Click here for a deeper conversation on this topic led by Hastings Director of Research Josephine Johnston.
Looking for the Psychosocial Impacts of Genomic Information
Monday, February 26 and Tuesday, February 27, 2018
For the last quarter century, researchers have been asking whether genomic information might have negative psychosocial effects. Anxiety, depression, disrupted relationships, and heightened stigmatization have all been posited as possible outcomes—but not consistently found. At this conference, we will ask what accounts for the discrepancy between these hypothesized outcomes and the effects that have been documented in empirical studies. Are we asking the right questions? Using the right tools? Looking in the right places? Or was the expectation of large, negative psychosocial impacts of genomic information overblown to begin with? Either way, where does research into the ethical and psychosocial implications of genomic medicine go from here?
With the rise of next-generation genome sequencing technologies and precision medicine initiatives, there is today, perhaps more than ever before, increased enthusiasm for using genomic information to help diagnose, predict, treat, and prevent human health problems. Alongside these hopes, however, concerns about the negative psychosocial effects of genomic information continue to linger, complicating the process of integrating new genomic tools into health care. Since the inception of large scale human genome research, scholars, professional associations and institutional regulators have cautioned about the risks of exacerbating a number of socially dangerous attitudes that have been historically tied to human genetics: “genetic essentialism,” “genetic determinism,” “genetic stigmatization,” “genetic discrimination” and the impact thereof for exacerbating perceptions of clinical reductionism, racism, and fatalism. In this presentation, we provide a working lexicon of all these “isms,” as a starting point for considering what emerging empirical studies say about their actual salience in applied genomics contexts. In the first part, we will offer an annotated taxonomy of the main worries that have informed our discourse to date, to flesh out what psychosocial phenomena empirical investigators might have expected the clinical translation of genomic research to provoke. In the second part, we ask where these worries came from in the first place. Through a historical discussion of analogous concerns in other medical fields, we explore why these concerns seemed to provide particularly important questions to address in assessing the impact of new genomic research.
“The Person on the Street Will Utilize Information about Their Genes in Line with Their Existing Beliefs and Preferences Rather Than Being Substantially Swayed in New Negative Directions” (10:30-11:30).
Both the strong optimistic hopes and extreme pessimistic fears about the impact that “genetic information” would have on personal and social health and wellbeing have not (yet) come to pass according to the cumulative research data. Fearful and hopeful predictions alike were founded on a model that envisioned people as relatively ignorant, simple-minded, and slavish receptors of expert or media messages about genetics. The many empirical research findings to date are instead consistent with a model of human beings as strategic users of whatever information they receive, whether it is genetic or social or anything else; i.e., they use whatever information they get to advance whatever ends that they already have. I will also review evidence which shows that, even in the face of genetically determinist messages, “lay people” are perfectly capable of maintaining complex multi-component models of causation; i.e., among the causes that lay people can entertain at once are heredity, behavior, individual choice, supernatural forces, chance, and other forces. The apparent genetic determinism and heightened discriminatory effect that appears in studies that prime genetic causation results from the situational selection of one of the available causal accounting components to achieve particular goals or to respond to particular conditions. These short-term activations are fully compatible with activations of other causal accounts simultaneously or subsequently. The strategic deployment of genetic determinism may nonetheless have social consequences of concern if genetic determinism serves the interests of the powerful. However, exploring such concerns would require methodologically daunting social-psychological rather than psycho-social or even merely sociological modes of study. In any case, efforts to improve health or forestall discrimination will be unlikely to achieve their objectives if they do not account for the robust nature of human beings as active observers and relatively complex thinkers who prioritize their own various goals over the uniformity or simplicity in causal accounting ostensibly valued by disciplinarily oriented researchers.
Session 2: Results from Empirical Explorations of the Impacts of Sharing Genomic Information with Individuals (Children, Adolescents, and Adults).
“Evaluating the Psychosocial Impacts of Providing Genetic and Genomic Health Risk Information: A Review of Systematic Reviews” (11:30-12:30).
The articulation of the original ELSI concerns in the 1990s motivated extensive empirical research over the following decades. My review aims to identify the overarching findings, gaps, and opportunities that have been gleaned from this expansive literature by examining existing systematic reviews, meta-syntheses, and meta-analyses. To accomplish this, I conducted a systematic literature search across three databases to identify publications that met the criteria of: being published since 1990, having performed a systematic literature review, and examining a psychosocial element in response to receipt of genetic or genomic information. I extracted information addressing interpretations relevant to the study aims. Based upon these findings, I will provide an overview of the current state of the evidence for psychosocial impacts of genomic information and address themes that are pertinent to future research in this area.
“Psychosocial Effects of Multigene Panel Testing for High and Moderate Penetrance Gene Variants in the Context of Cancer Genomics” (1:30-2:30).
I will begin with a brief overview of shifting trends in clinical cancer genetic testing. I will discuss the traditional model of serial testing for high penetrance cancer susceptibility genes (e.g., BRCA1/2), and will summarize past findings regarding the psychosocial effects of testing for high penetrance gene variants. I will also describe the increasing clinical adoption of cancer susceptibility multigene panel testing, along with its implications for greater availability of risk information about moderate penetrance genes (e.g., ATM), unexpected findings in high penetrance genes, and the identification of variants of uncertain clinical significance. I will review the current, limited literature on psychosocial effects of receiving such results from multigene panel testing. The presentation will conclude with a discussion of under-explored issues that arise from the use of multigene panel testing in cancer genomics: sources of psychosocial effects (e.g., uncertainty), additional negative psychosocial effects (e.g., medical mistrust, communication challenges among patients, families, and healthcare providers), and potential positive psychosocial effects (e.g., patient empowerment).
Session 3: Results from Empirical Explorations of the Impacts of Sharing Genomic Information in the Perinatal Context (regarding Fetuses and Newborns)
“The Impacts of Genomic Information on Parent-Child Bonding: Chasing the Perfect Pregnancy” (2:30-3:30).
The American College of Obstetrics and Gynecology considers prenatal chromosomal microarray analysis (CMA) an appropriate first tier test following detection of an ultrasound anomaly. The increased diagnostic yield of CMA is, however, accompanied by the possibility of finding copy number variants (CNVs) of uncertain clinical significance, incomplete penetrance, or variable expressivity, with phenotypes ranging from apparently normal to severely affected. Carrying a pregnancy to term without understanding or knowledge of the full implications of an uncertain CNV may come at great emotional and financial cost to the parent and to the child. For some families, education may mitigate these outcomes in the near term, but the long-term consequences remain unclear. In this presentation I will report on data derived from a longitudinal study of families in which a fetus was diagnosed prenatally with a potentially pathogenic CNV. These data suggest uncertain findings may incite parental distress that cannot be completely assuaged given the current limits on what is known. Without any symptoms other than a positive test result, physicians lack the diagnostic schema to permit clear conceptualization and action. Consequently, families remain in a state of prolonged uncertainty, and children may be perceived as vulnerable in ways that are unfounded yet highly impactful. The potential for over medicalization [the process of defining human experience as warranting medical evaluation, diagnosis, and/or treatment] of the child’s development heightens the risk of parental, and potentially of the physician’s, perception that the child is vulnerable, needs ongoing evaluation or resources, and may come to shape the child and family’s social experiences or identity.
“Pregnant Women Manage the Delivery of News about Their Fetus Better Than We May Assume” (4:00-5:00)
Early NHGRI ELSI-funded research explored the complexities of novel prenatal testing choices facing women/parents. In 1994, Karen Rothenberg and Elizabeth Thompson published an edited volume of essays (Women and Prenatal Testing: Facing the Challenges of Genetic Technology) that considered how prenatal testing might adversely affect the experience of being pregnant. In 2000, Erik Parens and Adrienne Asch published an edited volume of essays (Prenatal Testing and Disability Rights) that considered the implied messaging of prenatal testing, and its relation to the value/regard for those living with differences or disabilities. While tensions remain about the implications of prenatal testing and reproductive choice, decades of studies have failed to demonstrate significant negative psychological consequences for women and their partners. Most parents with access to services prefer to have a choice to learn genetic information about their developing baby prior to birth. They capably manage receipt of difficult information and make good decisions for themselves, although uncertain information can impede this process. Parents most traumatized by testing are those whose fetus is found to be affected with a condition and face an agonizing decision whether to continue the pregnancy. Yet even these parents recover from their loss, or adapt to raising an affected child. While prenatal screening and testing offer complex choices that lead to challenging decisions, evidence suggests that most mothers/parents cope with the stress that accompanies these valued options.
People are essentialist thinkers, meaning that they have a strong intuition that the natural world emerges as it does due to some invisible, internal, and immutable forces. Lay people’s understandings of genetics shares sufficient parallels with essentialist thinking meaning that people tend to think of genetic causes in similar ways to how they think about essences. A number of social psychological experiments will be discussed to show how people think about genetic causes in ways that are different from environmental causes, particularly in domains of moral responsibility. Furthermore, these essentialist ways of thinking about genetics are rather resistant to efforts to reduce them. For example, people fail to adequately calibrate for the strength of genetic causes. However, complex discussions of gene-environment interactions appear to weaken people’s essentialist responses.
“Bad Acts and Bad Genes: Effects on Decisions to Punish” (9:30-10:30)
Behavioral genetics, once largely the preserve of scientists exploring the relative influences of heredity and environment on behavioral traits, is now an increasingly frequent visitor in the courts. The interest of the criminal defense bar in the genetic roots of behavior lies in the presumed effect of genetic explanations on perceptions of individual responsibility. If a defendant’s criminal behavior, rather than being determined by conscious choices, were driven by unconscious genetic predispositions to commit antisocial acts, the person may seem less responsible for the outcome and therefore less deserving of punishment. Behavioral genetics, at least in principle, thus has become a tool for legal claims of reduced culpability and mitigated punishment. However, such effects on determinations of guilt and sentencing have been difficult to detect in actual cases — with rare exceptions — and are modest or entirely absent in the experimental data. I will explore some of those data, consider explanations for these findings, and speculate on the future of behavioral genetic evidence in criminal court.
“Implications of Genetic Explanations for Thinking about Mental Disorders” (11:00-12:00).
Mental disorders are increasingly thought of as biomedical illnesses with genetic roots. A considerable body of research has examined the effects of explanations that attribute mental disorders to biological causes, including genetic etiologies, and has linked biomedical explanations to feelings of hopelessness about the prospect of recovery, as well as to reductions in therapists’ compassion and warmth toward patients. More recent work has focused specifically on the consequences of telling people that they are genetically predisposed to mental disorders. This research has suggested that receiving such feedback can lead people to doubt their ability to cope with psychiatric symptoms and even affect people’s recall of their own experiences of such symptoms. Additionally, while the primary benefit of genetic explanations for psychopathology is that they reduce the extent to which people are blamed for their own symptoms, evidence suggests that this may be a “double-edged sword,” with reductions in perceived blameworthiness accompanied by reductions in the degree of self-control and agency ascribed to patients.
Session 5. Broad Reflections on Quantitative and Qualitative Approaches to Understanding Psychosocial Implications.
“Qualitative Research: Robust but Marginalized” (1:00-2:00).
In this presentation, I will explore the assumptions that drive the studies that tend to discover small psychosocial impacts of genomic information – especially as those assumptions have played out in the context of studying the impacts of newborn screening (NBS). What are some of the distinctions between quantitative and qualitative NBS findings? Can qualitative analyses help us understand varied responses to genomic information that may average out in quantitative studies? Should specific kinds of long-term effects be a “gold standard” in our assessment about impact, or are there ways to take other kinds of impact into account? How do we define “consistent findings,” and what is the relationship between frequency of a given finding in the published literature and its value for understanding a given phenomenon? In what ways has a growing reliance on meta-synthesis shaped both our assumptions about the worth of qualitative studies, and the methods we use to describe the insights they generate?
“Assessing the Impact of Genetic Susceptibility Testing: Where Do We Go from Here?” (2:00-3:00).
I will offer a review and critique of traditional approaches to measuring the psychological and behavioral effects of genetic susceptibility testing. I will highlight the limitations of commonly used measures in this area; these include a lack of sensitivity in detecting effects, an overreliance on psychiatric symptom inventories, and a relative lack of attention to the benefits of receiving genetic information. I will suggest some directions for future research in the field, including improved contextualization of measures, increased use of mixed-methods approaches, and recruitment of more diverse participant populations. Throughout the presentation, I will draw on my experience in examining the psychological and behavioral effects of genetic testing in areas including Alzheimer’s disease, cancer genomics, and direct-to-consumer genetic testing.
Final Session: Reflections on What Has Been Said at the Conference (3:15-4:00)