Factors associated with parental knowledge of neurofibromatosis type 1 (NF1): Parental affected status and genetic counseling
Abstract
Neurofibromatosis type 1 (NF1) is a genetic condition characterized by various cutaneous, neurological and psychological manifestations. The present study examined whether parental knowledge of NF1 is associated with a parent's NF1 status, affected or unaffected, and exposure to genetic counseling. Parents of children with NF1 were invited to complete an online survey answering true or false and multiple-choice questions to evaluate their overall knowledge of NF1. The study included 274 respondents, of which NF1 knowledge scores were significantly higher for unaffected parents (p < .001), and for parents who reported previously meeting with a genetic counselor (p < .001). Items pertaining to NF1-related cancer were least likely to be answered correctly. The results of the current study revealed lower overall NF1 knowledge in affected parents and knowledge gaps identifying areas where focused NF1 education may be beneficial.
1 INTRODUCTION
Neurofibromatosis type 1 (NF1) is an autosomal dominant disorder with an incidence of 1 in 3,500 (Hersh, 2008). NF1 is caused by a pathogenic variant in the NF1 tumor suppressor gene. Common features include café-au-lait spots, inguinal and axillary freckling, Lisch nodules, neurofibromas and scoliosis. Additionally, approximately 40 to 60 percent of NF1 individuals have a learning disability, such as attention-deficit/hyperactivity disorder and difficulties with reading and/or spelling (Hersh, 2008). Less frequently, more severe manifestations such as pseudarthrosis, optic or other low-grade gliomas, and malignant transformation of tumors occur. A de novo pathogenic NF1 variant provides an explanation for approximately half of individuals with NF1, whereas the other half inherited a pathogenic NF1 variant from an affected parent. NF1 is a completely penetrant condition with extreme inter- and intra-familial variability in expression. Relatively few NF1 genotype-phenotype correlated cases are documented.
Previous studies have demonstrated that overall NF1 knowledge is deficient among affected individuals and parents of children with NF1 (Benjamin et al., 1993; Crawford et al., 2016; Oates, Payne, Foster, Clarke & North, 2013; Ponder et al., 1998). Carrieri, Farrimond, Kelly, and Turnpenny (2016) suggested that limited NF1 knowledge may be due in part to the extreme variability of the condition and to affected parents often basing their own clinical concerns, as well as concerns for their affected children, on their family members’ experiences (Barke, Coad & Harcourt, 2016; Benjamin et al., 1993; Carrieri et al., 2016). Additionally, Benjamin et al. (1993) reported that NF1 knowledge of affected participants is often limited to their personal and familial NF1 experiences.
Genetic counselors play an integral role in the education process by explaining clinical features, genetic etiology, recurrence risk, and medical management recommendations associated with NF1 (Huson, 1989). Various studies have shown that receiving genetic counseling is associated with higher levels of NF1 knowledge among affected and unaffected individuals (Benjamin et al., 1993; Rosnau et al., 2017). Kolesnik et al. (2017) found that unless parents are familiar with the condition or express concern, most infants with NF1 do not receive regular developmental examinations. Parents of children with NF1 also report value in understanding the condition in order to support their child's care (Barke et al., 2016; Gallo, Angst, Knafl, Hadley & Smith, 2005). Previous research supports the importance of increased education of parents of NF1 children and highlights genetic counselors’ role in the process.
The current study hypothesized that NF1-related knowledge is more limited in affected parents compared with unaffected parents. Previous studies have found that one's own experiences, such as a personal diagnosis of NF1, may impact their impressions and assumptions about the disease (Carrieri et al., 2016). Also, the higher prevalence of NF1-associated learning disabilities may contribute to affected parents having lower NF1 knowledge scores compared with unaffected parents (Hersh, 2008). Secondly, the study hypothesized that parents who received genetic counseling will have higher levels of NF1 knowledge than parents who did not receive genetic counseling. In the present study, parents of children with a diagnosis of NF1 were invited to answer a multi-part NF1 knowledge questionnaire, where scores were compared based on parental NF1 status and exposure to genetic counseling. To our knowledge, there is no previous research comparing affected versus unaffected parental NF1 knowledge in a quantitative manner.
The results of this study will help genetic counselors and other NF1 providers recognize important factors that may impact NF1 comprehension with a potential for improved pediatric healthcare compliance. Affected parents who are more compliant in their child's NF1 healthcare may also become better advocates of their own NF1 management. Lastly, the findings of this study may highlight the importance of genetic counseling in NF1 care and increase the number of families referred for genetic counseling.
2 MATERIALS AND METHODS
2.1 Study population
Parents were invited to participate in the study through the Children's Tumor Foundation (CTF) NF1 registry if they met the following criteria by data query: have at least one child under the age of 18 who has been genetically or clinically diagnosed with NF1, are aware of their own NF1 status (affected or unaffected), and know the etiology of their child's NF1 pathogenic variant (inherited or de novo). This study was approved by the Long Island University Institutional Review Board (IRB No: 18/09-138.A1).
2.2 Recruitment
A letter of intent was created to gain access to the NF1 registry and approved by the CTF research coordinator and review team, as well as the Long Island University IRB. Following approval of the study, the NF1 registry was utilized to send an initial invitation by email to eligible registry members. A link was provided with a consent page containing a description of the study and a web-based survey through Survey Monkey. A reminder email was sent three weeks later. If participants had questions concerning the purpose of the study or other inquiries, they were given the contact information for the study's principal investigator (PI) and the CTF.
2.3 Instrumentation
Survey questions were developed to analyze different aspects of NF1 status and care in both children with NF1 and, when applicable, parents who also have NF1.
2.3.1 Study groups
Parent participants were asked, ‘Do you have at least one child with NF1?’ and were only included in the study if they answered yes. A single multiple-choice question, ‘Do you have NF1?’ was used to determine whether the parent respondent also had NF1. If they answered yes, participants were included in the ‘self-affected’ group. Participants were also asked whether their child's other parent had NF1. If they answered yes, they were included in the ‘other parent affected’ group. If participants answered no to both questions, they were included in the ‘unaffected’ group. Parents included in the affected and other parent affected subgroups were assumed to have a child who inherited a pathogenic variant in the NF1 gene. Parents included in the unaffected subgroup were assumed to have a child with a de novo variant, excluding the possibility of germline mosaicism.
2.3.2 Evaluation of NF1 knowledge
Neurofibromatosis type 1 knowledge items were adapted from questions used in the Benjamin et al. (1993), Rosnau et al. (2017), and Erblich et al. (2005) studies. Additional questions were formulated and reviewed by genetic counselors who specialize in NF1 diagnosis and management. The knowledge items consisted of 18 true or false questions, 11 multiple-choice questions, and one ‘select all that apply’ item that addressed general topics of NF1, including etiology, possible manifestations, variability, diagnosis,and recommended follow-up surveillance. Including the ‘select all that apply’ question where one point was given for each correct selection, the total possible NF1 knowledge score could range from zero to 38. The internal reliability of the NF1 knowledge items was high (Cronbach's alpha = 0.701). NF1 knowledge scores were indicated as percentage of correct responses (0%–100%).
2.3.3 Exposure to genetic counseling
A single multiple-choice question, ‘Have you met with a Genetic Counselor?’ was used to establish whether participants received genetic counseling concerning NF1 from a genetic counselor.
Lastly, there was a free response question at the end for participants to provide any comments regarding the study. Three parents of children with NF1 were recruited from CTF as representatives to provide feedback during survey development in order to assure clear and concise wording of questions. The complete survey can be found in the Data S1.
2.4 Statistical analysis
Three groups were defined to describe the parental NF1 status: self-affected, other parent affected, and unaffected. The other parent affected group was used in the study as it gives a unique perspective from a subgroup of participants that do not themselves have NF1 and may have alternative exposures to NF1 knowledge than parents of the other two groups. ANOVA was performed to determine whether NF1 knowledge scores differed as a function of parental group (self-affected, other parent affected, and unaffected). t Test was performed to determine whether NF1 knowledge scores differed as a function of participant report of having received genetic counseling or not. The data were analyzed using the R Programming Language.
3 RESULTS
3.1 Demographic analysis
A total of 396 respondents initiated the survey. Out of the 396 respondents, 303 completed the knowledge portion of the survey. From those 303 respondents, 274 met eligibility criteria and were included in the analyses. Of the 274 respondents, 52 parents (19%) stated they had a personal diagnosis of NF1, while 222 parents (81%) answered that they did not. Of the 222 parents who reported to be unaffected, 29 respondents selected that the child's other biological parent has NF1. In the open-ended response section of the survey, seven of the participants unaffected by NF1 stated that they had adopted a child with NF1. There were no responses where both biological parents had NF1.
Based on this information, three groups were defined to describe the parental NF1 status: self-affected (52 respondents), other parent affected (29 respondents), and unaffected (193 respondents). Most participants were Caucasian (86.5%) and female (84.7%) with an average age of about 42 years. Just over half of participants reported full-time employment (52.6%), and 77.3% indicated they held an associate degree or higher. Complete demographic information can be found in Table 1.
| Age | |
| Average | 42.1 years |
| Range | 25–82 years |
| Race | |
| White/Caucasian | 237 (86.5%) |
| Hispanic/Latino | 13 (4.7%) |
| Black/African American | 8 (2.9%) |
| Asian/Asian American | 6 (2.2%) |
| Native American | 3 (1.1%) |
| Other/Unknown | 7 (2.6%) |
| Gender | |
| Female | 232 (84.7%) |
| Male | 40 (14.6%) |
| Other | 2 (0.07%) |
| Employment | |
| Full-time employment | 67 (24.5%) |
| Part-time employment | 9 (3.3%) |
| Unemployed | 51 (18.6%) |
| On disability | 9 (3.3%) |
| Unknown | 3 (1.1%) |
| Education | |
| No high school diploma | 4 (1.5%) |
| High school diploma | 58 (21.2%) |
| Associate degree | 50 (18.3%) |
| Bachelor's degree | 88 (32.1%) |
| Master's degree | 54 (19.7%) |
| Doctorate | 19 (6.9%) |
3.2 Parental NF1 status and level of NF1 knowledge
The first study aim was to evaluate whether a parent's NF1 status (self-affected, other parent affected, or unaffected) is associated with their level of NF1 knowledge. A one-way ANOVA, F (2, 271) = 7.79, p < .001, determined there was a significant difference between the three groups. Follow-up post hoc comparisons using Tukey's HSD (honestly significant difference) test concluded that the self-affected group had significantly lower NF1 knowledge, as measured by NF1 knowledge scores, than both the other parent affected group, p < .001 and the unaffected group, p = .004. The mean NF1 knowledge scores for each group were as follows: self-affected (M = 77.68%, SD = 0.09), other parent affected (M = 84.36%, SD = 0.08), and unaffected (M = 82.77%, SD = 0.09). The results of the post hoc pairwise comparisons, which support the hypothesis that parental knowledge of NF1 is negatively associated with having a personal diagnosis of NF1, can be found in Figure 1.
3.3 Genetic counseling and level of NF1 knowledge
The second study aim was to evaluate whether genetic counseling is associated with parental knowledge of NF1. Overall, 71% (194/274) of respondents reported to having seen a genetic counselor, while 29% (80/274) of respondents had not seen a genetic counselor. Parents who had received genetic counseling scored significantly higher than those who had not [t (135.41) = 3.266, p < .001]. These results support the hypothesis that parental NF1-related knowledge can be increased with genetic counseling (Figure 2).
3.4 Lowest scoring NF1 knowledge items
Apart from the two main study aims, the percentage correct was calculated for each knowledge item to describe, which were the lowest scoring for the entire sample. When averaged, the whole sample had a total NF1 knowledge score of 81.0% (30.78/38.00). Respondents were less likely to correctly answer questions concerning the increased risks of cancer in NF1. For the question ‘Are women with NF1 at an increased risk for breast cancer?’, there was a correct response of 45.3% (124/274) overall, and ‘People with NF1 are generally more likely to develop cancer at a younger age?’, had a correct response rate of 35.7% (98/274). When asked, ‘What is the increased risk for a person with NF1 to develop cancer in their lifetime compared with someone without NF1?’, only 29.9% (82/274) of all participants answered correctly. Percent correct responses for each NF1 knowledge item, depicting these three NF1 knowledge items having the lowest overall percentages, can be found in Table 2.
| NF1 knowledge items | Correct (%) |
|---|---|
|
What is the increased risk for a person with NF1 to develop cancer in their lifetime compared to someone without NF1? (10%) |
29.9 |
|
People with NF1 are generally more likely to develop cancer at a younger age. (True) |
34.7 |
|
Women with NF1 are at an increased risk for breast cancer. (True) |
45.3 |
|
How many copies of a non-working NF1 gene must one have to be affected with NF1? (One) |
46.7 |
|
Are seizures associated with NF1? (Yes) |
54.7 |
|
NF1 can skip generations. (False) |
57.7 |
|
People with NF1 from different families will always have different symptoms. (False) |
59.1 |
|
What is the function of the NF1 gene in the body? (Stops tumors from growing in the body) |
65.3 |
|
Are congenital heart defects associated with NF1? (No) |
68.6 |
|
Half of people with NF1 have a family history of NF1. (True) |
69.7 |
|
How is NF1 diagnosed? (All of the above; Physical examination, genetic testing, prenatal genetic diagnosis) |
70.4 |
|
Is high blood pressure associated with NF1? (Yes) |
72.6 |
|
A baby with NF1 may be born with a tumor. (True) |
75.5 |
|
There is more than one gene that causes NF1. (False) |
77.7 |
|
One half or 50% of genetic information is passed down from mother to child. (True) |
79.2 |
|
What types of exams may be helpful when trying to make a diagnosis of NF1? (All of the above; Skin examination, imaging (such as X-Ray or MRI), eye examination, genetic testing) |
79.6 |
|
Are men or women more likely to be affected by NF1? (NF1 affects both men and women equally) |
79.9 |
|
Are clubbed feet associated with NF1? (No) |
84.7 |
|
A father can pass down an NF1 gene mutation to his daughters. (True) |
87.6 |
|
Are allergies associated with NF1? (No) |
88.0 |
|
Is infertility associated with NF1? (No) |
89.8 |
|
What is the most concerning symptom of a tumor to suggest it has become malignant (cancerous)? (When the tumor grows rapidly and becomes painful) |
90.5 |
|
Is attention-deficit hyperactivity disorder associated with NF1? (Yes) |
90.9 |
|
Is small head size associated with NF1? (No) |
91.2 |
|
Is scoliosis associated with NF1? (Yes) |
91.6 |
|
Are cataracts associated with NF1? (No) |
92.3 |
|
If a person with NF1 has a child what is the chance that the child will have NF1? (50%) |
93.1 |
|
How do the symptoms of NF1 change as a person gets older? (A person with NF1 can develop more symptoms as they get older) |
94.5 |
|
During childhood what kind of screening may be done for optic gliomas (eye tumors)? (MRI and/or annual eye exams)) |
94.9 |
|
Are lisch nodules (dark spots on the iris. the colored part of the eye) associated with NF1? (Yes) |
95.3 |
|
Are optic gliomas associated with NF1? (Yes) |
96.0 |
|
A person is born with NF1. (True) |
96.4 |
|
Tumors in NF1 can appear anywhere in the body. (True) |
96.4 |
|
A person with NF1 can develop tumors that may lead to vision loss or blindness. (True) |
96.7 |
|
Are bumps on the skin associated with NF1? (Yes) |
97.4 |
|
Are learning disabilities associated with NF1? (Yes) |
98.2 |
|
If a woman with NF1 has scoliosis, then her child will also develop scoliosis. (False) |
98.5 |
|
Café au lait spots (brown marks on the skin) are often the first sign that a person has NF1. (True) |
98.5 |
|
What causes NF1? (A mutation in a gene) |
98.5 |
|
If a woman with NF1 has one child with NF1 her second child will definitely not have NF1. (False) |
98.9 |
|
All cases of NF1 can be detected in a woman's pregnancy by ultrasound. (False) |
98.9 |
|
All people who have an NF1 gene mutation will develop cancer. (False) |
99.6 |
|
NF1 symptoms vary from one person to another. (True) |
100 |
4 DISCUSSION
The goal of the current study was to understand potential factors that may be associated with parental NF1-related knowledge in order to identify knowledge gaps and specific populations where increased education could be beneficial. As hypothesized, the results of the current study found both that affected participants had significantly lower levels of NF1 knowledge than unaffected participants and that participants who had not received genetic counseling had significantly lower NF1 knowledge scores than those who had. The positive association of genetic counseling with NF1 knowledge further supports conclusions of previous research (Benjamin et al., 1993; Rosnau et al., 2017). To our knowledge, this study is the first to identify an association between parental NF1 status and NF1-related knowledge and serves as an important addition to the existing body of research.
The large sample size of this study not only serves as the main strength of our study; it also reveals a sizable group of parents who desire involvement in NF1 research. Interestingly, most respondents who participated in the current study were unaffected parents. Compared with the Benjamin et al. (1993) study, which included 81 participants, 69% being affected adult participants and 31% being unaffected parents of isolated cases, our study had 274 participants with only 19% being self-affected. We are unable to account for why a lower number of affected parents participated in the study compared to unaffected parents. This finding may be interesting to explore in future research.
4.1 Lower NF1 knowledge among affected parents
The average total NF1 knowledge score in our sample of parents with and without NF1 was 81.0%, a result that is consistent with results by Rosnau et al. (2017) in a sample of 49. However, our data also revealed that parents with a personal diagnosis of NF1 had lower NF1 knowledge scores than parents who did not. A previous study by Carrieri et al. (2016) interviewed NF1 families and asked how they were influenced by the variability of symptoms and uncertainty of severity in NF1. The study discussed a common trend of affected participants underestimating or ‘downplaying’ the possible or actual severity of NF1 in themselves and family members. A hypothesis is that this emotional underestimation among NF1 families may contribute to the lower levels of knowledge witnessed among affected parents in our study.
Furthermore, Carrieri's research also found that surveillance of other affected family members’ symptoms served as the main tool to manage NF1 concerns among affected respondents (Carrieri et al., 2016). Riley et al. (2013) describe the importance of comprehension of phenotypic variability of genetic conditions and its crucial role for diagnosis and surveillance. Previous research, coupled with the results of our study, lead us to hypothesize that affected parents may limit their exposure of NF1-related knowledge to other affected family members. Therefore, affected parents may not be receiving current and accurate information from medical providers who are likely the most knowledgeable about NF1.
4.2 Higher NF1 knowledge with genetic counseling
The aim of the second study found that parents had higher NF1 knowledge scores if they reported having previously met with a genetic counselor. However, 29% of our participants reported they had not seen a genetic counselor. Given this positive association between meeting with a genetic counselor and NF1 knowledge, it is important to understand why some parents face barriers to meeting with a genetic counselor. It is possible that some participants selected ‘no’ to seeing a genetic counselor because they had received counseling from another type of health care provider, such as a geneticist or nurse practitioner. Previous research has also found that patients decided against genetic counseling due to the perception that they already fully understood their results and did not need further explanation (Darst, Madlensky, Schork, Topol & Bloss, 2013). Our study results identify an opportunity for further exploration into what perceptions or factors may determine exposure or uptake of genetic counseling and emphasize the positive association between NF1 knowledge and meeting with a genetic counselor.
4.3 Lowest scoring NF1 knowledge items
The three lowest scoring questions of the survey were questions regarding NF1-related cancer risks and the potential for early-onset NF1-related cancers. Some NF1 cancer-risk associations, such as women with NF1 having an increased risk for breast cancer leading to an earlier screening protocol (Laguado, Hurtado & Gomez, 2019), are from newer literature that was not defined until recently (Uusitalo et al., 2016). Our results highlight potential knowledge gaps in NF1-related cancer risks, as well as emphasize an opportunity for genetic counselors to discuss current literature and terminology concerning NF1-related cancer risks with families.
4.4 Study limitations
As mentioned previously, the lower knowledge scores in the affected group may be related to the higher prevalence of learning disabilities in some NF1 patients (Bilder et al., 2016; Kolesnik et al., 2017; Potvin, Hardy & Walsh, 2015; Wiener, Battles, Bedoya, Baldwin & Widemann, 2018). Although efforts were made to create a simple and concise survey, and survey questions were written with the consideration that participants may have learning disabilities, it still is an important factor when assessing NF1 knowledge of affected individuals. Patient representatives, including affected and unaffected parents, reviewed the survey prior to distribution to ensure adequate comprehension. Regardless, we acknowledge the potentially higher incidence of learning disabilities among affected parents may have impacted the results of this study.
Another study limitation relates to the survey question regarding whether parent participants had seen a genetic counselor. Participants may not be aware of what a genetic counselor is and, therefore, may have marked ‘no’ to meeting with a genetic counselor but did in fact receive NF1 genetic counseling from a doctor or other healthcare professional. If this were the case, we would not expect a statistically significant difference in the NF1 knowledge scores between those who reported meeting with a genetic counselor and those who did not. Clarification of what genetic counseling is and defining other healthcare professionals who may also provide adequate genetic counseling in future studies would further benefit research.
4.5 Research implications
These findings are important not only for parents of children with NF1 but also for health care providers responsible for explaining the multifaceted complexities of NF1 to affected and unaffected parents. Results of the current study suggest affected parents may require more in-depth education of NF1. Further research could continue to examine these groups of parents and investigate whether more or different types of educational tools increase NF1 knowledge. Looking at the field of genetics more broadly, future studies with the goal of increasing parental knowledge of highly variable conditions, such as NF1, may be advantageous to pediatric care and lead to improved management and outcome.
5 CONCLUSION
Parental knowledge of NF1 may play a role in children with NF1 receiving appropriate medical care and surveillance. Genetic counselors are often involved in the education of families with NF1 and may benefit from being more aware of factors that may influence parental knowledge and comprehension, such as parental NF1 status. This study has identified significantly lower levels of NF1 knowledge among affected parents, and parents who have not met with a genetic counselor. Furthermore, possible NF1-related cancer risks appear to be least familiar to parents of children with NF1. Recognizing possible determinants of NF1 knowledge, and gaps in knowledge, exposes areas in which genetic counselors and NF1 healthcare providers can work to increase NF1 education and healthcare.
AUTHOR CONTRIBUTIONS
Emily Solem agrees to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. Emily Solem, Michelle Primiano, Monika Zak Goelz, and Marshall McQuillen made substantial contributions to the conception and design of the work. Marshall McQuillen and Emily Solem made substantial contributions to the analysis and interpretation of data for work. Emily Solem drafted the work while Michelle Primiano and Monika Zak Goelz revised it critically for important intellectual content. Emily Solem, Marshall McQuillen, Michelle Primiano, and Monika Zak Goelz authorized final approval of the version to be published.
ACKNOWLEDGEMENTS
Heather Radtke, CGC at Children's Tumor Foundation played an instrumental role in the development and execution of the survey and manuscript.
COMPLIANCE WITH ETHICAL STANDARDS
Conflict of interest statement
Emily Solem, Michelle Primiano, Marshall McQuillen, and Monika Zak declare that they have no conflict of interest.
Human studies and informed consent
This study was approved by the Long Island University Sponsored Research IRB protocol and conformed to recognized standards stated in the Declaration of Helsinki. All persons gave their informed consent for this study, and participant anonymity is preserved throughout the article.
Animal studies
No non-human animal studies were carried out by the authors for this article.
Open Research
DATA AVAILABILITY STATEMENT
Data supporting the results in the study are archived in the following repository: http://github.com/marshallm94/NF1_Knowledge_Analysis
