OBJECTIVE: The goal was to measure parents' interest in obtaining predictive genetic testing for their children for a disease that has no treatment.
METHODS: We conducted a nationally representative, Internet-based survey of parents. Parents received 2 vignettes describing hypothetical diseases without treatments. The first vignette described a disease in which symptoms were severe and timing of symptom onset was uncertain. The second vignette described a disease in which both symptom severity and timing of symptom onset were uncertain. Parents reported their interest in predictive genetic testing for their youngest child in each vignette by using a 5-point Likert scale. We collected information on parents' demographic features and parents' general opinions about genetic testing. We conducted logistic regression analyses to evaluate the independent associations between these factors and parents' interest in testing for both vignettes.
RESULTS: The response rate was 71%. For the first vignette, 35% of parents were definitely or probably interested in predictive genetic testing for their youngest child, 35% were unsure, and 31% were definitely or probably not interested. Responses were similar for the second vignette. More than one fourth of parents consistently expressed interest in testing in both vignettes.
CONCLUSIONS: Approximately one third of parents are interested in predictive genetic testing for their children, even for disorders with no treatment. Uncertainty about the severity of symptoms was not related to parents' testing preferences.
Predictive genetic testing can provide individuals with data on their numeric risks of developing future diseases on the basis of their genetic profiles. Although predictive genetic testing is not yet offered routinely in doctors' offices, individuals can obtain such testing through Internet-based companies (eg, 23andMe, Mountain View, CA; deCODEme, Rekjavik, Iceland; Navigenics, Foster City, CA) or by asking their doctor to coordinate testing through medical laboratories outside their immediate health care settings (eg, through the laboratory directory of GeneTests, Seattle, WA). Predictive genetic testing is currently available for a number of different types of conditions, ranging from those of physiologic significance only, such as athletic endurance (Atlas Sports Genetics, Boulder, CO), to preventable or treatable conditions, such as type 2 diabetes mellitus,1 to conditions without means of prevention or treatment, such as amyotrophic lateral sclerosis.2
In pediatrics, the situation is further complicated by the fact that parents often are charged with the responsibility of deciding whether to obtain predictive genetic testing for their children. Currently, the American Academy of Pediatrics3 and the American Society of Human Genetics and the American College of Medical Genetics4 recommend against predictive genetic testing of children, especially under circumstances in which symptoms do not develop until adulthood and/or diseases do not have an effective treatment. This recommendation is based on consideration of the balance between potential benefits and harms, weighed in conjunction with the ethical principle of autonomy. For example, if the balance between benefits and harms is uncertain, then professional societies recommend deferring genetic testing decisions until minors can make competent autonomous decisions for themselves, in adolescence or adulthood.
These professional societies' recommendations have been provided largely in the absence of information regarding parents' preferences. Although anecdotal evidence of parental interest exists5 and some Internet-based companies provide predictive genetic testing for children,6 we do not have a clear understanding of how interested parents would be in such testing for their children. Moreover, parents' interest is likely to be influenced by a number of factors, such as family history, disease severity, and/or available treatments.
To inform the discussion about predictive genetic testing for children, we surveyed a nationally representative sample of parents to examine their interest in predictive genetic testing for their children. Given the range of predictive genetic tests available, we thought that it would be most interesting to examine parents' interest regarding the extreme circumstances of testing for disease when no treatment exists and when there is uncertainty about the age of symptom onset and disease severity.
Overview of Study Design
We conducted an Internet-based survey of a nationally representative sample of US parents in March 2007. This study was approved by the University of Michigan institutional review board.
We selected participants randomly from a nationally representative online panel of individuals ≥18 years of age that is maintained by Knowledge Networks (Menlo Park, CA), a survey research firm. This online panel, titled KnowledgePanel, was created through probability sampling of both online and offline members of the US population through random-digit dialing of listed and unlisted numbers. If necessary, participants were provided with Internet access and computer hardware free of charge.
For this survey, parents were oversampled. A parent was defined as a respondent who was the parent or legal guardian of a child ≤17 years of age who was living in the same household.
Data Collection Instrument
To establish a common level of understanding, we provided parents with background information on genes, DNA, and genetic testing (Appendix). The reading level for the background section of the survey was below eighth-grade level, whereas the reading level of the survey was ninth- to tenth-grade level.
Each parent was then presented with 2 vignettes that described hypothetical diseases without treatments for which a predictive genetic test exists (Fig 1). The first vignette described a disease that had severe symptoms and an uncertain age of symptom onset (eg, childhood or adulthood). The second vignette described a disease with symptoms that ranged from mild to severe and with an uncertain age of symptom onset.
We elicited parents' general opinions about genetic testing, and we used their responses to validate internally their responses to the vignettes. To do so, we asked parents to indicate which of the following statements best reflected their opinions about genetic testing. (1) Genetic testing for a disease should be performed only when an effective treatment for the disease is currently available. (2) Genetic testing for a disease should be performed even if no effective treatment for the disease is currently available. (3) Genetic testing for diseases should not be performed at all. (4) I do not have an opinion on genetic testing for diseases.
At the end of each vignette, we asked parents, “Would you want to have your child get this genetic test? Please answer for your youngest child.” We asked parents to answer for their youngest child by using a 5-point Likert scale. For our descriptive analyses, we grouped respondents into the following categories: definitely/probably versus unsure versus definitely not/probably not. Finally, we collected data on the following parent demographic features: age, gender, education, household income, and insurance status.
For our descriptive statistics, we used bivariate analyses to measure parents' interest in testing their youngest children in each of the disease vignettes (Fig 1). We conducted simple and multivariate logistic regression analyses to evaluate the associations of parent demographic features with parental interest in testing. For our logistic regression outcome, we compared parents who were interested (definitely/probably) in testing in both vignettes with all other parents in the sample.
We also conducted logistic regression analyses to evaluate the relationship between parents' general opinions about genetic testing and their interest in testing their child in each vignette. All analyses were performed with Stata 10 (Stata, College Station, TX), using sample weights provided by the survey vendor to draw nationally representative inferences.
The 1342 parents who participated represented a response rate of 71%. The demographic features of the parents are presented in Table 1.
Predictive Genetic Testing for a Disease With Severe Symptoms and Uncertain Onset
Parents were divided regarding predictive genetic testing for their children for a disease with severe symptoms and uncertain time of onset. Just more than one third of parents definitely or probably wanted testing for their children, whereas 35% were unsure about such testing (Fig 2).
Predictive Genetic Testing for a Disease With Uncertain Symptom Severity and Onset
When both disease symptom severity and time of symptom onset were uncertain, we found a pattern of parental interest in testing that was very similar to that found in the first vignette (Fig 2).
Consistent Interest in Predictive Genetic Testing in Both Disease Vignettes
More than three fourths (77.3%) of parents reported the same level of interest in genetic testing for their children in each vignette. Of the parents surveyed, 27.6% were definitely or probably interested in predictive genetic testing for their child in both vignettes. Parents with this level of interest in testing were more likely to self-report Hispanic ethnicity (Table 2).
Parents' General Attitudes About Genetic Testing and Their Interest in Predictive Genetic Testing for Their Children
Parents were approximately evenly divided in their general attitudes about genetic testing according to the availability of effective treatment; one third favored genetic testing only when an effective treatment existed, one third favored testing even if no effective treatment existed, and one third had no opinion (Fig 3). In each vignette, parents' interest in predictive genetic testing for their children was strongly associated with the parents' general interest in testing (Table 3).
To our knowledge, this study represents the first national survey of parents' interest in predictive genetic testing for their children for diseases for which no treatment exists. Professional medical organizations do not currently support predictive genetic testing for children in situations in which no treatment exists for the disease being tested or situations in which the disease does not present in childhood.3,4 However, we found that more than one third of parents were interested in predictive genetic testing for their children for a disease for which no treatment exists and the disease develops in adulthood. A similar proportion of parents were interested in testing when we posed a scenario that also incorporated a range of possible symptom severity. More than one fourth of parents were consistently interested in predictive genetic testing for their children, regardless of the range of potential severity of disease symptoms.
These findings have important implications, now that parents have the ability to access genetic testing through private companies that advertise via the Internet. If parents endorse testing in these “extreme” circumstances, then they may be equally or more likely to endorse predictive genetic testing when treatments or preventions exist, which represents the majority of predictive genetic tests available via the Internet. Such easy access to the latest genetic technology also challenges the role of the physician as a gatekeeper. Indeed, patients already can access medical testing without the aid of a physician, with one example being nongenetic screening tests offered by private companies (eg, Life Line Screening, Independence, OH), and patients seem to be doing so, given that the number and distribution of self-referred, body-imaging centers has increased.7
The availability of Internet-based genetic testing also may alter the influence of physicians' recommendations on patients' decisions. One important reason may be that these companies have framed genetic testing as an issue of rightful ownership of one's personal health data and not as a medical problem. For example, the Internet-based, genetic testing company 23andMe informs potential consumers, “Though we store and help you interpret it, your genetic information is yours to have and explore. 23andMe provides you with all your data and will never withhold it from you.”8 In addition, studies in adults found no significant association between physicians' recommendations and patients' intentions to pursue genetic testing.9 In addition, some patients think that they should be able to undergo genetic testing even when a physician recommends against it.10 Given these findings, it is plausible that parents might opt for genetic screening strategies that the medical establishment does not recommend.
The only demographic factor we found to be independently associated with parents' consistent interest in testing was parents' self-report of Hispanic ethnicity. This finding is supported by a national survey of adults, which found Hispanic subjects, compared with white subjects, to be more supportive of predictive genetic testing when no treatment exists, even controlling for other demographic variables.11 The reasons for these ethnicity-associated differences remain unclear and merit further investigation.
Our study has limitations that should be noted. First, the vignettes posed to parents were hypothetical and did not name a specific disease. To make the vignettes more compelling for parents, we described the disease as serious and potentially life-limiting. Although we cannot generalize to a specific disease, our findings provide important insights regarding parents' attitudes toward predictive genetic testing in situations in which no treatment exists. Our study provides a foundation on which future studies can evaluate how specific disease scenarios modify parents' attitudes toward predictive genetic testing for diseases without treatment.
Second, we acknowledge that parents might have interpreted the terms “no treatment” and “no effective treatment” differently. Some parents might consider any treatment short of a cure to be ineffective, others might consider a life-prolonging, noncurative treatment to be effective, and still others might consider palliative, noncurative treatments to be effective. Regardless of how parents interpreted these terms, predictive genetic testing can be obtained for even the most-stringent interpretation of what constitutes an untreatable disease, one example being amyotrophic lateral sclerosis. Most predictive genetic tests currently available through Internet-based companies are for diseases for which some intervention (palliative, life-prolonging, or preventive) exists. The availability of predictive genetic testing for children for chronic, adult-onset diseases (eg, diabetes mellitus, heart disease, and hypertension) might once have seemed far-fetched. Today, parents can obtain such tests through Internet-based genetic testing companies.6 Given this increased availability of genetic testing, along with a persistent therapeutic gap in genetic testing-based medical care,12 it is not inconceivable that genetic tests for conditions without effective treatment might become more readily available to parents in the future.
Third, we did not collect data on the age of the youngest child for which the parent answered, and it is possible that parents' interest was influenced by the ages of their children. Fourth, the reading level for our questions was ninth- to 10th-grade level, whereas the reading level for the background section on genetics was just below the eighth-grade level. This raises potential concerns regarding the ability of some participants to understand the nuances and complexity of the vignettes and predictive genetic testing. However, we think that the strong association between the general attitude that genetic testing should be performed for a disease even when no effective treatment exists and parents' interest in predictive genetic testing for their children when no treatment exists indicates strong internal validity and suggests that parents understood the concepts behind the vignettes. Finally, although we asked parents' about their interest in testing, we did not measure testing use. We think that current rates of such testing are too low to have been measured in our study, and we think that our study provides important insights into parents' potential behavior.
Examining issues regarding predictive genetic testing for children before the broad commercial availability of genetic testing technology should give health care providers an understanding of the issues that parents are likely to discuss with their physicians, as well as factors that may motivate parents' interest in predictive genetic testing. We hope that these findings prompt health care professionals and the medical community to be proactive in the face of the rapid advances in genetic testing technology that are bringing predictive genetic testing from the examination room to the family room.
What Are Genes and How Do They Relate to DNA?
Each person has a set of genes that provides instructions for the development and functioning of his or her body. These genes are carried in the form of DNA.
How Do Genes Affect Health?
Genes provide instructions for the production of proteins in the body. Diseases can occur when instructions from a gene produce a defective protein or no protein at all. Sometimes, severe diseases result from these defective proteins. In other cases, people have only mild symptoms or even no symptoms at all.
What Are Genetic Tests?
Genetic tests use a few drops of blood to show whether an individual has specific genes that may increase the chance of getting certain diseases. Not all genes that can increase the chance of developing a disease have been discovered, so genetic tests will change with future medical research.
What Can Be Done With Genetic Tests?
Sometimes, if doctors identify individuals who have disease-causing genes, they can offer earlier or more-effective treatment. Other times, no effective treatment is available.
This project was funded by the Clinical Science Scholars Program at the University of Michigan (Dr Tarini). The funder had no role in any of the following: design and conduct of the study; collection, management, analysis, and interpretation of the data; and preparation, review, and approval of the manuscript. This research was conducted as part of the C.S. Mott Children's Hospital National Poll on Children's Health (www.med.umich.edu/mott/npch).
- Accepted April 10, 2009.
- Address correspondence to Beth A. Tarini, MD, MS, Department of Pediatrics, Child Health Evaluation and Research Unit, Division of General Pediatrics, University of Michigan, 300 N. Ingalls St, Room 6C11, Ann Arbor, MI 48109-0456. E-mail:
This research was presented at the American College of Medical Genetics meeting; March 14, 2008; Phoenix, Arizona.
The views expressed herein do not necessarily represent the views of the University of Michigan.
Financial Disclosure: The authors have indicated they have no financial relationships relevant to this article to disclose.
What's Known on This Subject:
Professional societies recommend against predictive genetic testing of children for diseases that develop in adulthood and/or do not have treatments. However, these recommendations have been provided largely in the absence of information regarding parents' preferences.
What This Study Adds:
To our knowledge, this study represents the first national survey of parents' interest in predictive genetic testing of their children for diseases without treatments.
- ↵23andMe. Health and traits: type 2 diabetes. Available at: www.23andme.com/health/type2diabetes. Accessed January 9, 2009
- ↵Fanos JH, Gelinas DF, Miller RG. “You have shown me my end”: attitudes toward presymptomatic testing for familial amyotrophic lateral sclerosis. Am J Med Genet A.2004;129A (3):248– 253
- ↵Nelson RM, Botkjin JR, Kodish ED, et al. Ethical issues with genetic testing in pediatrics. Pediatrics.2001;107 (6):1451– 1455
- ↵Macur J. Born to run? Little ones get test for sports gene. New York Times. November 30, 2008:A1
- ↵23andMe. Core values. Available at: www.23andme.com/about/values. Accessed July 24, 2009
- ↵Gwyn K, Vernon SW, Conoley PM. Intention to pursue genetic testing for breast cancer among women due for screening mammography. Cancer Epidemiol Biomarkers Prev.2003;12 (2):96– 102
- ↵Holtzman NA, Watson MS, eds. Promoting Safe and Effective Genetic Testing in the United States: Final Report of the Task Force on Genetic Testing. Baltimore, MD: Johns Hopkins University Press; 1997
- Copyright © 2009 by the American Academy of Pediatrics