In this interview study, the general attitude of most mutation carriers of HD, SCA1, and SCA3 towards receiving onset information was positive. Following their presymptomatic genetic test result, most participants desired further information about their future disease. Importantly, participants were asked to consider this information within the current context where no disease-modifying treatment is available. This study found that mutation carriers of HD, SCA1, and SCA3 often estimate their disease onset by correlating it to the observed disease onset of family members and by relating the repeat length and onset of their (grand)parent to their repeat length. This suggests a need for clarity about personal disease onset and aligns with their general willingness to receive such information.

Participants provided a wide range of reasons for wanting to receive onset and progression information. The most important reasons mirror those given by at-risk individuals for undergoing presymptomatic genetic testing, which include reducing uncertainty, life planning, and informing relatives (Baig et al. 2016; Mandich et al. 2017; Tibben 2007). Our study results suggest that presymptomatic genetic testing alone does not fully fulfil some mutation carriers’ needs in this regard, and that obtaining more precise information about disease onset and progression could provide them with meaningful benefits. Additionally, preparing for the disease and reducing hypervigilance were also noted as important reasons for seeking onset and progression information. There were some differences between the two age groups in this regard. For example, participants in the younger age group (under 40) were more likely to focus on reproductive, educational, and career decisions, whereas older participants tend to focus more on providing information to their children, retirement planning, and end-of-life decisions.

Approximately one third of the participants stated they wanted to obtain all possible onset and progression information, while four participants would decline any possible information. The other half of the participants expressed a positive attitude towards receiving onset information in general, but many were also concerned with untoward psychological effects. If implemented, it seems important to thoroughly discuss the potential consequences of receiving onset information with mutation carriers. The findings of this study can assist counsellors in discussing important reasons for and against receiving this information if it becomes available.

In general, participants were more hesitant regarding obtaining progression information, with approximately half of them unwilling to receive this information at present. Some felt this information would not significantly add to their current knowledge, as they believed their own situation would mirror that of their affected family member(s). Additionally, some participants, mainly in the younger age group, believed such information only becomes relevant later in life. This suggests that when onset and progression predictions become available, it is important to discuss their implications separately as two distinct topics.

Preferences for the range of onset predictions varied between the two age groups. The older age group preferred a more specific onset range (one to two years) compared to the younger age group (two to five years). It appears that carriers who perceive the onset as closer to the present have a stronger desire for a more precise prediction. When developing a prognostic model, it is important to consider the possibility of differing preferences among age groups. Additionally, when such a model becomes available in clinical practice, information about age group preferences can be valuable for genetic counsellors.

Furthermore, participants mentioned a wide variation in percentages regarding a reliability threshold for such a test. Many found questions about test characteristics and terms like ‘reliability’ and ‘range’ difficult to grasp. The majority of participants believed that if a prediction is incorrect, the actual onset would still occur close to the predicted range, but this remains uncertain. If this belief is not correct, it could affect their preferences regarding a reliability threshold. A test with a reliability threshold significantly below 90% as suggested by many participants could complicate the counselling process and diminish the personal utility of these predictions. Understanding percentages and probabilities in healthcare-related situations is notoriously challenging for individuals (Apter et al. 2008). Since such predictions will inevitably come with some degree of uncertainty, extensive communication about the test’s reliability should occur after its implementation. Moreover, this raises the question of what an appropriate threshold for implementation should be, as well as how and by whom it should be determined.

In addition, we only discussed one potential type of test result regarding the onset (an age range). Other strategies are also possible, such as a test that predicts whether someone will show clinical symptoms within the next x years. One participant mentioned this type of communicating such a test result. These different methods of communicating test results may be used in the counselling to ensure that the communicated results are most in line with the counselee’s preferences and needs.

Some participants had a repeat length within the reduced penetrance range, which means they currently remain uncertain about developing disease symptoms in the future. Reduced penetrance carriers often stressed the uncertainty of their test results and did not attempt to estimate their own disease onset. While being uncertain about developing symptoms in the future could potentially increase their willingness to receive onset and progression information, we did not observe a difference in willingness between reduced and full penetrance carriers regarding receiving this information. However, the sample size of this qualitative study is too small to draw definitive conclusions.

A few SCA participants wondered about whether the disease had already started, which was not confirmed by their neurologist. This raises broader questions about how disease onset will be defined in a prognostic model. For instance, such a model may be able to detect changes that occur before the current moment of clinical onset as diagnosed by neurologists. This raises questions about whether the model will be trained to predict the current moment of clinical diagnosis, or whether it will predict a different moment of onset, e.g. in a prodromal stage. For instance, in HD research, Tabrizi et al. (2022) suggested distinguishing multiple disease stages. Some of these stages are based solely on increased biomarkers, in the absence of clinical symptoms. If such a system extends to clinical practice, what will the consequences be for mutation carriers? Would they consider themselves to be ill sooner? These are important questions that should be addressed during the development and after implementation of such a model.

In general, we observed a dynamic thought process among participants. Some participants explicitly indicated that the interview questions were challenging, while others, after contemplation, arrived at different viewpoints than their initial responses. Additionally, some participants found it difficult to decide whether they would opt for a prediction, as receiving it has both benefits and drawbacks, which indicates the importance of thorough counselling. It is possible that, despite expressing their willingness to receive onset and progression information during the interview, participants might ultimately make a different decision, similar to how individuals considering undergoing presymptomatic genetic testing in the 1980s and 1990s changed their minds (Craufurd et al. 1989; Tibben et al. 1992). However, this does not weaken the relevance of our study results, as our main aim was not to determine participants’ definitive decisions regarding onset and progression predictions but to map their possible preferences, views, and concerns. In addition, many participants had observed a parent or other relatives progress through multiple stages of the disease, which shaped their understanding of their own potential future. In the literature on presymptomatic genetic testing for HD, such experiences appeared to influence some reasons for and decisions about undergoing testing (e, g, Mandich et al. (2017). While firsthand exposure to the disease’s severity may have also shaped our participants’ responses, this experience is an inherent aspect of hereditary diseases.

Our study has several limitations. First, HD, SCA1, and SCA3 have some distinct characteristics and manifestations, such as the more pronounced cognitive impairment observed in HD. We aimed to address these differences by highlighting notable differences between the diseases where relevant. Second, to capture a broad range of perspectives, we sought to include a diverse group of participants for each disease and from both age groups. However, recruiting SCA1 and SCA3 carriers proved challenging, and as a result, the Dutch Ataxia patient organisation also distributed information about the study. This may have introduced selection bias. We did not interview men with SCA3, as only women with SCA3 were willing to participate, and we were only able to include two SCA1 carriers. The age ranges varied between the diseases, with a higher proportion of HD participants being over 40 compared to those with SCA1 and SCA3. Furthermore, individuals who strongly oppose receiving any prognostic information may have been hesitant to participate. In addition, we actively sought participants from diverse cultural backgrounds but were not successful. There are several possible explanations for this. For HD, the cultural diversity within the affected population might be limited (Rawlins et al. 2016). It is, however, also likely that individuals of these groups have either not undergone presymptomatic testing or are reluctant to participate in studies, possibly due to (cultural) taboos or language barriers. Third, during the interviews, participants were encouraged to bring their partners. In our opinion, this contributed to a deeper reflection on the answers provided by the participants. However, participants might have been hesitant to fully express their thoughts in the presence of their partner. For example, some might have been afraid of the reactions of their loved ones, though there were no indications of this in the interviews. Lastly, all participants received information regarding their presymptomatic genetic test result and its implications by their genetic counsellor. However, we did not explicitly assess participants’ prior knowledge and understanding of the disease or whether they sought additional information from other sources. This potential variation in prior knowledge might have influenced their responses, although we did not observe any critical knowledge gaps during the interviews.

Future research could proceed in several directions. Firstly, it should focus on understanding the preferences and concerns of groups with different cultural and religious backgrounds within the Netherlands regarding onset and progression information. Secondly, it is important to explore the preferences, views, and concerns of healthcare professionals as well. Some authors (MR, MS, AT, and LB) are in the process of conducting an interview study with healthcare professionals (genetic counsellors, neurologists, psychosocial workers, psychologists, and specialised nurses) to map their preferences, views, and concerns regarding more precise onset and progression predictions. Ideally, that will also stimulate discussion on best practices for counselling if such prognostic models become available. Thirdly, if such models do become available, the psychological impact of more precise predictions for onset and progression should be carefully monitored.