OBJECTIVE: Our objective was to describe how parents consider disease and test characteristics when making decisions about newborn screening.
METHODS: We conducted focus groups with parents from primary care clinics and interviews of parents from a genetics clinic (total of 45 participants). Participants discussed 7 vignettes about newborn screening that we developed and refined with the assistance of an expert panel. Two coders coded the data independently, compared coding, and resolved disagreements through discussion. Using framework analysis, we analyzed the data and identified how parents' preferences varied according to disease characteristics, test characteristics, and perceptions of the associated risks and benefits.
RESULTS: Study participants strongly supported population-wide screening for disorders with well-defined, effective treatments, even if the treatment (eg, a bone marrow transplant) had significant morbidity. However, particularly among primary care clinic participants, there were more-varied preferences and greater difficulty making decisions about disorders associated with older age at onset, less-accurate screening tests, or less-effective treatment. In those cases, many participants suggested optional screening. For all disorders, participants expressed a desire for more information to facilitate decision-making.
CONCLUSIONS: Participants supported newborn screening for treatable disorders but suggested optional screening for other disorders. The variable influences on parents' decision-making suggest that parents with diverse experiences, if they were included in decision-making regarding screening policies, could provide critical perspectives and help screening programs address parents' preferences and meet parents' information needs.
WHAT'S KNOWN ON THIS SUBJECT:
Newborn screening programs are expanding rapidly, and suggestions have been made to change the criteria for evaluation of screening tests. Parents of affected children typically support screening for various conditions, but other parents' preferences are less well known.
WHAT THIS STUDY ADDS:
Parents' preferences differed according to experience with genetic conditions. However, most parents desired more-detailed information and suggested some optional testing. Understanding parents' decision-making processes and information needs would support development of screening policies that better address variations in preferences.
Newborn screening programs have grown rapidly, which is attributable to a combination of new research, new technologies, and the work of advocacy groups.1,–,4 Historically, diseases that fit defined criteria, such as having a well-understood natural history, being identifiable at a presymptomatic stage, and benefiting from early treatment, were considered candidates for population-wide screening.5,6 In recent years, suggestions have been made to expand criteria to consider benefits to other family members, facilitation of new research, and prevention of diagnostic odysseys.1,5,7 However, little is known about the newborn screening decisions parents would make if they were given a choice.
Previous studies of parents' attitudes toward newborn screening involved families affected by the targeted disease8,–,10 or considered specific, relatively well-known disorders.11,12 Those studies generally demonstrated support for newborn screening for specific conditions, but the disease specificity limited their generalizability. The studies provided little information about general attitudes toward newborn screening for other disorders. Other studies focused on psychological aspects of the newborn screening process, such as waiting for confirmatory testing,13 false-positive results,14 and identification of carrier status,15 rather than parents' preferences.
Work in newborn and prenatal screening has shown that parents may be influenced by previous experiences with disease8 and understanding of the clinical outcomes associated with a specific disease.12 A deeper understanding of parents' potentially diverse views could assist policymakers in planning for future newborn screening tests. Our study aimed to elucidate the factors parents would consider when making a decision about newborn screening, as well as the risks and benefits they perceive for such screening.
We conducted a qualitative study including focus groups and interviews with parents of young children. The study was determined to be exempt from review by the Partners HealthCare institutional review board and was approved by the Cambridge Health Alliance institutional review board.
Parents were eligible to participate if they were English-speaking and received care at 1 of 3 pediatric primary care practices or a hospital-based genetics clinic. We intended to recruit parents with <3-year-old children; however, after data analysis, we realized that the youngest child of 1 parent from the genetics clinic was 4 years of age. We sent invitational letters, which were cosigned by a physician in the identifying practice, to a random subset of eligible parents from each clinic. At 3 sites, parents returned response cards if they did not wish to be contacted for the study; at the fourth, parents returned the card if they were interested in participating. We called potential participants to arrange for their attendance at a focus group. Parents from the genetics clinic initially were invited to participate in a focus group but, because of the greater geographic dispersion of this group, focus groups proved to be unfeasible. Instead, these parents were invited to either an in-person group interview or an individual telephone interview.
Study Guide Development
We developed a semi-structured question guide that included vignettes presenting hypothetical disorders for which screening could be performed. Each vignette was designed to depict a disorder with a specific set of characteristics, including disease severity, timing, and treatability (Table 1). We used open-ended questions to elicit parents' preferences and perceptions of risks and benefits. Vignettes were based loosely on real conditions (eg, vignette 1 on phenylketonuria and vignette 4 on Huntington's disease). Vignettes are presented in the Appendix. A team consisting of a pediatrician, a geneticist, a newborn screening expert, and a parent reviewed and refined the vignettes for clarity, accuracy, and thoroughness. The guide was pilot-tested in a focus group and interviews, to improve cognitive validity.16
Written (or verbal for telephone interviews) consent was obtained. Focus groups were held at the primary care sites and lasted ∼1.5 hours. Interviews lasted ∼30 minutes. Participants received $50 in remuneration.
Each participant completed a demographic survey. Focus groups and interviews used 5 to 7 vignettes from the study guide, according to the discussion length. The focus group leader read the vignettes and then asked open-ended questions about preferences and for clarification of participants' responses. For consistency with how decisions are made in nonresearch settings, the leader did not clarify misunderstandings unless she was asked directly. All focus groups and interviews were audio-recorded and transcribed. After each focus group, team members (Drs Lipstein and Nabi) debriefed and summarized the group's themes.17 After each interview, the interviewing investigator (Dr Lipstein) wrote a summary of key themes. We verified each transcript by using the recording and notes taken during data collection.
We used framework analysis to analyze the data.18 Framework analysis is particularly suited to studies in which preexisting ideas inform the initial analysis.19 Our initial coding framework organized a priori themes, determined from the existing literature, into domains and themes. We then added emerging themes, those not considered before data collection, and related subthemes from early focus groups until we reached thematic saturation (Fig 1). In addition, a parent consultant reviewed all focus group transcripts to identify major and minor themes. Any new themes identified were added to the coding framework. The framework was tested and refined by having the 3 coding investigators (Drs Lipstein, Kuhlthau, and Nabi) compare coding for 1 transcript. Thereafter, each transcript was coded independently by 2 investigators. We resolved coding disagreements through discussion. Although many subthemes, such as treatability, could fall under multiple domains and themes (for example, benefits and influences), for clarity only 1 code was used. We have maintained the same practice in reporting our results.
After coding, data associated with each domain were compared across vignettes. Recurrent themes were then analyzed across the data sets, to assess variations in each theme and associations between the theme and participant characteristics.20 This triangulation allowed us to assess more thoroughly areas in which opinions converged or diverged.21 We used NVivo 8 (QSR International, Victoria, Australia) for all coding.
We completed 6 primary care focus groups, with a total of 40 participants, and 4 interviews (3 individual interviews and 1 group interview with 2 participants) with parents from the genetics clinic. The majority of participants were mothers with at least a college education, and 69% of participants self-reported their race as white (Table 2).
General Screening Preferences
Parents strongly supported population-wide screening for disorders with well-defined, effective treatments, even if those treatments had significant morbidity. Opinions were less consistent for disorders that had a later onset or had less-definitive (eg, symptom-based or experimental) treatments. In those cases, parents often suggested optional screening (Table 3). In contrast to other parents, whose preferences varied more, all parents recruited from the genetics clinic supported screening, except for vignettes 4 and 5, for which 2 parents from the genetics clinic did not support screening.
Domain 1: Perceived Benefits of Newborn Screening
The most-consistently mentioned benefit was treatment, even when the treatment itself had significant risk (eg, bone marrow transplant) (Table 4). For disorders without well-established, effective treatments, parents saw benefit in the opportunity for children to receive symptom-based, experimental, or future treatments. Many participants, including all from the genetics clinic, thought that newborn screening would allow for early identification of the problem and would prevent diagnostic odysseys.
Many parents perceived benefit in knowledge of the diagnosis even if definitive treatment was not available. Parents mentioned financial planning, anticipating future care-giving needs, and, in 1 focus group, making informed reproductive decisions as benefits of newborn screening. In addition, parents mentioned that such knowledge might encourage them to value time with their child or might compel their child to live life more fully. Less-frequently mentioned benefits included participation in research and joining support or advocacy groups.
Domain 2: Perceived Risks of Newborn Screening
In general, as the vignettes described less-treatable conditions, parents perceived increasing risk associated with screening (Table 5). Psychosocial impacts, including stress and worry, were the most-emphasized risks, especially in the focus groups. Parents thought that negative psychosocial effects might lead to child overprotection or, conversely, might impair parent-child bonding. For disorders with a late onset or less-accurate tests, parents thought that the wait for symptoms would be stressful. Several medical risks were mentioned frequently, including treatment side effects, false-positive results leading to unnecessary treatment, and decreased quality of life from needing follow-up testing.
Broader societal implications included labeling of the child, which parents thought might affect insurance, education, or employment, as well as costs of treatment, particularly for children without insurance. There were minimal concerns about genetic privacy, costs of testing, or parent employability after a diagnosis. Similarly, the fact that testing might reveal health information about other family members was rarely mentioned.
Domain 3: Influences on Newborn Screening Preferences
Aside from issues mentioned as benefits or risks, parents mentioned other influences on their preferences and decision-making (Table 6). They typically supported screening if early symptoms of the disorder were nonspecific and might be mistaken for common illnesses. For disorders without established treatments, parents were most likely to favor screening when the disorder was severe or had an early onset. For disorders with later onset and no benefit from early treatment, many parents suggested waiting to screen. They commented that this would allow parents to enjoy their infant without the prospect of disease or, in the case of adult-onset disorders, would permit children to make their own testing decisions. The disease characteristics influencing screening preferences did not vary significantly between the primary care and genetics clinic groups. However, parents from the genetics clinic tended to be more interested in screening, regardless of disease characteristics.
Parents were less likely to desire a screening test that was less accurate or indicated only susceptibility of the child to develop a disorder, without an available, timely, diagnostic test (vignette 5). For each vignette, some parents forgot or did not understand that screening tests would be followed by diagnostic tests.
Not surprisingly, personal experiences influenced parents' preferences. The experience cited most often was prenatal testing. Parents who mentioned experiences with more-extensive prenatal screening tended to support newborn screening more strongly. However, those who mentioned experiences with false-positive prenatal screening results often expressed concerns about test accuracy. Both their personal health and their child's health seemed to influence parents' preferences for screening, but not in a predictable manner. For example, although parents recruited from the genetics clinic, who by definition had a child with a chronic disease, tended to favor screening, the few parents from primary care clinics whose children had experienced significant health problems were more hesitant. Similarly, some parents who mentioned family medical history had favorable attitudes toward screening, but others who related similar histories had more-negative attitudes.
Domain 4: Information Seeking
Parents frequently requested more information about test and disease characteristics than the vignettes provided (Table 7). They inquired about test accuracy, particularly false-positive rates, and how quickly screening and diagnostic test results would be available. They wanted more details about treatments, disease severity, the likelihood of various symptoms, and the heritability of the disorders.
In several cases, parents stated that additional information was unlikely to change their preference but they would be more comfortable stating a preference if they had such details. Moreover, some parents acknowledged having made decisions about actual newborn screening without such detailed information. Parents from the genetics clinic were least likely to want more information and thought it might be overwhelming.
For vignettes in which the natural history of the disease was not well established or the disease had both infantile and adult forms, parents struggled to state a preference, given the limited availability of precise information. Many parents also sought information regarding screening alternatives, such as prenatal screening or testing later in life.
We found that parents supported screening for disorders with proven treatments but demonstrated less consensus about others and often suggested that parents should have the option to choose or to decline screening in those cases. Notably, we found that, although each group acknowledged that others may feel differently, parents recruited from a genetics clinic were more likely to support screening for a broad variety of conditions than were parents from a primary care clinic. These results are consistent with those of earlier studies that focused on specific named disorders,9,10,12,22 newborn screening program features,13,23 or predictive genetic testing.24 Our results contrast with those of a recent survey conducted in the Netherlands (where newborn screening is optional but is accepted by >99% of the population), which found that most parents favored screening for untreatable conditions.25 Most of the risks and benefits mentioned by study participants are similar to those found by researchers studying specific populations and disorders8,10,14,23 or considered by policymakers,3,5,7,26 including benefits such as treatment and planning for the future and risks such as false-positive results and parental worry. However, our use of descriptive vignettes of unnamed disorders allowed us to develop a broader understanding of how parents process screening information and make decisions on the basis of perceived risks and benefits.
The same risks and benefits were mentioned across individuals, but how parents considered each varied depending on disease characteristics, test characteristics, and personal health experiences. Similar to other studies with parents of children with genetic conditions,9,10 we found that parents recruited from the genetics clinic tended to support newborn screening broadly and to have less concern about false-positive results, parental worry, or treatment side effects. This difference between the groups of parents is particularly important because most screening advisory committees that include parent representatives seek out parents of affected children.27,28 On the basis of the differences we found, the inclusion of only parents of affected children may not allow for consideration of the range of parent preferences regarding newborn screening tests. Programs could include a greater diversity of parent opinions by considering research addressing parental preferences, using informational focus groups and key-informant interviews, or expanding parental representation on advisory committees to include parents representing varying experiences and viewpoints.
Parents' suggestion of optional testing for some disorders may reflect, in part, the fact that the study was conducted in Massachusetts, where part of the newborn screening panel is mandated and part is optional.29 However, our results indicate not only that parents accept varying degrees of risk from screening but also that some may be early adopters of new tests and treatments.30 By considering the possibility of an optional portion for newborn screening, programs may be able to adopt tests earlier while supporting parents' decision-making.31
Regardless of their preferences for specific tests, nearly all parents desired more information. This is particularly interesting given that the information provided in the vignettes was more detailed than that provided by most screening programs, including that in Massachusetts.32,–,34 Despite recognizing they had decided to participate in the optional portion of the Massachusetts newborn screening panel without receiving detailed information, parents were interested in having significantly more information than our vignettes provided. One way to meet these information needs is through the design of materials about newborn screening that address parents' information needs and promote informed decision-making. In addition, more-consistent prenatal presentation of newborn screening information, as recommended by the American College of Obstetrics and Gynecology, may better prepare parents for screening decisions.35
In reviewing the Web sites for states' newborn screening programs, it is apparent that most list diseases according to physiologic category, such as urea cycle defects or endocrinopathies. Our data suggest that screening programs may better serve parents by providing materials that explain clearly the differences among screening, diagnostic, and susceptibility tests and categorize disorders according to topics that influence parents' preferences, such as test accuracy, efficacy of treatment, and typical age of onset. Organization of the information in this way may help parents understand and contextualize the screening tests, which, in turn, may help them make informed decisions.
As in any qualitative study, our results may not be generalizable to all parents. We recruited participants from primary care practices that serve diverse parts of the community and then invited parents to a focus group held at their practice. This allowed us to increase the diversity of our study population while maintaining participant comfort through inclusion of a relatively homogeneous participant group in any given focus group.17 The inclusion of parents from diverse backgrounds helped to demonstrate both the variation in parent preferences and the areas in which most parents agreed. Despite recruitment from multiple settings, most participants were highly educated. To promote discussion among participants, all spoke English.
The lack of dialogue in the interviews might have limited opinions; conversely, in the focus groups, parents' opinions might have been influenced by their peers. However, using 2 different methods of data collection allowed us to increase the diversity of our participants, as measured by their experiences with genetic disorders. Moreover, variations in the process of expressing preferences are consistent with variations in how decisions are made outside the research setting.36
The fact that we included parent consultants in designing the study and analyzing the data increases our confidence that our results have face validity and are not unduly influenced by a priori ideas. In addition, the use of rigorous qualitative data analysis limited the influence of our own perspectives on the analysis. Vignettes were based on actual disorders or groups of disorders, but we presented all vignettes as hypothetical diseases to increase generalizability and to limit biases that might have arisen from familiarity with specific disorders.
In general, parents in our study preferentially supported newborn screening for disorders that fit traditional screening criteria.6,37 In other words, they wanted universal screening for disorders that had defined, disease-specific treatments, accurate tests, early onset, and well-understood natural histories. However, like many in the research and policy arenas,1,5,7 some parents suggested reasons to screen beyond the traditional criteria, such as exploratory treatments and financial planning. Rather than changing the criteria according to which tests are evaluated for mandatory screening,1,5,7 we suggest that not all newborn screening tests need to be mandated and programs should consider including an optional component. Such an approach would be more consistent with current parent preferences24 than the broad adoption of newborn screening tests that do not fit traditional criteria. By including parents in the discussion, determining the best methods to inform parents about newborn screening, and considering alternatives to exclusively mandatory screening, newborn screening programs may be better equipped to adopt future tests in a manner that respects parents' preferences and role in decision-making.
APPENDIX: STUDY VIGNETTES
The focus groups and interviews began with the following introduction. The introduction was followed by vignette 1. Thereafter, the order of vignettes varied between focus groups, with vignettes 7 and 3 being used only if time allowed.
“Imagine there are new tests that could be used to test all newborns for specific health problems. The chance a child has the health problem is very small. You can assume that the test can be done with blood that is already collected for other tests. In other words, no extra needle pokes are required. For all the tests, a positive test means the child may have the health problem, but further tests, which are available for all but one of these health problems, would be needed to make sure. I am going to describe the tests one at a time and invite you to discuss why newborns should or should not receive that particular test.”
Health Problem 1
This test is for a health problem in which babies appear healthy but, unless they are given special formula in place of regular formula, breast milk, or other high-protein foods, they will develop mental retardation. Once the mental retardation has occurred, the children may receive special education and therapy services, but the damage cannot be reversed. Children who are detected by a newborn test can be fed the special formula and will grow up fairly healthy and with normal abilities.
Health Problem 2
This test is for a health problem in which children initially grow and develop normally. Sometime between age 6 months and 2 years, they begin to have frequent vomiting and seizures and slowly develop brain damage. Treatments for this health problem have been studied, but none consistently prolongs the life of children with this health problem. Treatment is available for some symptoms but cannot cure or prevent the health problem. Children with this health problem typically die before their teenage years.
Health Problem 3
This test is for a health problem that is not well understood. It was discovered because some children with severe disabilities and seizures tested positive for this health problem. However, some children and adults who have never had any symptoms also have positive tests. There is no way to know from the newborn test whether or not the child will develop symptoms. If they do occur, symptoms are treated in the same way as they are for other children with the same symptoms.
Health Problem 4
This test is for a health problem that does not appear until one's early 40's. Before that age, people who test positive have the same health as others. As adults, they develop difficulties with both their physical and mental abilities. There is no way to prevent the disease from happening either before or after symptoms occur. Some symptoms may be treated; however, the disabilities get worse over several years and eventually lead to early death.
Health Problem 5
This test is different than the others. Rather than testing for the health problem, a positive result means that the child has a higher-than-normal chance of developing the health problem. Some children who have this high chance will develop the health problem, but others will not. Some children who have a negative test will develop the health problem, but most will not. There is no test that confirms who will or will not develop the health problem. Once the health problem develops, it can be treated with medicine specific for this health problem. There is no way to prevent the health problem from happening, and treatment cannot be started before symptoms occur.
Health Problem 6
This test is for a health problem in which children seem healthy for the first couple of months but then become very ill. Although some of the symptoms can be treated, the only cure is bone marrow transplantation. Without a bone marrow transplant, children with this health problem generally die by age 2. Transplants are not always successful, and when they are the transplant itself can lead to long-term health problems. If the transplant is done before a child develops symptoms from this health problem, it is much more likely to be successful.
Health Problem 7
This test is for a health problem in which most people develop severe symptoms as infants. These symptoms include mental retardation, heart problems, and seizures. Children with such symptoms usually die before age 2. However, about 20% of people with this health problem do not have any symptoms until early adulthood. As adults, they become weak and develop severe physical and mental disabilities, which eventually lead to early death. Treatment is available for this health problem. It improves symptoms and increases life expectancy in children with the health problem. It is unknown if the treatment prevents adults from developing symptoms. The treatment itself has side effects and must be given in special clinics.
The study was funded by a grant from the Deborah Munroe Noonan Memorial Fund. Dr Lipstein's effort was supported by the Health Resources and Services Administration, through grant T32 HP10018 to the Harvard Pediatric Health Services Research Fellowship Program.
We thank the staff members and physicians of Cambridge Pediatrics, Patriot Pediatrics, and Weston Pediatrics, as well as their patients, for assisting with this study.
- Accepted June 21, 2010.
- Address correspondence to Ellen A. Lipstein, MD, MPH, Center for Innovation in Chronic Disease Care, Cincinnati Children's Hospital Medical Center, MLC 7027, 3333 Burnet Ave, Cincinnati, OH 45229. E-mail:
This study was presented in part at the Pediatric Academic Societies meeting; May 4, 2010; Vancouver, Canada.
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
- Paul DB
- Wilson JMG,
- Jungner G
- Parsons EP,
- Clarke AJ,
- Hood K,
- Lycett E,
- Bradley DM
- Campbell E,
- Ross LF
- Tluczek A,
- Koscik RL,
- Farrell PM,
- Rock MJ
- Maxwell JA
- Krueger RA,
- Casey MA
- Bryman A,
- Burgess RG
- Ritchie J,
- Spencer L
- Pope C,
- Mays N
- Miles MB,
- Huberman AM
- Tarini BA,
- Singer D,
- Clark SJ,
- Davis MM
- Plass AM,
- van El CG,
- Pieters T,
- Cornel MC
- Howell RR
- 27.↵Advisory Committee on Heritable Disorders in Newborns and Children. Policies and procedures for operation and the development of recommendations for screening newborns and children for heritable disorders and for the Heritable Disorders Program: membership. Available at: www.hrsa.gov/heritabledisorderscommittee/governance/policiesandprocedures.htm#membership. Accessed October 23, 2009
- Rogers EM
- Fant KE,
- Clark SJ,
- Kemper AR
- 34.↵New England Newborn Screening Program. What parents should know—a summary. Available at: www.umassmed.edu/nbs/whatindex.aspx?linkidentifier=id&itemid=65350. Accessed October 23, 2009
- Copyright © 2010 by the American Academy of Pediatrics