APPENDIX C

Protecting the Health Services Research Data of Minors

Ross A. Thompson, Ph.D.

There are many issues relevant to the confidentiality, security, and privacy of personally identifiable health information (PHI) used for health services research purposes. These include (1) the nature of the confidentiality protections of privately funded health services research studies (which are not generally subject to institutional review board [IRB] approval), (2) the purposes for which the data were originally gathered, (3) the purposes for which they are used in secondary reanalysis, (4) the nature of the consent procedures originally used and the confidentiality assurances that are part of the consent process, (5) protections of the data while the study is under way, and (6) who the relevant actors and agencies fundamental to such procedures are (e.g., hospital IRBs, insurance companies managed care providers, physicians researchers).

These are difficult issues that are relevant to the PHI of all research participants, regardless of developmental stage. For example, problems of confidentiality emerge in very large databases when health events of extremely low frequency are studied and data relevant to these particular cases can easy lead to their identification. The steps that can be taken to ensure the confidentiality of health data such as these are unlikely to vary significantly depending on whether the health events beset adults or infants (e.g., a multiply challenged baby in the neonatal intensive care unit).

However, when children are research participants, there may be more unique risks to the confidentiality and privacy of their PHI and special concerns in the secondary analysis of their health data. This paper is devoted to framing the issues associated with protecting the health services research data of minors



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 159
Protecting Data Privacy in Health Services Research APPENDIX C Protecting the Health Services Research Data of Minors Ross A. Thompson, Ph.D. There are many issues relevant to the confidentiality, security, and privacy of personally identifiable health information (PHI) used for health services research purposes. These include (1) the nature of the confidentiality protections of privately funded health services research studies (which are not generally subject to institutional review board [IRB] approval), (2) the purposes for which the data were originally gathered, (3) the purposes for which they are used in secondary reanalysis, (4) the nature of the consent procedures originally used and the confidentiality assurances that are part of the consent process, (5) protections of the data while the study is under way, and (6) who the relevant actors and agencies fundamental to such procedures are (e.g., hospital IRBs, insurance companies managed care providers, physicians researchers). These are difficult issues that are relevant to the PHI of all research participants, regardless of developmental stage. For example, problems of confidentiality emerge in very large databases when health events of extremely low frequency are studied and data relevant to these particular cases can easy lead to their identification. The steps that can be taken to ensure the confidentiality of health data such as these are unlikely to vary significantly depending on whether the health events beset adults or infants (e.g., a multiply challenged baby in the neonatal intensive care unit). However, when children are research participants, there may be more unique risks to the confidentiality and privacy of their PHI and special concerns in the secondary analysis of their health data. This paper is devoted to framing the issues associated with protecting the health services research data of minors

OCR for page 159
Protecting Data Privacy in Health Services Research in light of how the PHI of minors is commonly treated in health services research. In brief summary, these include the following issues: Children constitute a heterogeneous population. Within the broad population of legally defined “minors,” research studies document considerable variability in developing judgment, self-understanding, and psychosocial functioning. This means that: the capacities of children and youth to assent to the uses of their PHI and to meaningfully understand assurances concerning confidentiality, privacy, and research risk develop considerably with increasing age, with adultlike capabilities evident early in adolescence; the psychological risks posed by the inappropriate disclosure of minors' PHI (e.g., perceived privacy violations, feelings of embarrassment or humiliation, threats to their medical or legal interests) change complexly with development but, by late childhood, approximate the risks experienced by adults; collected PHI, older children and adolescents may be concerned about access to health data collected when they were younger and have a strong interest in giving independent consent to its use. These considerations are relevant to all minors, especially to adolescent populations who can in some jurisdictions consent independently of their parents to certain forms of medical treatment (e.g., substance abuse or mental health treatment). Children constitute a uniquely vulnerable population because of their limited rights under the law and their limited capacities for autonomous decision making. This means that special provisions are needed to ensure their protection from research risks, which include, but extend beyond, parental proxy consent on their behalf. Problems in proxy consent arise from (1) fundamental difficulties in distinguishing the interests of children from those of their parents or other custodians, (2) the fact that consent also involves accepting provisions for control over research materials, knowledge of research findings, and conditions governing children's elective withdrawal from research participation in which parents' and children's interests may also differ, and (3) lack of clarity about whether proxy consent endures for the entire course of the research investigation, including longitudinal or secondary analyses, regardless of children's developing capacities (and interests) in asserting and protecting their own rights as research participants as they mature. Special considerations in biomedical data result from the uniquely sensitive nature of such data, and the potential immediate and longer-term implications of PHI for children in the context of family dynamics. These require im-

OCR for page 159
Protecting Data Privacy in Health Services Research mediate determinations about who has access to, and control over, children's PHI that take into consideration life-course concerns for children. Taken together, it is clear that the health information of minors should be considered very differently from the PHI of adults in a manner that reflects their developing capabilities and a life-span consideration of children's interests. A CASE ILLUSTRATION These issues related to protection of the research data of minors are not exclusive to PHI. To illustrate, consider the following hypothetical case study (based on actual published research): A research team inaugurates a large, longitudinal study of children 's social and emotional adjustment in the 1970s. This research enlists a representative sample of more than 2,000 children in the early gradeschool years from a large urban population. Passive consent procedures are used (i.e., parents receive a letter describing the nature of the study and are asked to contact the investigators if they wish their child NOT to participate). The children are assessed annually on six occasions after the study begins, with assessments including measures of peer relationships, self-perceptions, academic competence, and emotional functioning. Fifteen years after the study begins, a new researcher joins the team with interests in the prediction of child maltreatment. After obtaining IRB approval, the names and other identifying information of children in the original research are matched against the state 's child abuse registry, after permission to do so is obtained from state officials. This results in a large sample of children identified as having been abused or neglected when the original research was in progress. Furthermore, the social service agencies for counties in the area are contacted and invited to participate in the research by sharing the child protective services case records of the identified children in the sample. Based on these records, the timing, type, chronicity, and severity of maltreatment is determined. Careful efforts are taken to ensure the confidentiality of all research materials. As a consequence, a uniquely informative investigation of the antecedents and correlates of child maltreatment using a prospective longitudinal design with a large, representative sample results. Matched subsamples of maltreated and nonmaltreated children are compared to address fundamental questions about the impact of the experience of abuse or neglect, and its timing and severity, on various measures of psychosocial and emotional adjustment in childhood. A number of questions are raised by this case illustration that are also relevant to the protection of health services research data by minors: Given the sensitive nature of the secondary analyses (and potential perceptions of privacy invasion), were the researchers ethically obligated to seek permis-

OCR for page 159
Protecting Data Privacy in Health Services Research sion from families in the study when the investigation later turned to the prediction of child maltreatment? Or did the original consent procedures apply to the secondary data analysis that also involved accessing confidential state records? Did the initial use of parental passive consent procedures—rather than active consent, in which the parent must contact the researchers to consent volitionally after having been informed about the research purposes—alter the ethical obligation to seek further consent after the research team began to investigate more sensitive issues? Or were original consent procedures adequate regardless of whether passive or active consent was used? Because a significant proportion of the children originally enlisted into the study had reached the age of majority before the secondary analyses began, were the researchers ethically obligated to contact them for permission to use data gathered during their minority? Or were the original consent procedures, involving parents but not children, still sufficient at this later period? Without opportunities for further informed consent, was it possible for the participant families or children (now, in some cases, young adults), to protect themselves against the potential risks involved in the secondary analyses of these data in concert with protected (and confidential) state records? ISSUES TO CONSIDER There are no easy answers to these questions, but posing them thoughtfully is essential to consider judgments about how the needs and rights of minors can be protected in health services research data that may be subjected to secondary analysis. There are additional considerations, discussed below, that add further complexity to thinking about any potential special ethical review requirements of research involving minors. Children as a Heterogeneous Population Although the term “children” is commonly used in these contexts to refer to all persons below the age of majority, it is instructive to realize that the term encompasses infants, preschoolers, gradeschool children, and adolescents within a single conceptual umbrella. Although the umbrella may be sufficient for legal purposes, it ill-fits the heterogeneity of capabilities, interests, and needs characterizing the population it covers. With respect to the reflective judgment required for informed consent, for example, there is considerable research evidence that by early adolescence, young people are capable of making informed consent decisions about medical treatment and research participation that are comparable in quality to those of adults (Abramovitch et al., 1995; Abramovitch et al., 1991; Lewis et al., 1978; Melton et al., 1983; Ruck et al., 1998; Ruck et al., 1998; Weithorn, 1982, 1983; Weithorn and Campbell, 1982). At somewhat earlier ages (i.e., during the gradeschool years), children's informed consent capabilities are more uneven or

OCR for page 159
Protecting Data Privacy in Health Services Research inconsistent, although their judgment can be strengthened through the use of simple interventions, such as providing short, educational information about rights and prerogatives as a research participant (Abramovitch et al., 1995; Rau and Fisher, 1999). Thus, the capacities of children to assert an active, responsible voice in judgments concerning their research participation increases significantly with age, with mature competence reached well before the legal age of majority, but with the capacity for mature consent strengthened through age-appropriate educational interventions. In a somewhat comparable manner, the psychological risks to which children are vulnerable also change significantly with age. Moreover, these risks change in complex ways: some decrease with increasing age, others increase as children mature, and still others remain essentially stable over the course of development (Thompson, 1990a, b, 1992). This challenges the prevailing assumption that children become less vulnerable in research contexts as they mature. For example, although children become progressively less prone with increasing age to becoming distressed, overwhelmed, or disorganized in research settings because of the development of emotional and behavioral self-regulation and coping skills, children become more susceptible to other risks from their research participation as they mature. These include susceptibility to threats to self-concept and self-esteem (sometimes arising from performance evaluation),; vulnerability to feelings of shame, embarrassment, or humiliation (sometimes from concerns about the improper disclosure of personal information); and concern about expressed or implied social comparison evaluations. Likewise, although children become progressively less prone to being deceived or coerced by research procedures as they develop more mature and insightful judgment about the motives of other people, they also become more vulnerable to concern about perceived privacy violations in the use of their research data. Indeed, for many adolescents, concerns with personal privacy extend to the disclosure of personal information, such as their research data, even to parents (see Wolfe, 1978). There are several implications of this developmental analysis of research risk: The psychological risks associated with research participation and deriving from the imroper disclosure of research data vary as minors mature, but do not necessarily decrease linearly with increasing age. For some risks, vulnerability increases. Moreover, children 's vulnerability to the research risks most pertinent to adults increases sharply over time; these include perceived violations of privacy, the embarrassment and humiliation that may derive from inappropriate disclosure of personal health data, and some of the tangible consequences of unwarranted disclosure (e.g., difficult family processes or compromised medical or legal circumstances; see below). Estimating the nature of the research risks to which children are vulnerable is thus a developmentally graded assessment, and in longitudinal research these risks may change as children mature over the course of the investigation. In other words, the research risks relevant to an investigation that was inaugurated when children were preschoolers are not necessarily the same as those

OCR for page 159
Protecting Data Privacy in Health Services Research relevant by the time the same children have become adolescents. This means that a revised risk–benefit calculus becomes necessary as longitudinal research proceeds (Thompson, 1992). In the secondary analysis of data (as with longitudinal data), the constellation of risks to which children are vulnerable is also likely to change with the passage of time since the original data were gathered. In addition, older children and adolescents, like young adults, are likely to feel much differently about the uses of data gathered when they were very young than they were capable of feeling at the time these data were initially obtained. Concerns over personal privacy, threats of embarrassment or humiliation, and more tangible concerns related to the potentially inappropriate disclosure of research data now become personal issues (whereas formerly they were addressed on behalf of the child through parental proxy consent procedures). Any assessment of the ethical responsibilities of researchers should take these developmentally changing concerns into consideration. Children as a Uniquely Vulnerable Population A longstanding tradition of special protections for children and youth in research derives from the special configuration of child, parental, and state interests related to children's research participation. Within moral theory, the unique characteristics of children account for their limited self-determination and the beneficent paternalism they receive from others (Baumrind, 1978; Melton, 1987). Their limited experience and immature reasoning capabilities together mean that although children (as persons before the law) are entitled to some of the rights of privacy and self-determination granted to adults, their capacities to exercise these rights are limited. They are limited, in part, because of the responsibilities entrusted to others to safeguard their welfare. As a consequence, adults (especially parents) make fundamental decisions concerning the research participation of their offspring, exercising proxy consent on behalf of children and making other decisions concerning research on behalf of their children' s interests. Moreover, the state also assumes a special interest in the child's well-being, independently of the authority of parents, because of its responsibility under the parens patriae doctrine. This is one reason why the ethical review of research protocols involving children by state-appointed agencies (e.g., IRBs) is typically more searching, even though parents also exercise proxy consent on behalf of their offspring. This means that children are almost uniquely powerless social actors in decisions concerning their research participation and the disposition of their research materials. Although children's assent is encouraged by existing federal regulations, it may be difficult for them to dissent meaningfully from research participation not only because of limitations in judgment, but also because their invitation to participate typically occurs in a context of prior parental permission, institutional support (whether the institution is a school, childcare center,

OCR for page 159
Protecting Data Privacy in Health Services Research hospital, or other setting), and adults' interests in furthering the research enterprise (Abramovitch et al., 1991). For the same reasons, children's social power to resist research procedures that they find unduly distressing, psychologically invasive, or coercive may also be quite limited. Furthermore, children experience limited social power not only over decisions concerning their research participation but also over other elements of the research process, such as the disclosure of research data and assurances concerning the disposition of research materials, their withdrawal from research participation after the study has begun, and obtaining the benefits (if any) of research participation. Each of these ordinary prerogatives of the research participant is exercised instead by adults as proxies for the child. Of course, in most cases, adults (especially parents) make decisions in the interests of children. However, it is unwise to assume that the interests of parents and offspring are always identical in these situations and that the motives underlying parental consent are always consistent with children's interests. Recognizing this, the federal regulations governing research with children (Department of Health and Human Services [DHHS], 1991) not only require special review considerations in studies involving children, but also encourage the child's assent to research participation when children are capable of doing so meaningfully. (Indeed, in 1978 the National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research recommended that a child's objection to research participation constitute a binding restriction except in extraordinary circumstances and that the assent of children age 7 and older be required for their research participation. These provisions were not, however, incorporated into the final DHHS regulations.) These provisions seem to reflect the following: (1) children's unique needs and social vulnerability mandate special consideration in the ethical review of research, (2) adults (usually parents) must exercise proxy consent on behalf of children, but (3) proxy consent alone cannot be the only assurance that children are not subject to unreasonable risk. These provisions raise further the question of whether, when data gathered on an earlier occasion are enlisted later into new research purposes (e.g., secondary or longitudinal analyses), children should be capable of making their own, independent decisions concerning access to and the disposition of their research materials if they have developed sufficiently mature judgment to do so. Problems in Proxy Consent Just as there is value in recognizing that parents usually make thoughtful judgments concerning the research participation of their offspring, there is value also in recognizing the circumstances in which proxy consent does not necessarily protect children's interests. When parents derive financial benefit, access to services, or other personal rewards from the research participation of offspring for example, proxy consent may not adequately protect children's interests. (Indeed, philosopher Paul Ramsay, 1970, 1976, 1977 has argued that any nontherapeutic research with children is morally impermissible because even

OCR for page 159
Protecting Data Privacy in Health Services Research proxy consent inevitably confounds adults' interests with those of children.) In abusive or adversarial parent –child relationships, adults may be motivated to deny permission if children's research participation might contribute to the detection of maltreatment, substance abuse, or other parental problems. With respect to medical research (e.g., PHI), proxy consent by parents may be problematic in situations when biological assessments of offspring are needed to evaluate the medical condition of another family member (e.g., in genetic screening studies or DNA analysis of tissue samples), or parents may be motivated to conduct risk assessment (presymptomatic) testing of offspring for disorders that are not immediately relevant to the child's well-being (e.g., Huntington's disease). Potential problems in proxy consent are even more apparent when consent is given not by a biological parent, but by a representative of a government agency when children are wards of the state (see National Bioethics Advisory Commission, 1998). In these circumstances, inappropriate incentives may result in considerable pressure for children to participate in research studies that may not be in their best interests. In some situations, adolescents may be hesitant to obtain parental consent for medical treatment of sensitive conditions (e.g., substance abuse or mental health problems) because of privacy concerns. Current legal policies in many states recognize this in allowing adolescents to consent independently to certain forms of treatment without parental consent. This raises important issues concerning the confidentiality of medical records arising from treatment and adolescents' control over the disclosure of this information to family members and others outside the family. Likewise, behavioral researchers have long recognized that parental consent may be an impediment to research participation by older children and adolescents in research studies of sensitive topics (e.g., sexuality, drug or alcohol use) in which parental consent is also likely to violate the privacy interests of youth. In these circumstances, confidentiality and the control over access to research data must also be carefully considered. The potential problems of proxy consent are magnified somewhat by two additional considerations. First, consent also typically includes accepting provisions for and assurance of responsibility for many other aspects of research participation in which parents and children 's interests may not be identical. These include, for example, provisions governing access to the research data gathered from children, issues of privacy and confidentiality of research materials, children's knowledge of the results of the research inquiry, and conditions governing children's elective withdrawal from research participation. In certain situations, the decisions of parents may be influenced by factors different from those relevant to children's needs and interests. Second, consent is often assumed to endure for the duration of the research investigation. As noted earlier, this may be problematic in longitudinal studies or in the secondary analysis of original data that occurs at which children have matured to an age when their own independent judgments concerning the disposition of their research data are relevant. Children's independent consent is im-

OCR for page 159
Protecting Data Privacy in Health Services Research portant not only to respect their increasing maturity concerning judgments about research participation (and their rights as research participants), but to enable them independently to protect their prerogatives as research participants by participating in a consent process in which provisions about confidentiality and privacy, risks and benefits, freedom to withdraw and access to research findings are discussed. In light of these considerations, it becomes apparent also why many research scientists believe that passive consent procedures are inappropriate to research involving children (e.g., Fisher, 1993). In passive consent, parents receive notice of a research study in which their children will participate unless the parent contacts the investigator to prohibit participation. By contrast, active consent procedures require the parent's volitional consent before children can participate in research. The fundamental problem with passive consent procedures is the assumption that parents who do not respond have assented to research participation when, in fact, they may have failed to receive relevant information, it may have been lost or misplaced after it was received, or parents may have otherwise neglected to indicate their objections to their child's research involvement. Passive consent is also problematic because entailed in the consent process is the acceptance of many other provisions governing research participation (e.g., provisions concerning disclosure, confidentiality, and protection of research data; conditions governing children's withdrawal from research participation; assurances concerning risks and benefits from research involvement; and guarantees concerning penalties owing to the failure to participate in research) for which active, not passive, consent is necessary. Active consent is especially important in situations where parents are acting as proxies in protecting their children's needs and interests. For similar reasons, blanket research consent provisions incorporated into insurance and consent-to-treatment documents that permit PHI (and, quite often, the PHI of family members) to be used for secondary research purposes may be equally inappropriate from the perspective of research ethics. Without clear information concerning the nature of the secondary research purposes and relevant assurances concerning privacy and confidentiality, it is difficult for an individual to know what he or she is consenting to. Moreover, when such blanket consent is requested at the time of medical treatment, these circumstances may make it very difficult to decline. Special Considerations in Biomedical Data The need for parental proxy consent and children's assent, the developmentally graded risks faced by children owing to the inappropriate disclosure of research data, and the growing capacities for personally informed consent that emerge as minors mature are considerations shared by behavioral and biomedical researchers who enlist children into their studies. There are, however, a number of concerns that are more specific to health services research data, such as those used for secondary research purposes, because of the broader implica-

OCR for page 159
Protecting Data Privacy in Health Services Research tions of the data for children and their families (see, generally, National Bioethics Advisory Commission, 1999). These make considerations associated with the control and dissemination of PHI especially germane, particularly when data are collected and/or used by agencies not usually governed by the Common Rule. These issues can be illustrated by considering the genetic testing of children and adolescents. With the recent significant advances in knowledge generated by the Human Genome Project, the challenges posed by the uses and potential misuses of personal genetic information have received increased attention. Genetic testing of children and youth can occur for many reasons, as in genetic screening or pedigree analyses (e.g., to trace the family history of a disease) or in DNA analysis of tissue samples for research or therapeutic purposes. As knowledge of the human genome increases, it is likely that genetic testing of children will increase in the years to come. Whatever the reasons that testing initially occurs, genetic testing may yield information of a sensitive nature for children and adolescents, such as the discovery of an inherited vulnerability, possibly for a stigmatizing (e.g., mental illness, alcoholism) or life-threatening disease (e.g., Huntington's disease, cancer). There are immediate and long-term implications of the knowledge yielded by genetic testing of children and adolescents that are directly relevant to the dissemination and confidentiality of the results of their testing. In the immediate context, knowledge of inherited vulnerabilities can be distressing and confusing, especially given the uncertain, probabilistic prognostic implications of this information (indeed, most adults are unlikely to grasp the indeterminate implications of a genetic marker for an inherited disorder). This requires sensitive considerations of whether children and youth are permitted access to their testing results, conditions under which this information is provided, and the availability of support and guidance to help ensure that youth respond constructively (e.g., by enlisting pathology-preventive behavioral practices, if possible) rather than nonconstructively (e.g., through personally abusive practices or the development of depressive or anxious pathology) to the news that they have an inherited vulnerability (Gardner et al., 1992). This might involve enlisting a fellow health care professional, a trusted extended family member, or an adult friend, to discuss distressing health care findings with an adolescent when immediate family members may not be the most suitable counselors, perhaps because of their own conflicts or interest or owing to family dissension. These considerations lead to a series of important questions. Who else is allowed access to the information yielded by a child's or an adolescent's genetic testing? Who decides if, when, and how the young person is provided with this knowledge? To what extent are the youth's preferences to know—or not know—the results of the test determinative at the time the testing is conducted or in the future? To what extent does a child or adolescent have control over whether this information is shared with family members or with others outside the family? There is value in creative avenues for protecting children while also ensuring that their privacy and self-determination are safeguarded. One research

OCR for page 159
Protecting Data Privacy in Health Services Research team, for example, protected adolescents' “right not to know” the results of genetic test for Huntington's disease, but also put the testing results in a registry to which youth would have direct and exclusive access after age 18 (see Fisher et al., 1996). These questions are important because of their relevance to the potential extended implications of the information yielded by genetic testing. In a long-term context, knowledge of inherited vulnerabilities can have implications for employment, insurability, and health care coverage that may be based on an accurate or inaccurate understanding of the broader implications of markers within the genotype for inherited disorders. This raises similar questions about the control of the information yielded by a young person's genetic testing and the importance of guarding against the risks inherent in the unwarranted (and unwanted) disclosure of personal health information such as this, especially for institutions not typically governed by the Common Rule. Importantly, PHI improperly disseminated during a child's minority can have longer-term implications for that person's well-being after he or she has become an adult. The secondary analysis of health care data adds further questions to these. This arises, in part, because as children mature to majority age, they may be entitled to an active role in decisions concerning the use of their original testing results. Furthermore, the interests that governed initial parental proxy consent to genetic testing may be much different from those governing the young adult's personal concerns about the dissemination of this information. This suggests that procedures governing the uses of PHI obtained from minors must be considered within a life-span context for the individual. The consent processes and assurances that occurred at the time a child was initially tested may not generalize to later conditions in which the adolescent or young adult can better represent her or his own interests and preferences, and may have an interest in doing so. CONCLUSIONS A central challenge in considering the ethics of research with children is the juxtaposition of the interests of children, their parents, and the research community within the context of children's limited decision-making capabilities and our cultural commitment to beneficent paternalism on their behalf (e.g., Koocher and Keith-Spiegel, 1990). When considering how best to protect the health services research data of minors, especially in the context of the secondary analysis of these data, the issues become especially complex because of the changing developmental needs and interests of minors over the course of the investigation, which may be very different at the time of the secondary analysis of data than they were at the time the data were originally gathered (and consent procedures were originally instituted). These problems are encountered, in somewhat different ways, by behavioral researchers who conduct longitudinal studies in which children's involvement in research is maintained through their increasingly active participation in decision-making on their behalf as they mature. In biomedical research most relevant to PHI, however, the challenges are

OCR for page 159
Protecting Data Privacy in Health Services Research additionally complicated because family needs can be confounded with children's needs, the dissemination of children's PHI can create family conflict, health care institutions may or may not be governed by traditional procedures associated with protection from research risks, and the inappropriate disclosure of children's PHI can have lifelong implications for children that affect them well after they have reached the age of majority. With respect to the practices of IRBs in their review of research protocols involving children, several kinds of questions are warranted. Do research procedures undergo a more searching examination when children are research participants, particularly with respect to assessing in a developmentally graded manner the potential risks of their research participation? IRBs assume a special responsibility to children as research participants and cannot assume that parental proxy consent will always safeguard children's interests. Are active parental consent procedures used when children are below the age of majority? Are consent procedures sufficiently specific and clear that parents can make reasonable judgments concerning the nature of the research procedures and the uses of research data? Do research procedures provide for children's assent to research participation independent of parental consent? Are these procedures appropriate to the child's age and conducted in a manner that avoids undue incentives or pressure on the child? Do these procedures attempt to provide information to children, suitable to their understanding of their rights concerning the privacy and confidentiality of the information yielded by research, their freedom to withdraw from participation, and related prerogatives? (Some IRBs require documentation of the proportion of children who do not assent to research procedures and who seek to withdraw from participation once the research has begun.) With older children and adolescents, do research procedures provide for their active participation in the process of research consent, recognizing the maturity of their independent judgment? When research is longitudinal in nature or when secondary analyses of previously collected data are conducted, is a follow-up IRB review needed to determine whether additional consent procedures are required from the children who participated in the original research or from their parents? Are there new dimensions of potential research risk arising from the fact that the children are now older or from changes in research purposes or goals? When sensitive data (e.g., the results of biomedical assessments) are obtained, is thoughtful consideration devoted to questions such as (1) who has access to this information, (2) whether, and under what circumstances, the child or adolescent is informed about testing results; and (3) provisions for the child to obtain direct access to this information, and control over its dissemination, after reaching the age of majority? As a general rule, it is valuable to consider the protection of health services research data for minors within a life-span context, taking into consideration the

OCR for page 159
Protecting Data Privacy in Health Services Research longer-term implications of decisions made early on behalf of the child, as well as the child's own preferences and goals (Fisher, 1997). Procedures that respect children's immediate and long-term interests in the privacy, confidentiality, and protection of their PHI, in the context of respecting their rights to age-appropriate self-determination, are especially warranted. REFERENCES Abramovitch, R., Freedman, J. L., Henry, K., and Van Brunschot, M. ( 1995). Children's capacity to agree to psychological research: Knowledge of risks and benefits and voluntaries. Ethics and Behavior, 5, 25–48. Abramovitch, R., Freedman, J. L., Thoden, K., and Nikolich, C. ( 1991). Children's capacity to consent to participation in psychological research: Empirical findings. Child Development, 62, 1100–1109. Baumrind, D. ( 1978). Reciprocal rights and responsibilities in parent–child relations. Journal of Social Issues, 34, 179–196. Department of Health and Human Services (DHHS) ( 1991). Code of Federal Regulations. Protection of Human Subjects. 45 CFR 46, Subparts A and D. Washington, DC: U.S.Government Printing Office. Fisher, C. B. ( 1993). Integrating science and ethics in research with high risk children and youth. Social Policy Report of the Society for Research in Child Development , 7, 1–27. Fisher, C. B. ( 1997). A relational perspective on ethics-in-science decisionmaking for research with vulnerable populations. IRB: A Review of Human Subjects Research, 19, 1–4. Fisher, C. B., Hoagwood, K., and Jensen, P. S. ( 1996). Casebook on ethical issues in research with children and adolescents with mental disorders. In K. Hoagwood, P. S. Jensen, and C. B. Fisher (Eds.), Ethical Issues in Mental Health Research with Children and Adolescents . Pp. 135–266. Mahwah, NJ: Erlbaum. Gardner, W., Thompson, R. A., and Smith, M. G. ( 1992, April). Genetic Counseling About Risks of Chronic Diseases: Developmental Perspectives on Adolescents' Best Interests and Legal Rights. In E. J. Susman (chair), Ethical Issues and the adolescent: Research, Health Care, and Social Policy. Symposium conducted at the meeting of the Society for Research on Adolescence, New Orleans, LA. Koocher, G. P., and Keith-Spiegel, P. C. ( 1990). Children, Ethics, and the Law. Lincoln, NE: University of Nebraska Press. Lewis, C. E., Lewis, M. A., and Ifekwunigue, M. ( 1978). Informed consent by children and participation in an influenza vaccine trial. American Journal of Public Health, 68, 1079–1082. Melton, G. B. ( 1987). The clashing of symbols: Prelude to child and family policy. American Psychologist, 42, 345–354. Melton, G. B., Koocher, G. P., and Saks, M. J. (Eds.) ( 1983). Children's Competence to Consent. New York: Plenum. National Bioethics Advisory Commission ( 1998). Research Involving Persons with Mental Disorders That May Affect Decisionmaking Capacity. Rockville, MD. National Bioethics Advisory Commission ( 1999). Research Involving Human Biological Materials: Ethical Issues and Policy Guidance. Vol. 1. Report and Recommendations of the National Bioethics Advisory Commission . Rockville, MD. Ramsay, P. ( 1970). The Patient as Person. New Haven: Yale University Press.

OCR for page 159
Protecting Data Privacy in Health Services Research Ramsay, P. ( 1976). The enforcement of morals: Nontherapeutic research on children. Hastings Center Report, August, 21–30. Ramsay, P. ( 1977). Children as research subjects: A reply. Hastings Center Report, April, 40–42. Rau, J.-M. B., and Fisher, C. B. ( 1999). Assessing and Enhancing the Research Consent Capacity of Children and Youth. Unpublished manuscript, Department of Psychology, Fordham University, New York. Ruck, M. D., Abramovitch, R., and Keating, D.P. ( 1998 a). Children's and adolescents' understanding of rights: Balancing nurturance and self-determination . Child Development, 64, 404–417. Ruck, M. D., Keating, D. P., Abramovitch, R., and Koegl, C. J. ( 1998 b). Adolescents' and children's knowledge about rights: Some evidence for how young people view rights in their own lives. Journal of Adolescence, 21, 275–289. Thompson, R. A. ( 1990a). Vulnerability in research: A developmental perspective on research risk. Child Development, 61, 1–16. Thompson, R. A. ( 1990b). Behavioral research involving children: A developmental perspective on risk. IRB: A Review of Human Subjects Research, 12, 1–6. Thompson, R. A. ( 1992). Developmental changes in research risk and benefit: A changing calculus of concerns. B. Stanley and J. E. Sieber (Eds.), Social Research on Children and Adolescents: Ethical Issues. (Pp. 31–64) . Newbury Park, CA: Sage. Weithorn, L. A. ( 1982). Developmental factors and competence to make informed treatment decisions . Pp. 85–100G. B. Melton, Legal Reforms Affecting Child and Youth Services New York: Haworth. Weithorn, L. A. ( 1983). Children's capacities to decide about participation in research. IRB: A Review of Human Subjects Research, 5, 1–5. Weithorn, L. A., and Campbell, S. B. ( 1982). The competency of children and adolescents to make informed consent treatment decisions. Child Development, 53, 1589–1598. Wolfe, M. ( 1978). Childhood and privacy. Pp. 175–222 I. Altman and J. F. Wohlwill (Eds.), Children and the Environment. New York: Plenum.