses pre-test and post-test by question area. Different wedge colors refer to different question domains assessed. The length of the wedge corresponds to the percentage of correct responses on the knowledge quiz. Lighter-shaded wedges represent pre-test percentage of correct responses, and darker wedges represent post-test percentage of correct responses. Question types included: T/F (true/false); MC (multiple choice); Y/N (yes/no).
Study design
Study procedures included: pre-disclosure education (week 0); disclosure of amyloid PET results using a structured process (week 2); and a post-disclosure dementia risk-reduction counseling session (week 6), followed by telephone follow-up at 1 month and 6 months after the risk-reduction counseling visit. Here, we focus on the pre-disclosure education session (week 0). Detailed information on study design and feasibility are reported elsewhere [10].
During the education session, we assessed participants’ baseline knowledge of dementia, amyloid testing, the role of testing in research, and the potential psychosocial consequences of disclosure using a 15-question quiz with multiple choice and true/false questions (for questions see Table 2 and Supplemental Material). Next, trained study staff guided participants through a 15-min interactive presentation on causes of dementia, AD brain changes, risk factors for AD, meaning of amyloid PET results, and potential psychosocial outcomes (such as stigma, anxiety or other mood changes, or consequences for insurability) after learning results. Using a slide deck and script, staff reviewed key information, regularly eliciting feedback from participants and providing the opportunity to ask questions. Then the quiz was administered again to re-assess knowledge. If any questions were answered incorrectly, the teach-back method, an evidence-based approaches to education in healthcare settings, was used to address these remaining knowledge gaps [11]. Participants were required to demonstrate adequate understanding (as defined by a quiz score of 12/15) to continue in the study. The quiz and educational intervention were designed by experts in geriatrics, AD, and psychology.
Knowledge quiz items and number (percentage) of correct responses pre-education and post-education session
Descriptive statistics were used to evaluate sample characteristics. Response distributions for each quiz item and the total number and percent of items correct were used to measure pre- and post-education baseline knowledge. Because the quiz scores were not normally distributed, we assessed within-person differences on pre- and post-education total quiz scores using a Wilcoxon signed-rank test.
RESULTS
Sample characteristics
102 participants enrolled into the Amyloid Disclosure Study. The sample was on average 72.1 (SD = 4.9; range = 65–82) years old, predominantly female (67.7%), White (95.1%), had a family history of dementia (64.7%), and averaged 16.1 (SD = 2.4; range = 12–20) years of education.
Knowledge at baseline
The average quiz score of participants before the education session was 12 of 15 questions correct (SD = 1.7). Participants demonstrated high levels of knowledge about progression of brain changes in AD (n = 98, 96.1% participants answering correctly), relationship between lifestyle and cognition (n = 97, 95.1%), the role of amyloid as a marker of elevated brain risk and meaning of an ‘elevated’ (n = 102, 100%) or ‘non-elevated’ scan (n = 101, 99.0%), and that amyloid is a risk factor for cognitive decline (n = 95, 93.1%) but does not definitively predict AD dementia (n = 98, 96.1%). Participants also knew potential psychosocial consequences of learning amyloid results, such as psychological anxiety or distress, or discrimination by insurance companies (n = 95, 93.1%). Almost all participants knew that results were being disclosed as part of a research study (n = 99, 97.1%), and participants (n = 102, 100%) felt confident in their decision to learn results.
The area of least knowledge at baseline was around causes of dementia, with 24 (23.5%) answering this question correctly. Other areas of lower knowledge included how brain changes due to AD are measured by means other than amyloid (n = 42, 43.1% answered correctly), and early symptoms of AD dementia (n = 70, 68.6%). Fifty-seven participants (55.9%) were unsure how APOE, the most important genetic risk factor for the development of late-onset AD, was defined. While participants generally knew potential psychosocial consequences of learning their amyloid result, slightly less than half (n = 48, 47.1%) knew that the Genetic Information Non-Discrimination Act (GINA) does not protect against discrimination based on amyloid PET scan results.
Knowledge after the educational session
Total quiz scores improved from pre-education (median = 12 [IQR = 11–13]) to post-education (median = 15, [IQR = 14–15]) (W = 22, p < 0.001). The biggest gain was in correctly identifying the causes of dementia, which only 23.5% of participants answered correctly prior to education, compared to 68.6% after education. Other topics in which scores improved after education included how AD brain changes are measured (41.3% before and 70.6% after); the definition of APOE (55.9% before and 89.2% after); and the extent of existing protections against discrimination through GINA (47.1% before and 94.1% after). For one question about early symptoms of AD, knowledge decreased (68.6% to 59.8%).
DISCUSSION
As interest in AD biomarker disclosure expands, there is a growing need to provide individuals with knowledge about AD and biomarker testing prior to obtaining and learning their results. Providing education respects individuals’ autonomy by allowing them to make an informed decision [4, 7] and may allow individuals to better understand their results [6]. To date, one study has assessed information recall and emotional state after disclosure by asking cognitively unimpaired individuals in a research registry to imagine themselves receiving biomarker results, and then view video vignettes in which biomarker results were shared using different communication strategies [12]. To our knowledge, we present the first study to provide data about the level of knowledge among individuals undergoing testing before biomarker disclosure, along with the efficacy of a pre-disclosure educational intervention on increasing knowledge.
We assessed efficacy of an educational intervention on improving knowledge about AD and biomarker testing among older adult participants already involved in an observational AD study who received their amyloid PET results. Participants had high levels of baseline knowledge regarding AD biomarkers, pathophysiology, the impact of lifestyle on cognition, and of potential psychosocial consequences of disclosure. Participants had lower baseline knowledge around the causes of dementia, how AD is diagnosed with non-biomarker measures, genetic risk for AD, and existing legal protections against discrimination on the basis of biomarker test results. After our educational intervention, knowledge improved in several areas, suggesting that this was an efficacious approach to providing education about AD biomarkers to a sample already involved in AD research. The education session took about 15 min to administer and may thus be feasible for many settings. Our approach used the teach-back method, which is effective and widely used in many clinical contexts [11].
While participants were well-informed about amyloid biomarkers, they demonstrated less knowledge about more general information, such as causes of dementia or genetic risk factors. This likely reflects the population enrolled: participants had given informed consent to be in longitudinal research involving amyloid PET scans, had previously already undergone amyloid PET scans, and were familiar with amyloid biomarkers. Thus, they likely had high levels of knowledge about amyloid biomarkers even before they enrolled in the study. However, this research experience may not have translated to knowledge about other aspects of AD important to treatment, diagnosis, and psychosocial outcomes. Our intervention was efficacious in addressing several knowledge gaps, and a similar process addressing these content areas could be used for pre-disclosure education in other research settings. Given the potentially high familiarity with biomarkers for individuals already enrolled in research, it may be reasonable to reduce the education on this topic, while increasing education about issues like dementia more generally, psychosocial impacts, and legal/privacy concerns. We present our single center experience, and more work is needed in different research cohorts (e.g., racially and ethnically minoritized participants or socio-economic status) and settings (e.g., longitudinal cohorts versus trials). Research with several iterations of educational materials to identify an optimal balance of topics for a specific cohort may be necessary.
Of note, we observed a decline after the education session on scores for one question, which asked participants to identify early symptoms of AD. Before the educational intervention, most participants had correctly identified mood changes and cognitive decline as symptoms, but fewer participants did so after the intervention. The education session reviewed early symptoms in the context of preclinical AD pathophysiological changes that later manifested with symptoms. Some participants may have misunderstood “early” to mean “presymptomatic”, indicating a need for improved terminological clarity, such as reviewing the distinction between preclinical, asymptomatic AD, and symptomatic, early-stage AD [13].
For clinical populations, substantially different educational content may be necessary. Surveys of the US population have shown knowledge about dementia and AD, including of common symptoms and treatment, to be limited [14, 15]. Further, different populations may vary in terms of health literacy, views on aging, and access to social supports and resources after diagnosis [16, 17]. Clinical populations will likely need extensive education on dementia, biomarkers, and the implications of testing prior to disclosure [8], which may vary based on cognitive status [18]. The need for education clarifying the potential implications of biomarker testing will likely increase if multiple biomarkers are used as part of a comprehensive risk assessment that increases the possible number of outcomes (as well as the associated uncertainty) [19].
The purpose of education may also vary between clinical and research populations. In research disclosure, education is part of the informed consent process, intended to help participants understand the potential benefits and risks of harms of disclosure in general terms. Research is not primarily oriented toward direct personal benefit, and participants are often motivated by altruism or a desire to advance scientific knowledge [20, 21]. In clinical disclosure (which could include both cognitively unimpaired and impaired patients), education would be oriented towards making decisions that are best for the treatment of an individual patient. In situations where there is a clinical choice, the decision-making process may be more in line with what is termed shared decision making [22], a widely used, effective approach to making medical decisions when outcomes are uncertain or there is more than one reasonable option [23, 24]. In shared decision making, education goes beyond the review of risks and benefits that is typical in research, to review the implications of a test or treatment for the patient’s unique situation [25]. As a consequence, the educational approach we outline could be a starting point, but would need to be tailored to each patients’ clinical picture, and discussed in relationship to their personal goals and values. This may require a multi-disciplinary approach, involving medical providers, social workers, and nurses. Further research is necessary to understand the consequences of AD biomarker testing and disclosure in clinical populations, and the best means to improve education and shared-decision making about testing [8].
Limitations of this study include the focus on current research participants who had high levels of knowledge prior to the intervention and were highly motivated to learn their biomarker results. Our population was mostly non-Hispanic White and highly educated, limiting applicability to the US population of older adults. Approaches to education, including accessible language, are critically needed to ensure people with less health literacy, education, and familiarity with the English language are given ample opportunities to learn about and understand AD biomarker testing and results. All study staff were trained in the teach-back method, and the observed effect may have been partially due to the skill of the study team member in providing the education and time available to address all questions, rather than the content of the education itself.
We have presented a brief educational intervention that can increase knowledge about AD biomarkers, dementia, and the psychosocial aspects of disclosure in a research cohort of cognitively unimpaired participants. Further research is necessary to optimize educational interventions for different populations, and to determine the impact of education on disclosure outcomes.
Footnotes
ACKNOWLEDGMENTS
We extend our deepest thanks to the WRAP and WADRC participants and staff for their invaluable contributions to the study.
FUNDING
This publication was supported by funding from the National Institute on Aging (Supplement to NIA R01 AG021155 (PI: Johnson, Supplement Lead Investigator: Clark).
CONFLICT OF INTEREST
SCJ has in the past two years served on advisory boards to Roche Diagnostics, Prothena, AlzPath, Merck, and Eisai. His institution has received research funding from Cerveau Technologies. FK is an Editorial Board Member of this journal but was not involved in the peer-review process nor had access to any information regarding its peer-review. The rest of the authors have no relevant conflicts of interest to disclose.
DATA AVAILABILITY
The data supporting the findings of this study are available on reasonable request, and are not publicly available due to privacy restrictions.
