By Katharine Press Callahan, MD, MSME, CHOP Attending Neonatologist and Ethicist 

It’s in your DNA. 

This phrase is familiar, but what exactly is “it” that is in DNA?  

Remarkable progress in genetic technology has enabled doctors to search genomes for changes more broadly and more quickly. This is appealing in the neonatal intensive care unit (NICU), where there is a high prevalence of genetic disease and a need for greater prognostic information. The speed of modern genetic testing means results are available on the scale of days or weeks and can potentially influence ICU decision making. For a handful of genetic diseases, targeted genetic therapies exist.   

My work, however, stems from a fear that though we can increasingly find “it,” we don’t always know what it means, nor what to do with it. A genetic test result for an ill neonate may be poorly correlated with the questions it was sent to answer or have uncertain implications for a child’s current care or future life. These issues exist in part because our current understanding of genetic science is incomplete. Additionally, genetic tests frequently produce information about future health or development without a clear next step. Measuring the benefits and harms of information itself is not straightforward. The value of such information rests largely on perceptions of those receiving it, namely parents and clinicians.   

Challenges measuring the benefit of genetic information  

Utility is the likelihood that a test will result in an improved outcome. The utility of genomic testing cannot be tied to a single health outcome since genetic tests yield varied results with even more varied effects on care. Length of stay or quality-adjusted life years are unifying outcomes applied to other medical interventions that produced varied results. Such measures are not easy to apply to genetic tests, for which the aim is not always reducing length of stay or prolonging life. If a genetic test result precipitates redirection of care, this shortens life but is frequently considered a benefit, for instance. As a consequence, intermediate measures such as ‘diagnostic yield’ and process measures such as ‘changes in management’ are often used as proxies for utility.  

I conducted a systematic review of how studies of genomic medicine in the NICU assess and report utility. While studies tend to highlight cases where therapies are changed, these cases are rare, and even in such cases the expected degree of benefit is typically unclear. Changing an antiepileptic medication is more valuable if seizures are permanently stopped than if seizures continue, but both results would be counted as equivalent in current metrics. At the same time, researchers typically overlook more common benefits, such as the ‘personal utility’ of a parent being able to able to join a disease-specific support group or a clinician feeling more confident they are not missing another diagnosis.   

Working with the Baby Eagle project, I am designing new measures of utility, and already it is clear that changing the metric changes the results. Clinicians and researchers need to be clear about what the benefits of genetic testing are, particularly as they are weighted against potential harms.   

Potential harms of genetic information

Genetic information that is uncertain or informs prognosis — particularly disability — poses ‘informational hazards.’ Such information may be misunderstood, used in ethically questionable ways, or may interrupt nascent relationships between families and new babies. I began studying such hazards among neonatologists. Using a hypothetical case study with a randomized design, neonatologists were randomized to see one of two versions of a patient case, which varied only on whether they included or omitted certain genetic information.  

My findings were concerning. Neonatologists were more likely to recommend palliative care over intensive care for patients with uncertain genetic information and patients with genetic diagnoses that portended future developmental disability but did not affect short-term prognosis. If these biases extend to clinical practice, they represent unacknowledged harms of genetic information.   

My current work explores parents’ ambivalence about genetic information. The most unequivocal evidence of perceived harms is that a substantial proportion of parents decline recommended genetic tests. While parents are generally committed to pursuing any medical technologies that may improve their child’s outcome, they are sometimes skeptical about whether genetic tests will address problems they care about. They also worry that tests may produce information they do not want — bad news without a solution or information that could be used in discriminatory ways. Ultimately, improving our understanding of both benefits and harms will be essential in truly personalizing ‘personalized medicine.’  

Is genetic information exceptional?

The phrase, “It’s in your DNA,” reflects a widely held belief that genetic information is ‘the truth.’ In many practical ways, the role of genetic information is not dissimilar from other medical tests and procedures in the NICU. Brain imaging can also provide information about the future, and about disability, often with uncertainty, for example.  

Yet, somehow, genetic information seems to feel different. Perhaps this is because genetics provides the link between biologic families. Perhaps it is because genetics relates (albeit in complicated ways) to sex, race and other elements of our identities. Perhaps it is because, historically, diagnosed genetic conditions tended to be severe. Whatever the reason, this information seems to have particular significance for families in the NICU, adding weight to clinicians’ management of benefits and harm.  

Contact

callahankp@chop.edu 

Related articles  

1. Callahan KP, Flibotte J, Skraban C, et al. Influence of Genetic Information on Neonatologists’ Decisions: A Psychological Experiment. Pediatrics. Published online February 16, 2022:e2021052130. doi:10.1542/peds.2021-052130 
2. Deem MJ. Whole-genome sequencing and disability in the NICU: Exploring practical and ethical challenges. Pediatrics. 2016;137:S47-S55. doi:10.1542/peds.2015-3731I 
3. Callahan KP, Mueller R, Flibotte J, Largent EA, Feudtner C. Measures of Utility Among Studies of Genomic Medicine for Critically Ill Infants: A Systematic Review. JAMA Netw Open. 2022;5(8):e2225980. doi:10.1001/jamanetworkopen.2022.25980 
4. Callahan KP. Discarding Information. JAMA. 2023;330(16):1527-1528. doi:10.1001/jama.2023.19005