Newborn Screening in the United States: Ethical Issues
Abstract and Keywords
The purpose of newborn screening is to identify and treat infants with certain conditions prior to the onset of symptoms in order to reduce morbidity and mortality. The development of new technology, including genomic sequencing, has expanded the number of conditions that can be detected in the newborn period. Whether infants should be screened for conditions for which there is no available treatment is a subject of debate. The retention and secondary use of de-identified residual newborn screening dried blood samples without explicit parental permission have been controversial. In light of these challenges, some question whether mandatory newborn screening continues to be justified. This chapter will explore key ethical issues faced by state newborn screening programs in the United States.
(p. 653) Introduction
Newborn screening is a public health program designed to identify and treat infants with certain conditions prior to the onset of symptoms in order to ameliorate or prevent disability and/or death. Shortly after birth, blood obtained from an infant’s heel is placed on a filter card and sent to a laboratory to test for biochemical and genetic markers for certain endocrine, metabolic, genetic, and infectious disorders. Newborn screening tests are used to identify infants who may be at increased risk for certain diseases so that these infants can be referred for diagnostic testing and, if necessary, treatment.
In the United States, newborn screening programs are operated by the state departments of health in all fifty states plus the District of Columbia. Almost all of the four million infants born in the United States each year undergo newborn screening, and approximately 12,500 infants are identified yearly with a newborn screening condition (CDC, 2012).
Despite this success, newborn screening has been controversial for three reasons. First, new technology has made possible the identification of a greater number of conditions. For many of these conditions, the natural history and benefit of treatment are not well understood. The determination of what conditions should be targeted is a constant challenge for state newborn screening programs. Second, there have been high-profile controversies over the retention and secondary use of residual blood samples left over after the screening tests are complete (Lewis, 2014). Third, newborn screening is mandated by state law in virtually all states; parental consent is not required although most states allow parents to opt out for religious or philosophical reasons. This chapter (p. 654) explores key ethical issues that newborn screening programs currently face and may face in the future.
The Development of Newborn Screening in the United States
Newborn screening began as a blood test performed shortly after birth to prevent the intellectual disability caused by untreated phenylketonuria (PKU), a rare, hereditary, metabolic disorder in which affected individuals cannot process the protein phenylalanine. Once a patient with PKU becomes symptomatic, irreversible brain damage has already occurred.
In 1960, Dr. Robert Guthrie developed a bacterial inhibition assay to detect elevated blood phenylalanine levels before the onset of symptoms. This discovery enabled the identification of affected individuals prior to the onset of symptoms so that they could be treated, thereby preventing the devastating consequences of this disease (Guthrie, 1992). By 1973, forty-three states had enacted legislation to screen newborns for PKU (AAP/HRSA Newborn Screening Task Force, 2000).
Expansion of Newborn Screening
Over time, new technology has made possible the simultaneous detection of increasing numbers of rare conditions using the same newborn screening filter cards that are used to detect PKU. The early evolution of state newborn screening programs occurred by the addition of a single, treatable condition, one at a time, to state newborn screening panels based on the premise that early identification and treatment would improve the health outcomes of affected infants. The development of tandem mass spectrometry (tandem MS/MS) technology in the 1990s challenged this paradigm. The use of the tandem MS/MS technology allowed the simultaneous identification of dozens of metabolic conditions—some of which were poorly understood, and some with no available treatment—using the same newborn screening filter cards that were used to detect PKU. Within a short period, most screening programs within the United States began to use tandem MS/MS as the principal tool for newborn screening analysis (Crowe, 2008).
In 1999, an American Academy of Pediatrics Task Force report commissioned by the US Department of Health and Human Services (HHS) noted that states varied widely with respect to the number of conditions included in their mandatory screening panels and in the mechanism for determining which tests should be included in the panels. Some states required screening for as few as two conditions, while other states required screening for more than forty.
(p. 655) While these inconsistencies in practice reflected differences in community values, public health technical capabilities, and political and economic realities, the practical effect of these inconsistencies was that infants across the country did not “have equal access to newborn screening and its potential to prevent morbidity and mortality” (AAP/HRSA Newborn Screening Task Force, 2000, 399). The Task Force called for a more uniform national policy for the selection of newborn screening tests.
In response to this recommendation, in 2002, the American College of Medical Genetics (ACMG) was commissioned to develop a uniform panel of conditions for screening. The subsequent 2006 ACMG report assigned conditions to one of three categories: (1) core panel, (2) secondary targets (conditions that are part of the differential diagnosis of a core panel condition), or (3) not appropriate for newborn screening. Of the eighty-four conditions considered, the ACMG report identified twenty-nine conditions as primary targets for newborn screening and an additional twenty-five conditions that could be identified in the course of screening for core panel conditions. Thirty conditions were determined to be inappropriate for newborn screening. The ACMG report recommended that both core panel conditions and secondary target conditions be included on state newborn screening panels, a significant expansion in the number of conditions targeted by newborn screening (ACMG, 2006).
The ACMG report was controversial. Some commentators argued that it would be unethical not to expand newborn screening for serious, potentially treatable conditions despite the lack of data on the efficacy of screening (Howell, 2006). Others advocated a more cautious approach, citing the lack of sufficient research about the impact of screening (Botkin et al., 2006). This debate raised legitimate questions about whether it is ethically appropriate to screen newborns for a large number of poorly understood conditions via a mandatory, state-based public health program.
Although the ACMG report acknowledged that the benefit to the child being screened should be the overriding consideration, an additional criterion for consideration was the benefit to the family and society of early identification of the condition. This broader definition of “benefit” was used to justify inclusion of conditions on the core panel for which there was only anecdotal evidence of effective treatment. The report cited the potential benefit of obtaining genetic information that could establish that others in the family may be at genetic risk of disease, the ability to inform subsequent reproductive decisions for the child’s parents, and the possibility of shortening the diagnostic odyssey as important factors that also should be considered. These considerations beyond direct benefit to the infant tended to lower the bar for inclusion of new tests on the recommended panel.
The methodology used to develop the ACMG recommendations was criticized for its failure to address key ethical issues, including the implications of mandating tests with uncertain results (Botkin et al., 2006). Critics of the recommendations suggested that instead of adopting the recommendations on a population-wide level, state programs move forward under a research paradigm to evaluate the risks and benefits of early detection by newborn screening (Botkin et al., 2006).
(p. 656) Despite these criticisms, the Advisory Committee on Heritable Disorders and Genetic Diseases in Newborns and Children (the Advisory Committee) adopted the ACMG panel as its Recommended Uniform Screening Panel (RUSP), and these recommendations were endorsed by the Secretary of HHS (Sebelius, 2010). The Advisory Committee is a federal advisory committee created to provide evidence-based recommendations for including (or excluding) conditions on the RUSP.
The RUSP establishes a standardized list of disorders that are recommended by the Advisory Committee, but individual state departments of health operate newborn screening programs. Each state decides which disorders will be included on its newborn screening panel, and states continue to vary with respect to the mechanism by which new conditions are added to the panels and the level of evidence required to make these decisions. By 2008 almost all of the states had added the twenty-nine core conditions to the state newborn screening panel, and most states had begun screening for a majority of the secondary target conditions (President’s Council on Bioethics, 2008). The Advisory Committee subsequently has recommended additional conditions for inclusion on the RUSP.
Future of Newborn Screening
The debate about the circumstances under which it is appropriate to add new conditions to state newborn screening panels has continued unabated. The New York experience is one example in which disease advocates successfully lobbied for inclusion of a condition on a state newborn screening panel despite the lack of clear evidence to support its inclusion. In 2006, in response to parental advocacy, the New York legislature voted to add Krabbe disease to the state panel, even though it was not included on the RUSP. In 2010, the Advisory Committee declined to add Krabbe disease to the RUSP, citing insufficient evidence regarding the benefits and harms of treatment of this disease (Kemper et al., 2010).
The argument that conditions should not be added to state panels without sufficient evidence has been countered by the view that identification of infants with rare conditions for which there currently is no available treatment will allow those children to participate in research on innovative therapies that may prevent the morbidity or mortality associated with that disease (Alexander and Van Dyck, 2006). However, this justification arguably is beyond the scope of the public health mission of newborn screening.
The prospect of conducting genomic sequencing as part of newborn screening further complicates this debate. In 2013 the US National Institutes of Health (NIH) provided grants totaling $25 million to four institutions to “explore the implications, challenges, and opportunities associated with the possible use of genomic sequence information in the newborn period” (NIH, 2012). These grants were intended to stimulate research to expand the scale of data available for analysis in the newborn period and to advance our understanding of specific disorders identifiable via newborn screening through DNA-based analysis. Despite the enthusiasm for this technology, state newborn (p. 657) screening programs and the US health care delivery system are not equipped with sufficient personnel to interpret sequence variants and provide appropriate clinical follow-up to the infants who would be identified as at increased risk of disease on such a large scale (Botkin and Rothwell, 2016).
Moreover, while genomic sequencing in the newborn period could produce results that might be of direct, immediate benefit to the health of the infant, other types of results with varying importance to the health of the child will be discovered for many newborns. If genomic sequencing were incorporated into newborn screening, state newborn screening programs would have to decide which results would be returned to infants’ families. In 2013, the ACMG published a list of fifty-six genes (revised to fifty-nine genes in 2016) that are highly associated with serious health conditions, for which we have treatment or prevention strategies, and for which we have sensitive and specific tests (Green et al., 2013). The ACMG recommended that, when sequencing is done for any clinical indication, the laboratory should look for variants in these fifty-nine genes. Many of these conditions are adult-onset diseases, such as breast or colon cancer, adding a whole new paradigm to the newborn screening enterprise. Although the ACMG specifically excluded newborn screening in its recommendations, research to evaluate the return of results of genomic sequencing in the newborn period currently is underway, as noted above.
In the past, genetic testing for adult-onset conditions has not been deemed to be in a child’s best interests, in part because testing in childhood precludes an individual from being able to make his or her own choice about whether to undergo testing for that condition in the future. The ACMG recommendation is inconsistent with this principle, but advocates of this practice have argued that it would be unethical not to look for this information and provide it to parents who might be at increased risk but otherwise would have no reason to undergo genetic testing themselves. In this situation, the ethical duties of clinicians to inform and educate parents and other family members who also may be affected is unclear, but counseling adult family members, including parents, about their increased genetic risk for adult-onset conditions is clearly beyond the scope of the public health mission of newborn screening.
Implementation of the ACMG recommendation in the newborn screening context would place tremendous strain on newborn screening programs. Current estimates are that one of the ACMG variants will be found in about 2 percent of those undergoing genome sequencing. If sequencing were performed on 4 million infants per year, 80,000 infants per year would be identified with a finding, compared with the 12,500 infants identified through newborn screening each year under the current paradigm. Under this new scenario, screening for the ACMG conditions could become the primary focus of newborn screening programs and draw programs far beyond their public health mission.
A second set of questions would arise over the disclosure of carrier results to families. Carriers are heterozygous for an autosomal recessive condition and are usually asymptomatic. However, if two carriers reproduce, there is a 25 percent chance with (p. 658) each pregnancy that an affected child will be born. Therefore, knowledge about carrier status can be helpful to adults of reproductive age. Several current newborn screening tests, such as those for cystic fibrosis and sickle cell disease, identify carriers, and most state programs alert parents when their child is identified as a carrier. Although carrier screening results in newborn screening have been disclosed for several decades, whether children identified as carriers through newborn screening programs use this information when they reach reproductive age has not been determined.
It has been estimated that each individual carries between 50 and 100 variants classified as disease causing; therefore, it is likely that if sequencing is used in the context of newborn screening, a high percentage, if not all, of newborns will be identified as a carrier for one or more conditions, many of which will be quite rare at the population level (1000 Genomes Project Consortium, 2010). The value of this information to newborns is questionable, particularly because adults of reproductive age can obtain carrier screening when considering or during pregnancy. Due to these considerations and others, several groups have stated that the use of genome sequencing as a screening tool in newborn screening programs is premature at best (Botkin et al., 2015; Johnston et al., 2018).
The continued variation between states regarding the operation of state newborn screening programs also raises distributive justice concerns. First, state newborn screening panels continue to screen for different numbers of conditions despite the creation of the RUSP. These differences mean that babies born in different states are screened for different conditions and therefore do not share equally in the potential benefits of newborn screening. Second, states vary with respect to the follow-up and treatment provided for infants identified with rare conditions through newborn screening (Botkin, Anderson, and Rothwell, 2012). In states that provide screening but not treatment for a particular condition, families may be unable to afford treatment, and affected infants may not benefit from screening. Third, when DNA-based analysis techniques are used, the selection of variants for molecular panel testing can exacerbate disparities if the variants chosen are not reflective of a genetically diverse and ethnically admixed population. The use of this technology can be problematic if the panels were designed prior to sufficient study of the mutation spectrum of nonwhite populations. In that case, the use of these panels would have limited value in diverse populations (Pique et al., 2017).
The Retention and Secondary Research Use of Residual Newborn Screening Dried Blood Samples
More blood is collected on the filter card than typically is needed. As a result, a small quantity of the blood sample often remains unused after completion of newborn screening testing. Residual newborn screening dried blood samples (DBS) can be used for newborn screening program operations, such as quality assurance and quality control, (p. 659) public health surveillance, and public health and biomedical research (Olney et al., 2006). Since newborn screening is performed on almost all of the babies born in the United States, DBS can be a tremendous source of population-level data.
Due to the mandatory nature of newborn screening, parents typically are provided limited education about the screening and the policies related to the retention and secondary use of DBS. Historically, parental consent was not required for the retention and secondary use of de-identified DBS because the samples were not considered human subjects for purposes of the federal human subjects research regulations. Nevertheless, the retention and secondary research use of de-identified DBS without explicit parental consent has been controversial. Whether explicit parental consent should be required to retain de-identified DBS and use them for secondary research, or, alternatively, whether parents should be informed about a state’s retention and use policy and then be given the option to refuse, has been a subject of debate.
Parental concerns about the retention and secondary use of de-identified DBS without consent led to litigation against the state departments of health in Texas (Beleno v. Texas Dept. of State Health Services, U.S. District Court, Western District of Texas, No. SA09CA0188 ), Minnesota (Bearder v. State, 806 N.W.2d 766 ), and Indiana (Doe v. VanNess, Marion County Superior Court, No. 49D011409CT031 ). In the Texas and Indiana litigation, the parents claimed that the state’s retention and secondary use of DBS without parental consent violated constitutional privacy rights. In the Minnesota litigation, the parents claimed that the state violated the genetic privacy provisions of the state Government Data Practices Act.
The Texas litigation was settled, and the Texas legislature passed new legislation that requires parental consent for the secondary research use of DBS. The Indiana lawsuit was dismissed. The Minnesota Supreme Court held that DBS could be used for newborn screening program operations without consent, but any other secondary use would require parental consent. As a result of the litigation, millions of archived DBS and decades’ worth of newborn screening data were destroyed (Lewis, 2015).
In 2014, the U.S. Congress passed the Newborn Screening Saves Lives Reauthorization Act (Reauthorization Act) to extend federal programs that assist states with the improvement and expansion of state newborn screening programs (Newborn Screening Saves Lives Reauthorization Act, Pub. L., 2014, 113–240). This law also required that DBS be considered human subjects for purposes of federal human subjects research regulations; therefore, parental consent was required for the use of DBS in federally funded research whether or not the samples were retained with identifiers. The Reauthorization Act also anticipated updated human subjects research regulations, and the provision that required that DBS be treated as human subjects and therefore subject to informed consent requirements for secondary research use was only to be in effect until the new, updated regulations were promulgated.
Ultimately, the final version of the regulation retained the old (pre-Reauthorization Act) provisions that require informed consent only for the use of identifiable blood or tissue samples, with no exceptions for DBS. However, parental consent is required in states, such as Texas, Minnesota, and Michigan, in which state law requires parental (p. 660) consent for the secondary research use of DBS. This controversy illustrates a larger tension in biomedical science between the value of residual clinical samples for research purposes and the public expectation that patients will be informed and offered choice regarding the research use of their biospecimens.
Should Newborn Screening Be Mandatory?
The original justification for newborn screening was the prevention of the devastating consequences of untreated PKU. Newborn screening was considered a public health initiative, and the goal to protect the health of infants was imperative. The need for informed consent prior to the performance of medical procedures did not become entrenched in American law until the 1970s. When newborn screening programs were becoming routinized, the question of whether to obtain parental permission for newborn screening did not receive broad consideration. Today, most states require newborn screening, and states vary regarding whether parents are permitted to refuse the screening (Lewis, 2014).
The increasing pressure to add tests for conditions for which effective treatment is unavailable, coupled with the controversy surrounding the retention and secondary use of DBS, makes the question of how to allocate decision-making authority for newborn screening among parents, states, and physicians less clear. The original justification for mandating newborn screening was the identification and timely treatment of infants with rare disorders in order to prevent the morbidity and mortality of disease. Some have argued that since the mission of newborn screening programs has expanded beyond the immediate health benefit of the infant to include broader benefits to the family and society, the original justification is no longer sufficient to warrant mandatory screening (Ross, 2011). Opponents of informed consent requirements for newborn screening argue that it is unethical to risk missing even one child with a treatable condition and that ethical arguments notwithstanding, logistical considerations, including the fact that there is no budget for training health care providers to provide informed consent or financial support for their time to do so, make informed consent impractical, if not impossible (Paul, 1999).
The relevant professional societies differ with respect to the optimal nature of the consent and refusal process. The American Academy of Pediatrics and the American College of Medical Genetics and Genomics “support the mandatory offering of newborn screening for all children,” and recommend that “parents should have the option of refusing the procedure” (AAP Committee on Bioethics, Committee on Genetics, and ACMG Social, Ethical, and Legal Issues Committee, 2013, 621). The American Society for Human Genetics supports improved parental education about newborn screening, (p. 661) but it “does not advocate a change in most state programs that mandate screening but permit parental refusals” (Botkin et al., 2015, 13).
If protection of newborns remains our primary concern, then all newborns should be screened. Nevertheless, it is possible to respect the wishes of parents with regard to less treatable disorders and research participation and still achieve the goals of testing all newborns for the more treatable disorders. A shared decision-making model, in which screening is mandatory for treatable conditions and parental consent is required to screen for less treatable conditions and for participation in research, would be one mechanism to achieve the dual goals of protection of the infant’s health and respect for parental autonomy. Such a two-tiered approach would require significant changes in the operation of state newborn screening programs, both at the level in which newborn screening samples are obtained and in the laboratories in which screening is performed.
The controversy related to DBS highlights the need for improved parental education about newborn screening and related issues. Parents must be better educated about the newborn screening process and given a role in decision-making regarding discoveries about the genetic makeup of their children. It is important that this education process includes information about the potential research use of DBS in states where DBS are retained for this purpose. The overarching goal should be to educate parents and the public about the purposes of newborn screening programs and how the information obtained will be used.
Newborn screening programs have prevented needless suffering by thousands of infants and their families. In the future, these same programs have the potential to address more conditions that cause disability and suffering for children and families. Newborn screening programs will continue to evolve, but the primary public health mission of protecting the health of children should remain paramount.
1000 Genomes Project Consortium. 2010. “A Map of Human Genome Variation from Population-Scale Sequencing.” Nature 476 (7319): 1061–1073.Find this resource:
AAP (American Academy of Pediatrics) Committee on Bioethics, Committee on Genetics, and ACMG (American College of Medical Genetics and Genomics) Social, Ethical, and Legal Issues Committee. 2013. “Ethical and Policy Issues in Genetic Testing and Screening of Children.” Pediatrics 131(3): 620–622.Find this resource:
AAP (American Academy of Pediatrics)/HRSA (Health Resources and Services Administration) Newborn Screening Task Force. 2000. “Serving the Family from Birth to the Medical (p. 662) Home: Newborn Screening: a Blueprint for the Future—A Call for a National Agenda on State Newborn Screening Programs.” Pediatrics 106(2, pt. 2): 389–422.Find this resource:
ACMG (American College of Medical Genetics and Genomics). 2006. “Newborn Screening: Toward a Uniform Screening Panel and System.” Genetics in Medicine 8: 1S–252S.Find this resource:
Alexander, D., and Van Dyck, P. 2006. “A Vision of the Future of Newborn Screening.” Pediatrics 117(5, pt. 2): S350–354.Find this resource:
Botkin, J. R., Anderson, R., and Rothwell, E. 2012. “Expanded Newborn Screening: Contemporary Challenges to the Parens Patriae Doctrine and the Use of Public Resources.” In Medicine and Social Justice: Essays on the Distribution of Health Care, edited by R. Rhodes, M. P. Battin, and A. Silvers (Oxford: Oxford University Press).Find this resource:
Botkin, J. R., Belmont, J. W., Berg, J. S., Berkman, B. E., Bombard, Y., Holm, I. A., et al. 2015. “ASHG Statement: Points to Consider: Ethical, Legal, and Psychosocial Implications of Genetic Testing in Children and Adolescents.” American Journal of Human Genetics 97(1): 6–21.Find this resource:
Botkin, J. R., Clayton, E. W., Fost, N. C., Burke, W., Murray, T. H., Baily, M. A., et al. 2006. “Newborn Screening Technology: Proceed with Caution.” Pediatrics 117(5): 1793–1799.Find this resource:
Botkin, J. R., and Rothwell, E. 2016. “Whole Genome Sequencing and Newborn Screening.” Current Genetic Medicine Reports 4(1): 1–6.Find this resource:
CDC (US Centers for Disease Control and Prevention). 2012. “CDC Grand Rounds: Newborn Screening and Improved Outcomes.” MMWR 61(21): 390–393.Find this resource:
Crowe, S. 2008. “A Brief History of Newborn Screening in the United States.” Staff Discussion Paper, President’s Council on Bioethics. http://bioethics.georgetown.edu/pcbe/background/newborn_screening_crowe.html.
Green, R. C., Berg, J. S., Grody, W. W., Kalia, S. S., Korf, B. R., Martin, C. L., et al. 2013. “ACMG Recommendations for Reporting of Incidental Findings in Clinical Exome and Genome Sequencing.” Genetics in Medicine 15(7): 565–574.Find this resource:
Guthrie, R. 1992. “The Origin of Newborn Screening.” Screening 1: 5–15.Find this resource:
Howell, R. R. 2006. “We Need Expanded Newborn Screening.” Pediatrics 1117(5): 1800–1805.Find this resource:
Johnston, J., Lantos, J. D., Goldenberg, A., Chen, F., Parens, E., Koenig, B. A., members of the NSIGHT Ethics and Policy Advisory Board. 2018. “Sequencing Newborns: A Call for Nuanced Use of Genomic Technologies.” The Hastings Center Report 14 Aug. 2018. https://doi.org/10.1002/hast.874.Find this resource:
Kemper, A. R., Knapp, A. A., Green, N. S., Comeau, A. M., Metterville, D. R., and Perrin, J. M. 2010. “Weighing the Evidence for Newborn Screening for Early-Infantile Krabbe Disease.” Genetics in Medicine 12(9): 539–543.Find this resource:
Lewis, M. H. 2014. “Newborn Screening Controversy: Past, Present, and Future.” JAMA Pediatrics 168(3): 199–200.Find this resource:
Lewis, M. H. 2015. “Lessons from the Residual Newborn Screening Dried Blood Spot Litigation.” Journal of Law Medicine and Ethics S43(1): 32–35.Find this resource:
NIH (National Institutes of Health). 2012. Genomic Sequencing and Newborn Screening Disorders Request for Applications (Bethesda, Md.: NIH). https://grants.nih.gov/grants/guide/rfa-files/RFA-HD-13-010.html.Find this resource:
Olney, R. S., Moore, C. A., Ojodu, J. A., Lindegren, M. L., and Hannon, W. H. 2006. “Storage and Use of Residual Dried Blood Spots from State Newborn Screening Programs.” Journal of Pediatrics 148(5): 618–622.Find this resource:
Paul, D. 1999. “Contesting Consent: The Challenge to Compulsory Neonatal Screening for PKU.” Perspectives in Biology and Medicine 42(2): 207–219.Find this resource:
(p. 663) Pique, L., Graham, S., Pearl, M., Kharrazi, M., and Schrijver, I. 2017. “Cystic Fibrosis Newborn Screening Programs: Implications of the CFTR Variant Spectrum in Nonwhite Patients.” Genetics in Medicine 19: 36–44.Find this resource:
President’s Council on Bioethics. 2008. The Changing Moral Focus of Newborn Screening (Washington, D.C.: President’s Council on Bioethics). https://bioethicsarchive.georgetown.edu/pcbe/reports/newborn_screening/.Find this resource:
Ross, L. F. 2011. “Mandatory versus Voluntary Consent for Newborn Screening?” Kennedy Institute of Ethics Journal 20(4): 299–328.Find this resource:
Sebelius, K. 2010. “Letter to R. Rodney Howell, M.D.” https://www.hrsa.gov/sites/default/files/hrsa/advisory-committees/heritable-disorders/reports-recommendations/response-rusp-2010052.pdf.