Our most important finding was that the vast majority (81%) of cognitively normal participants recruited from a population-based brain research registry wish to know their genetic risk for developing dementia when participating in genetic research. When presented with three hypothetical dementia risk scenarios corresponding to different APOE genotypes, participants were more interested in genetic risk disclosure and more likely to participate in medication trials, when the presented likelihood of developing dementia was higher. APOE-ε4 screening within an online research registry thus seems to be a well-received method to facilitate recruitment of individuals carrying APOE-ε4-genotypes for preclinical trials that include the disclosure of genetic risk to participants.
Our finding of a high interest in genetic risk disclosure is in line with previous studies in both individuals at risk for AD dementia [33, 34] and in cognitively normal elderly [35–37]. Most frequently endorsed reasons for interest were to contribute to scientific research, but also to inform themselves and their relatives about their genetic risk for dementia and make long-term arrangements, as similar to other studies [38, 39]. It is not surprising that we found a high willingness for research participation and genetic disclosure among research volunteers of a brain research registry, as these individuals may be more interested in personal health-related information. Nonetheless, our sample also examined the attitudes of participants without family members with dementia of which less is known, and showed similar levels of interest
In line with previous findings, we found that males are more interested in risk disclosure than females [34, 40]. We did not confirm the previously reported association with age [34, 41, 42]. In our study, interest in risk disclosure was also related to education, as higher educated individuals were more often interested and lower educated individuals more often undecided. Other studies also found an increased interest in higher-educated individuals [35], however others showed opposite association [36]. Variations in results may be due to differences in age or specific settings and populations. Nonetheless, these results give direction for tailored educational tools and dissemination about risk disclosure and research participation. Additionally, emphasising the importance of this matter, are the results of the sensitivity analysis revealing that participants abstaining to answer the hypothetical scenarios were lower educated, more often had subjective memory problems and less often had a family member with dementia. Tailored educational tools and dissemination could for example consist of videos with detailed explanation about genetic risk and possible impact, and use language that is appropriate at B1 proficiency to abate uncertainties about information provision and enable a well-informed decision.
Interest in genetic risk disclosure was high in all hypothetical scenarios corresponding the different APOE genotypes (73% − 79%). Previous studies showed that expressed interest in risk disclosure exceeds actual participation in disclosure. For instance, only 24% of adult children of AD patients with an interest in risk disclosure progressed to actual disclosure [43], which was comparable to results of genetic resting for Huntington disease in the 1980s [44–46]. Nonetheless, participants’ interest in risk disclosure was somewhat higher in response to scenarios in which their hypothetical risk was higher. We expected the lower risk scenario the be more favourable, because of the higher chance to rule out dementia. However, our findings indicate that higher genetic risk is especially relevant for individuals to know, despite the uncertainty and larger personal impact. Possible explanation of interest in risk disclosure in all scenarios could be due to selection bias. Research volunteers within the Dutch Brain Research Registry, specifically the ones participating within this study, would probably have high interest in genetic susceptibility testing independent of risk scenarios. An additional explanation could be that participants did not perceive the low-risk scenario (10%, corresponding to the ε2ε2 or ε2ε3 genotypes) as significantly or meaningfully lower than the presented general population risk (15%). It must be noted, that the concept of risk is generally difficult to comprehend and we did not verify how the hypothetical scenarios were understood. Nevertheless, Roberts et al. (2000) also found that the pros of risk disclosure outweighed the cons for many individuals, and that individuals might underestimate the limitations and risks of genetic testing and disclosure [47].
Previous studies found that disclosure does not lead to significant depression or anxiety symptoms in APOE-ε4 carriers in the short term, both in controlled research trials [24, 25, 48] and in direct-to-consumer testing [49]. Additionally, disclosing APOE genotype may have a positive impact in terms of health and lifestyle changes, even after being informed none of these changes were proven to prevent AD dementia [35, 50]. In the current study, we found that more than half of participants indicated that if they were to find out they had an increased dementia risk, they would adopt positive lifestyle changes (‘I would be mentally more active/exercise more/eat healthier.’). However, actual behavioural changes after genetic disclosure are expected to be different from behavioural intentions expressed in response to hypothetical scenarios. It is known that changing behaviour is difficult because of complex interplay between intrapersonal and external factors like motivation, behavioural capacity and self-efficacy [51]. Moreover, endorsed behavioural changes may be associated with our specific study sample. As mentioned above, individuals in the Dutch Brain Research Registry may be more engaged with their personal health. In line with this, a study of Christensen et al. (2015) showed that differences in behavioural changes after genetic disclosure were related to different recruitment strategies [52]. So, implications from this study and other genetic disclosure studies may not apply to populations that are less prone to proactively seek out genetic susceptibility testing. However, our result do underscore that a large number of individuals wish to know their genetic risk and want to take preventive actions, for example participate in clinical trials or change health- and lifestyle behaviour [53]. This could be due to the growing awareness of the relationship between healthy living and dementia risk reduction [54–56].
Younger age, having first-degree relatives with dementia, and the presence of subjective memory complaints were associated with a higher self-estimated dementia risk compared to the general population, but not with interest in risk disclosure. One in three participants believed their personal risk was higher than the general population risk, which could be explained by the large proportion of participants with first-degree relatives with dementia and subjective memory complaints. Slooter et al. (1998) estimated that 25% of the general population aged 55 years and older have a first-degree relative with dementia, compared to 44% within our study [57]. ‘Dementia runs in the family’ or ‘I have memory complaints’ were also the most endorsed reasons for higher self-estimated risk. However, self-estimated high dementia risk was not related to interest in risk disclosure. This indicates that motives for risk disclosure are not exclusively depended on one’s dementia risk perception, and that reasons can be very personal.
Commercially available direct-to-consumer genetic screening tests including APOE genotype, have become more widely available and the interest among the general population in genetic susceptibly testing increases. [49, 58, 59]. With the current increase in interest in commercially available APOE-genetic screening tests [43], like 23andMe, and associated requests to explain genetic results [55], and growing demand of personalized dementia risk reduction [53], the need for accurate education about genetic risk and disclosure impact arises. Historically, genetic disclosure was assigned to medical doctors providing education and information about impact. This is however neither efficient nor scalable for a research setting. Previous participant registries have successfully used (remote) APOE genotyping as screening to recruit participants [21, 22] and provided frameworks for scalable genetic counselling and (telephonic) disclosure within a research context [29, 60]. However, more research is needed to align APOE-genetic disclosure protocols within a research setting for cognitively normal adults without first degree relatives with dementia [61]. Currently, evidence about safe disclosure to cognitively normal research volunteers is emerging [23, 61]. Our study provides additional evidence of the generally positive attitudes of cognitively normal adults towards genetic screening for research purposes, and underscores the high interest in disclosure.
Strengths and limitations
This study had several limitations; the present sample was selected from the Dutch Brain Research Registry [31], in which participants have registered because they are potentially interested to participate in brain research and genetic disclosure which limits the generalisability of our findings to the general population. However, our large sample provided the opportunity to study the effect of multiple participant characteristics (age, gender, education level, having first-degree relatives with dementia) on the interest in risk disclosure. Secondly, participant selection from a research registry provided us with a representative of individuals who want to participate in scientific research or prevention trials. Unfortunately, we were not able include information about social-economic status or racial background. We acknowledges the importance to diversify the samples enrolling in genetic susceptibility testing research, to generalize the findings to the general population. Another possible limitation is the online nature of the questionnaire, as we could not control whether participants interpreted the questions correctly, and hypothetical scenarios were understood. Additionally, the present study investigated interest in risk disclosure using hypothetical research participation and personal risk scenarios. It is therefore important that future studies examine actual participation in genetic risk disclosure and its long-term impact in a real-life setting.