Validation of the “Perceptions Regarding pRE-Symptomatic Alzheimer’s Disease Screening” (PRE-ADS) Questionnaire in the German Population: Attitudes, Motivations, and Barriers to Pre-Symptomatic Dementia Screening

INTRODUCTION

Dementia, the seventh leading cause of mortality, affects over 55 million people worldwide [1] and is one of the most feared diseases [2]. Nearly 10 million new cases are recorded annually, although it is estimated that 75% of people with dementia (PwD) are undiagnosed at death, and up to 90% in some low- and middle-income countries [1]. As research has shown, routine screening could lead to a higher diagnosis rate, and early treatment may at least partially delay cognitive decline [3, 4], advances in dementia diagnosis have led to an increasing number of studies on the acceptability of early dementia screening [5–12] in older adults, but there is a paucity of data examining community attitudes toward pre-symptomatic screening.

The discovery of biomarkers associated with the pathophysiology of Alzheimer’s disease (AD) [13] has led to a paradigm shift from a clinical diagnosis based on functioning to a biological diagnosis based on biomarkers and PET [14] and to a new understanding of dementia as a continuum [15, 16]. The pathophysiological process begins decades before the first symptoms become clinically apparent [17], in a preclinical, asymptomatic stage, in which there is biomarker evidence of AD pathology, followed by a prodromal stage of mild cognitive decline with neuropathological changes typical of AD [15]. Therefore, biomarkers and genetic susceptibility screening are gaining increasing attention in health care, research, and direct-to-consumer (DTC) settings [18–20].

The utility of biomarkers such as noninvasive neuroimaging, cerebrospinal fluid (CSF), containing Aβ_1-42, Aβ_1-40, tau, phospho-tau, and neurofilament light chain (NfL), and genetic risk factors of AD is controversial because of limited prognostic value [21, 22], and because pre-symptomatic screening is without therapeutic consequences as long as safe treatment options are not available. In addition, it may also harm asymptomatic individuals through ineffective overdiagnosis and overtreatment [23]. The clinical validity of both CSF biomarkers and “amyloid PET positivity” is intensely investigated. The problem is that not all individuals with positive biomarkers or amyloid PET will become symptomatic [24]. Although biomarkers for AD may have an undeniable value for research, the benefit of pre-symptomatic screening appears limited [25].

Since the discovery of the Aβ-encoding APP gene in 1987 [26], developments in AD genetics have led to the identification of four genes involved in AD [27]. Three are known deterministic genes in which mutations are associated with early-onset autosomal dominant AD (EOAD) before the age of 60–65 years, accounting for 1–5% of all AD cases [28] and approximately 80% of EOAD cases [29]: the amyloid precursor protein (APP) on chromosome 21 [30], presenelin 1 (PSEN1) on chromosome 14 [31], and presenelin 2 (PSEN2) on chromosome 1. Recently, SORL1 has been identified as the fourth familial, sometimes even autosomal-dominant, AD gene. It is estimated that potentially damaging SORL1 variants affect as many as 2.75% of all unrelated people with early-onset AD, and 1.5 percent of those with late-onset AD (LOAD) [32].

For LOAD, genome-wide association studies (GWAS) have shown that many genetic loci contribute to AD risk [33], while apolipoprotein E (APOE), with its isoform ɛ4 (APOE4) on chromosome 19 appears to be the major susceptibility gene [28]. While the ɛ2 allele plays a protective role [34], APOE4 increases the risk in a dose-dependent manner: A single ɛ4 allele increases the risk of AD approximately threefold, while homozygotes have a 15–30-fold increased risk [35]. However, these susceptibility genes are only associated with an increased risk for the disease development, while many other (environmental) factors play a role in whether the disease is manifested, in contrast to familial AD or Huntington’s disease, where genetic tests have a predictive value of clinical relevance [36].

While biomarkers and genetic testing may be useful in clarifying a clinical diagnosis, the question of their utility in asymptomatic individuals arises in the absence of effective therapy and despite the paucity of evidence for biomarkers and amyloid scans [23, 37], as well as despite the lack of sensitivity and specificity of the biomarkers and their unclear predictive ability [14]. A biomarker must have a sensitivity of at least 90%, a specificity approaching 100%, a high degree of reliability, and the ability to detect one or more fundamental features of the relevant neuropathologic changes. These requirements are not met by imaging markers or by APOE4 [38, 39]. Biomarker data are limited in asymptomatic individuals but suggest that abnormal biomarkers are common and not indicative of dementia in the majority of individuals [40] and add only little value in predicting whether a person will develop dementia in the next few years [41]. Because of the uncertainty of biomarker results [42, 43], it is questionable whether a condition with abnormal biomarkers should be considered a disease or an at-risk condition in the absence of clinical symptoms [14]. Therefore, even testing for APOE4 is not recommended in asymptomatic individuals [28, 44]. In addition, testing asymptomatic individuals for dementia raises ethical questions about the appropriateness of using biomarkers or genetic tests and about the personal benefits of pre-symptomatic testing. The basic principles of bioethics—autonomy, beneficence, non-maleficence, and justice [45]—can help to identify moral values in the debate about the ethical desirability of pre-symptomatic screening for AD. Respect for autonomy emphasizes the importance of a person’s knowing or not knowing what is going on in their brain and their risk status for AD [25, 21]. This requires that the person understands what to expect and that the person understands the uncertainty of the results. The benefits of pre-symptomatic screening must always outweigh the risk of harms related to discrimination, overdiagnosis, and psychological harm, as well as the burden and risk of testing itself [25]. Other harms include irreversible decisions, changes in self-perception, family disruption, especially when hereditary risk factors may be involved [24], internalized stigma (how individuals feel about themselves) and public stigma (how others judge them) [25], as well as political questions about voting, driving, and economic questions about balancing an employer’s right to a productive workforce with an employee’s right to avoid discrimination on the basis of dementia risk [26], leading to a dilemma between privacy, protection, and family responsibilities[46].

Benefits are often based on misconceptions and unrealistic expectations [14], such as better planning for the future, improved health and well-being, adoption of a healthier lifestyle, as well as a desire to better understand their brain health status [47]. According to Erdmann et al. (2018), the benefits outweigh the harms only among people with symptoms who seek support [48]. Although the benefit of knowing the risk of AD is currently questionable, it may be of great benefit in the near future once therapy is available [4, 49]. Finally, justice concerns the limited availability of biomarker and genetic testing but also the risk of discrimination based on an AD diagnosis [25].

The motives for pre-symptomatic screening are not always clear. They may be intrinsically motivated, or they may be driven by fear or hope of preventing AD. This hope makes them vulnerable to misconceptions and unrealistic expectations [14]. Arguments for testing include people’s right to know (or not know) their risk, to plan and prepare for the future, both financially and emotionally, to spend more time with family, to purchase long-term care insurance, and to support basic research [18, 50–52]. As long as the prognostic value is unclear and the risk prediction inaccurate, this argument has only a limited validity [24]. Labeling large numbers of asymptomatic people with pre-symptomatic dementia or at risk without being sure they will develop symptoms seems to have little or no benefit and creates a group of “patients in waiting” [14]. There is also the risk of medicalization [53], as unnecessary tests and treatments can also have side effects [23, 54], adverse psychological and social outcomes [14], and require financial resources that are needed elsewhere. The often-heard argument that it is possible to change one’s lifestyle with knowledge of a risk is not really convincing as it is possible—and even desirable in terms of a healthier lifestyle—to change one’s lifestyle without knowledge of a risk [55]. But once a person is labeled with a pre-disease, he or she will remain so for life [56]. In the case of genetic testing, this also has implications for the family since genes, strictly speaking, affect not only the person tested but the entire family [57].

With the ever-increasing number of biomarkers and the ability to test outside of clinical trials using the DTC genetic testing, the question arises regarding the attitudes, motivations, and barriers to pre-symptomatic testing in the general population. Previous studies have focused on the attitudes, knowledge needs, benefits and barriers to dementia screening of a specific population, mostly relatives of PwD [52], using self-administered survey questionnaires or structured interviews [18, 58–61]. Validated scales include the “Perceptions Regarding Investigational Screening for Memory in Primary Care” (PRISM-PC) [7], which assesses attitudes toward dementia screening in primary care, and the “Dementia Screening and Perceived Harms” (SAPH) [11, 62], which assesses the acceptability of dementia screening in primary care. As these scales were developed to investigate the acceptability of early dementia screening, but not necessarily for pre-symptomatic screening, and lack questions about the need for knowledge or the type of support that people would like to receive before, during, and after the disclosure of the test results, Makri et al. (2023) [63] developed the “Perceptions regarding pRE-symptomatic Alzheimer’s Disease Screening” (PRE-ADS) questionnaire [63], which was based on the PRISM-PC [7], a literature review and a panel of clinical experts from the Greek Association for Alzheimer’s Disease and Related Disorders (GAADRD). The scale would be helpful for further research into ethical implications, knowledge of the sensitivity and validity of biomarker and genetic tests, their benefits and limitations, and exploration of cultural differences.

The aim of this cross-sectional study was a) to examine the psychometric properties of the German version of the PRE-ADS in the general population and b) to examine the psychometric properties of the subscale, “Acceptability of Screening”. In addition, we aimed to further investigate potential influences on participants’ attitudes, motivations, barriers, and acceptability of screening based on their demographic background or personal experience with dementia.

MATERIALS AND METHODS

RESULTS

DISCUSSION

The aim of this cross-sectional study was to analyze the psychometric properties of the German version of the PRE-ADS in the general population. The questionnaire was practical for online administration in less than 10 min. Both the 25-item scale and its three factors, “Concerns about Screening”, “Intention to be Screened”, and “Preventive Health Behaviors”, showed acceptable to good internal consistency. The PRE-ADS-D discriminates clearly between younger and older adults, but not between participants with and without affected family members or between those who have received dementia training and those who have not. The 5-item subscale “Acceptability of Screening” revealed that 51.2% of all participants expressed a positive attitude towards their willingness to undergo a pre-symptomatic or diagnostic AD screening. Internal consistency for this subscale was excellent. The subscale did not discriminate between younger and older people but showed a tendency to a higher acceptance of those over 65 years and discriminated between those with and those without a family history of dementia as well as between those who had attended a dementia course and those who had not. In addition, a significant association was observed between the fear of developing dementia and the presence of a family member with dementia.

The internal consistency is comparable with earlier reports of the original Greek version of the PRE-ADS [63] and to scales that focus on dementia screening, such as the PRISM-PC or modified scales [7, 10–12, 78, 79].

We found the strongest evidence for a three-factor structure for the PRE-ADS-D, in contrast to the study by Makri et al. (2023), which reported the four factors “Perceived Harms of Testing”, “Acceptance of Testing”, “Perceived Benefits of Testing”, and “Need for Knowledge” [63]. However, when we looked closely at the items that loaded on each factor, we found that 21 out of the 25 items loaded on their intended factor, with the remaining four items forming the fourth factor in the original study, “Need for Knowledge” (items 6, 7, 8, 14). It is noticeable that these items either show lower factor loadings (0.38–0.57) or cross-loadings on other factors. Similar studies that are based on the PRISM-PC by Boustani et al. (2008) [7] have also shown slightly different results in terms of the number of factors compared to the original proposal by its creators [11, 12]. Differences in factor loadings and model structure may be attributed to variations in sample characteristics [12]. In contrast to the predominantly student-based sample in Makri et al. (2023) [63], the sample in our study represents a more diverse range of the general German population: Participants in this study ranged in age from 18 to 99 years and had a variety of sociodemographic characteristics, including different occupations, education levels, marital status, family history of dementia, and different health and mental health status. In addition, attitudes, motivation, and barriers toward pre-symptomatic AD screening can differ across different countries due to culture-specific factors and variations in healthcare systems, as observed in dementia screenings in general [12, 80].

The first factor “Concerns about Screening” consists of 10 reverse-coded items that express participants’ concerns if they knew they were at high risk for developing dementia (items 9, 10, 11, 12, 15, 16, 18, 19, 20, 21) and corresponds exactly to the factor “Perceived Harms of Testing” of Makri et al. (2023) [63]. These concerns can be categorized into individual psychosocial effects, such as emotional suffering (anxiety and depression) or stigma (items 11, 12, 18, 19, 20, 21), and family burden (emotional and financial) (items 9, 10, 15, 16). Examples include item 11 (“I feel that I would be overwhelmed by mental pain”) and item 15 (“My family would suffer financially”). These findings align with the results of the other studies that utilized the PRISM-PS scale where similar items were included in the “Perceived Harms and Benefits of Testing” scale, and similar factors were identified as harms of testing, such as “stigma”, “(emotional) suffering”, “family burden”, and “negative impact of screening on independence” [7, 12, 80]. It is well known that pre-symptomatic testing for any type of disease can indeed carry several risks, including discrimination, over-diagnosis, internalized stigma (individuals’ self-perception), public stigma (how others perceive them), as well as political and economic considerations [27–29]. Since the benefits of genetic testing in asymptomatic individuals should always outweigh the potential concerns and harms, it is critical to identify the potential harms of AD screening and assess their perceived importance in order to incorporate them into future counseling programs [25, 80].

The second factor, “Intention to be Screened”, incorporates all questions that assess the participants’ preference to find out if they have a higher risk of dementia (item 1) or the willingness to be regularly tested for the presence or the risk of dementia by different screening methods such as regular completion of a questionnaire (item 2), blood sampling (item 3), imaging procedures (item 4), or biomarker screening (item 5). It also includes item 6 (“In order to decide to be tested for the presence of AD, I would need more information and details”) (Table 2). Here, the results differ from those of Makri et al. (2023) [63], where items 1 to 5 load on the second factor, but item 6 loads on factor 4 (“Need for Knowledge”). In the present study, the inclusion of item 6 in the “Intention to be Screened” factor highlights the relationship between an individual’s level of information about pre-symptomatic and diagnostic dementia screening and their attitudes toward screening. Prior knowledge plays an important role in shaping an individual’s intention to undergo screening. This finding is consistent with the basic concepts and primary goals of genetic counseling, in which the provision of understandable, non-directive information is critical for the counselee to have sufficient knowledge in order to facilitate informed consent and to make an autonomous decision to undergo testing [81].

The remaining 9 items loaded on the third factor, “Preventive Health Behaviors” (Table 2). Early detection of one’s predisposition may lead to the adoption of a healthier lifestyle (items 13 and 25) and may also allow individuals to have more time for overall processing, discussion, and planning for the future with family members. This includes discussing health issues, long-term care services, and financial planning (items 17, 22, 23, and 24).

In contrast to Makri et al.’s (2023) [63] third factor, “Perceived Benefits of Testing”, our study’s third factor, also included items 7, 8, and 14. Participants indicated that if they theoretically received a positive test result, they would seek counseling to gather advice in general (item 7), cope better psychologically or discuss their feelings and thoughts (item 8), and learn about prevention and treatment options (item 14). Interestingly, these remaining 3 items were added to the PRISM-PC by Makri et al. (2023) [63] and, together with item 6, contributed to the formation of the fourth factor in their validation study, called “Need for Knowledge”. In the German study, these items seem to focus on a different aspect, namely that information is seen as a way to achieve the goal of adopting preventive health behaviors. Through counseling, participants would ultimately benefit from disclosure of the results. Therefore, the PRE-ADS [63] is the first scale to address the knowledge and support needs of individuals before, during, and after pre-symptomatic screening. Gooblar et al. (2015) [79] only included the need for counseling during the process of pre-symptomatic testing in their scale, and Christensen et al. (2011) [82] have added a question about the individuals’ motivation to stay informed about new developments in the treatment and prevention of AD.

One explanation for the fact that the questionnaire results differ between older and younger people, but not between those with a positive or negative family history, is probably that the questionnaire combines too many topics, and it is questionable whether the participants could make an autonomous decision that would include being sufficiently informed. Thus, it would probably be advantageous to ask the questions about benefits and harms first so that the subject can do a cost-benefit analysis. Similarly, questions about possible consequences and lack of information should be asked before questions about acceptability. In addition, there is the question of how much the subjects know about the influence of genetics, how EOAD differs from LOAD, and what conclusions can be drawn from biomarkers or imaging techniques regarding the onset of the first symptoms. This includes education about possible prevention and therapies, which are not yet available for dementia, which is why at least pre-symptomatic genetic testing for AD is considered ethically questionable by professional societies [83]. In the case of genetic testing, there is the additional ethical problem that genetic data are sensitive data, which are also family data [84]. This also explains the finding that there was a significant association between the fear of developing dementia and having a family member with dementia.

Since any medical intervention requires informed consent, in which all questions from the patient must be answered by the physician, a valid questionnaire that captures the acceptability of pre-symptomatic screening would be highly relevant in practice. Braun et al. (2014) [12] used subscale B of the PRISM-PC for this purpose and developed a mini-questionnaire from the original 8 questions, which finally contained 6 questions, 5 of which related to an AD diagnosis, while one question related to a “higher risk of developing AD”. The subscale has excellent internal consistency and has been shown in a study of participants in the Berlin Aging Study II (BASE-II) to be acceptable to 72.2% of older adults for routine screening for AD. Following this attempt, we also created a subscale “Acceptability of Screening” consisting of the first 5 items of the PRE-ADS, which belong to the second factor. Consistent with the study by Braun et al. (2014) [12], this scale showed excellent internal consistency. While the focus of the questions in the Braun et al. (2014) [12] scale was on the diagnosis of dementia, the items of the PRE-ADS mainly ask about the risk of developing AD: The first item asks whether someone wants to know their general risk of developing AD, item 3 asks about genetic testing, while items 4 and 5 ask about imaging and biomarkers to diagnose AD (Table 2). As the PRE-ADS is a questionnaire about pre-symptomatic AD and with the knowledge that the pathology of AD can be detected with biomarkers 30 years before the onset of symptoms [85], it would be desirable to adapt the items and to make the difference between pre-symptomatic testing and screening for diagnosis more precise. Item 2 (I would like to be tested regularly with a short questionnaire) is a clear question about the desire to be diagnosed with AD.

Although several studies have examined the acceptability, perceived harms, and perceived benefits of dementia screening [8, 11, 12, 80, 86–88], there remains a dearth of research on attitudes toward pre-symptomatic screening. As the demand for pre-symptomatic screening continues to grow worldwide, the PRE-ADS-D questionnaire could play an important role in future research and help clinicians, policymakers, and other health professionals to better understand the needs and attitudes of the German general population regarding pre-symptomatic testing and AD screening [63]. EFA structured these “attitudes” in terms of their “Concerns about Screening” (factor 1), their “Intention to be Screened” (factor 2), and the chance to benefit from pre-symptomatic testing by adapting “Preventive Health Behaviors” (factor 3). The correlations between the factors were low and insignificant, except for the relatively strong positive correlation between factor 2 “Intention to be Screened” and factor 3 “Preventive Health Behaviors” (τ(254) = 0.331, p < 0.001). This relationship is underscored by many similar studies, as many of the described benefits are the most compelling reasons for seeking pre-symptomatic screening [44, 78, 89].

However, it is very essential to interpret the results of this validation study with caution, considering that only the autosomal dominant EOAD, which accounts for 1–5% of all AD cases, is predictive of approximately 80% of EOAD cases [29, 90]. Early detection of this predisposition can be useful for an individual to prepare for the disease in terms of professional and personal life planning, always respecting the need for genetic counseling and the national legal and ethical frameworks for the predictive diagnosis of genetic diseases [91, 92]. Unlike EOAD, LOAD is determined by a complex interplay between genetic factors (polygenic risk from multiple susceptibility genes), environmental, and lifestyle factors [93, 94]. While APOE ɛ4 is the strongest genetic risk factor for AD, its effect accounts for only 27.3% of the estimated heritability [95]. According to Livingston and colleagues [94, 96], modifiable factors such as hearing loss, physical inactivity, hypertension, and obesity play a significant role in LOAD and have a greater impact than genetics.

The benefits of pre-symptomatic genetic screening should clearly outweigh any potential harms or risks associated with the testing process or the test results in order for it to be considered legitimate [25]. One could argue that early detection of one’s genetic predisposition to developing AD through pre-symptomatic screening could lead to preventive behavior that limits the risk of dementia, as discussed in the recommended strategies by Livingston and colleagues [94, 96], a behavior change that is consistent with our third factor, “Preventive Health Behaviors”, and that was also observed in the REVEAL Study [44]. However, preventive strategies, such as engaging in regular physical activity, are health recommendations that should generally be followed as they promote overall well-being and reduce the risk of various health conditions, including cardiovascular disease [97, 98], and are not specifically targeted at preventing dementia. In addition, testing for susceptibility genes associated with LOAD can only place individuals in a “risk” category but not resolve their uncertainty [29, 35, 90]. The fact that individuals with positive biomarker screening may not develop AD, while those without a genetic predisposition may still develop AD, underscores the complexity and variability in the relationship between biomarkers, genetic factors, and the development of LOAD [24, 94]. A positive biomarker test can be emotionally distressing for the individual and their family, causing unnecessary anxiety, fear, depression, and stigma [53].

Nevertheless, it is critical to emphasize that biomarker screening for AD has undeniable value for research purposes. A larger database increases the potential to identify additional genes that may play a significant role in predisposing to AD and provide more accurate prognostic information [99]. However, given the current predictive value, the risk of psychological impact, ethical implications and the lack of treatment options, it is not currently advisable to undergo predictive testing for LOAD [21, 22, 100]. The American Geriatrics Society’s (AGS) Ethics Committee, therefore, currently recommends genetic testing for LOAD only for diagnostic purposes [83]. Similarly, The German Association for Psychiatry, Psychotherapy, Psychosomatics, and Neurology (DGPPN) and the German Society of Neurology (DGN) advise against predictive testing for the APOE4 allele and suggest that predictive genetic testing accompanied by genetic counseling should be limited to family members of individuals with monogenic forms of dementia [91].

In summary, there are many ethical implications associated with pre-symptomatic genetic screening for AD. As the demand for predictive testing increases, and such tests are more and more accessible to a broader spectrum of society [101], it is necessary to develop informative and consulting programs in order to minimize possible harms. A critical step is to explore society’s needs and attitudes toward pre-symptomatic AD screening [63, 80]. Therefore, the PRE-ADS-D questionnaire could be an important tool to explore significant differences in individuals with different socio-demographic characteristics regarding their acceptability, perceived harms, and benefits of pre-symptomatic dementia screening and contribute to the results of other similar studies [9, 78, 80, 87, 102]. These findings can also help to design and tailor health counseling services to meet the individual needs of counselors while respecting cross-cultural differences [58, 63, 80].