Benefits and Burdens of Newborn Screening: Public Understanding and Decision-making

Stuart G Nicholls; Brenda J Wilson; Holly Etchegary; Jamie C Brehaut; Beth K Potter; Robin Hayeems; Pranesh Chakraborty; Jennifer Milburn; Daryl Pullman; Lesley Turner; June C Carroll


Personalized Medicine. 2014;11(6):593-607. 

In This Article

Benefits & Burdens of Expanded Newborn Screening

Many newborn screening programs were established in the latter half of the 20th century – initially testing urine[11] and later progressing to analysis of bloodspots – with a number of bloodspot screening programs arising in the late 1960s and 1970s.[12]

Traditionally, screening programs have emphasized the need for a clear understanding of the natural history of a condition together with a clear and immediate benefit to the child.[13] As indicated above, a motivation for NBS is the early identification of asymptomatic individuals in order to facilitate early intervention with the aim of reducing irreversible damage or mortality. Examples of conditions included within screening panels that meet these criteria include phenylketonuria (PKU), a disorder of protein metabolism that leads to irreversible neurologic damage with severe cognitive impairment when untreated,[14,15] or medium-chain acyl-CoA dehydrogenase deficiency (MCADD), which can result in life-threatening episodes that are preventable if the diagnosis is known.[16] In the case of PKU, treatment of affected newborns with early dietary modification can prevent the severe developmental disabilities associated with the condition.[14,15] For MCADD, a delayed diagnosis increases the risk of death or severe morbidity. In both cases detection within the first few days or weeks of life is necessary to maximize the benefit to be gained from treatment.

Since the start of the 21st century there has been a rapid expansion in the number of conditions included within newborn screening programs, particularly in the USA (see Watson et al. for a detailed report on expansion in the US[17] and Botkin et al. for a commentary[18]). Recent additions to screening panels, such as severe combined immune deficiency,[19] continue to emphasize the role of prompt identification and intervention for the newborn, yet others less clearly meet established criteria of being amenable to treatment or have been less well characterized with respect to the natural history of the condition, including development from latent to declared disease. As such, a number of recently included, and recently proposed conditions – such as Pompe disease, Fabry disease and Type 1 diabetes[20–25] – have been the subject of debate and evaluation.[26]

This expansion has brought with it not only an increase in the number of conditions for which screening is conducted, but also a broadening of disease boundaries and a reformulation of the principles on which screening programs are founded. Some proponents of NBS have argued for an expanded interpretation of benefit on the basis that initial criteria were created for adults and forego the family benefits that may be gained from the knowledge of risk of disease in a child.[27] Others contend that even archetypal cases that are seen as clear examples of the benefit of screening – such as PKU – have not been an unmitigated success,[18,28,29] and that only by screening has it been possible to gain a better understanding of the natural history of disease, the variant forms, and therefore improve treatment plans.[30] As such, there has been debate regarding the appropriate application of the original principles and revised criteria have been developed by different bodies and institutions. While these revised criteria retain the original principles of identifying asymptomatic individuals with the aim of early identification to improve outcomes, they also expand the focus beyond the individual to include consideration of the family, or society more broadly, but have also become more specific to the screening program under review, with the inclusion of more detailed assessment criteria.[17,31] Recent attempts have sought to synthesize these emerging criteria (Supplementary Table 1; see online at[32,33] It is to these expanded and emerging criteria that we turn our attention (Supplementary Table 1

Recent evaluation criteria employ both an expanded notion of benefit and of the beneficiaries of screening than were common in the earlier screening literature.[28] Both the UK National Screening Committee and the 2006 American College of Medical Genetics and Genomics (ACMG) report include family or societal benefits to differing degrees. In particular, there has been an emphasis placed on psychological and sociological benefits gained from the identification of disease and avoiding the diagnostic odyssey.[34–37] Hence, while NBS has traditionally been based on a 'public health emergency' model, in which early disease identification and urgent action is required to save life or prevent catastrophic morbidity, there has been a shift toward a 'public health service' model[38] in which major additional goals are the provision of information to inform decision-making, reduce the time to clinical diagnosis[34–36] or to provide more moderate interventions to reduce morbidity.

A more contentious connotation of benefit has been the use of reproductive decision-making as a primary justification for expanded screening.[39,40] While the use of newborn screening as a method of gathering information to inform future reproductive decisions has been justified on the basis that clinical diagnosis of an affected child may only occur after a second child is conceived,[41,42] this has generally been perceived as a secondary benefit to direct health interventions for the child undergoing screening. Recent studies of healthcare professionals, however, indicate that this view may be changing to emphasize reproductive choice as an outcome.[43,44] Bombard and colleagues, for example, note that "providers do indeed perceive the incidental generation of reproductive risk information through NBS as of one its primary benefits".[43] Indeed this has been reported as an explicit aim within the screening program for Duchenne muscular dystrophy (DMD) in Wales.[45] The suggestion that the provision of reproductive risk should be a primary benefit of newborn screening has caused concern for others.[43,44,46,47] In particular, some authors have argued that placing reproductive decision-making as a primary goal of newborn screening uses the child as a means for discovering parental reproductive risks, and is unethical,[48] transgressing the maxim that individuals should be treated not merely as a means to an end, but as an end in themselves. However, data on the actual use of information in future reproductive decisions, and the benefits that parents derive from such information, is mixed.[49,50]

A further implication of citing reproductive choice as a primary benefit of newborn screening is that it lessens the urgency of screening. If informing future reproductive choices is a primary goal, there are options for gaining such information other than newborn screening. It could be derived through parental carrier testing or early childhood screening. When intervention is not time-sensitive, the urgency argument disappears. A corollary of this is that under such nonurgent conditions, and where the primary benefit is to the parent, there is a diminished justification for mandating screening as competent individuals are usually ascribed the freedom and responsibility to make healthcare decisions for themselves.

Expanding the notion of benefit to include familial or social benefits – for example, using newborn screening as a method of generating information regarding the prevalence and natural history of a condition as opposed to having this information a priori – has also generated an implicit expansion of the burdens of screening. Such burdens arise, to some extent, from the inverse of many of the benefits: the production of greater amounts and more complex data brings with it the burden of how best to manage such complexity in ways that promote understanding while minimizing the potential psychological or emotional burden. The generation of data that are potentially useful for population and clinical research generates the burden of managing obligations with respect to privacy, parental understanding and decision-making. Indeed, while early identification and avoidance of the diagnostic odyssey is viewed as an important benefit, managing results, worry and anxiety that may come with an initial screen-positive result are burdens for NBS programs and resource intensive for implicated healthcare providers. With the expansion of newborn screening platforms to include a greater number of diseases or individual assays, and particularly if newborn screening expands to include whole-genome sequencing, there may well be an increase in the number of false-positive results (for an in depth discussion see Kohane et al.[51]) and a concomitant increase in the associated burdens.

A further implication of the expansion of screening is the potential for results of unknown clinical significance (sometimes referred to as 'overdiagnosis'[52,53] see also Raffle and Gray[11] for a discussion of this with respect to screening programs in general) or overtreatment.[1,53] When the natural history of the condition is not well understood, a child may be harmed through overtreatment, especially when the child might never have developed symptoms.[54] A recent case discussion of screening for Krabbe disease illustrates these difficulties.[55–57] Krabbe disease is an inherited, autosomal recessive, neurodegenerative disease caused by a deficiency in a lysosomal enzyme, galactocerebrosidase. Screening for Krabbe disease as part of the NBS program was implemented in New York by executive order. However, in five years of testing only four cases were identified at a cost of US$3.5 million. In addition, 20 parents of children were given information as to a potential diagnosis but with uncertainty regarding if, how or when symptoms might appear and what could be done.[55] The treatment involves hematopoietic stem cell transplantation using umbilical cord blood. Although treatment could potentially lead to better outcomes, it is experimental, involves chemotherapy, and presents substantial risks of morbidity and/or mortality.[55] Indeed, the long term impact of treatment, and uncertainty of identifying infant- or adult-onset variant forms, means that some children may be exposed to harms that may outweigh any immediate benefits. Ross[54,58] also cites the example of 2-methylbutyryl-coA dehydrogenase deficiency, a condition that had been diagnosed in only a handful of instances worldwide prior to its addition to the ACMG screening panel. Based on screening results, the identified children were placed on restrictive diets, although subsequently most appeared asymptomatic and it is now believed that the disorder identified is a normal variant.[54,58,59]

While the two examples differ in their context: one derives from uncertainty of results from a sought condition, the other an 'incidental' result brought about by the technology used to identify another condition; both represent examples where instigating treatment under uncertainty could have potential harms for the child in question. Aside from the potential for physical harms through restricted diets when variants are benign or invasive treatment, both also have the potential to perpetuate a lack of closure, thus extending parental stress, or anxiety.

In summary, newborn screening programs in the 21st century provide undoubted benefits where early identification of asymptomatic individuals can lead to early interventions that can reduce morbidity or mortality. Moreover, there exist opportunities to provide families with information that may inform future decisions and remove uncertainty where previously a diagnosis has been lacking. However, the expansion of screening also brings burdens for both families and newborn screening programs when the natural history of the condition is less well characterized, and where evidence suggests more modest or uncertain clinical benefits. Parents and programs must establish ways of dealing with this uncertainty and develop evidence-based approaches to mitigate the potential harms while continuing to derive the benefits from newborn screening. At a more general level, the balancing of such benefits and burdens and assessment of how they could, and should, be addressed by newborn screening programs needs to remain a focus of discussion.

In the remainder of the paper we consider the existing evidence regarding public – and primarily parental – attitudes toward the identified benefits and burdens of newborn screening, and current understanding of how parents integrate these into their decision-making.