Introduction

Adopted children are a diverse paediatric population. Many present with complex physical, developmental and/or health-related problems.1–4 For each child entering the looked-after system, and for whom adoption is the agreed plan, an initial health assessment by a medical advisor is a statutory requirement.

There are two circumstances in which assessment may involve clinical genetic assessment. First is when a child has a clinical problem or physical features that suggest the condition may be genetic in origin. Paediatricians and clinical geneticists may agree that a child can be tested to obtain a diagnosis to account for their problem. The second circumstance arises when a looked-after child has a family history of a genetic disorder that is not apparent but which that child might either develop later or transmit to their own children in the future. Such circumstances can be further divided into clinical assessment that may involve carrier or predictive genetic testing.

Carrier testing is usually carried out on unaffected individuals at risk of recessive or sex-linked conditions, such as cystic fibrosis or haemophilia. While there may not be any direct health implications, a positive result will affect future reproductive decisions. Predictive testing applies to autosomal dominant conditions usually with late onset, such as Huntington’s disease. Testing positive means the individual will develop the disorder, although onset is variable. In such cases, the general consensus in the genetics community is that children should not be tested unless there is a clear medical benefit in doing so. If there is no medical benefit, then children should not be tested because it removes their autonomy and their right to an open future. The recommendation is that children should be given the choice of genetic testing when they are capable of making their own decisions.5–8

The recent introduction of next-generation sequencing technologies has substantially increased both the coverage and resolution of genetic information. While genetic testing may refer to relatively targeted investigations involving the search for mutations in one or more specific genes, ‘genome-wide’ testing can identify multiple variants across the entire genome. Recent advances in diagnostic testing pose challenges to paediatrics and adoption as new methods—microarray-based comparative genomic hybridisation (aCGH)—have largely replaced the light microscopy approach to chromosome analysis as the first-line investigation for identifying a likely cause of a child’s developmental, learning and behavioural difficulties.9 10 Microarray can be used to quickly scan a genome for chromosomal imbalances at a very high level of resolution.11 The increased sensitivity of detecting small deletions and duplications (collectively known as copy number variations (CNVs)) has improved diagnostic yield from 3% in traditional karyotype testing to approximately 10%–15% in aCGH.10 12 It also means that for a significant proportion of cases, diagnostic testing still fails to identify a genetic cause, which challenges the widely held assumption that genetic testing is definitive and straightforward.9 In fact, genome-wide screening can produce four types of outcome: finding a diagnosis for the clinical problem, finding no abnormality, finding something of unknown significance or finding an abnormality of likely clinical importance but not causing the clinical problem at hand (known as an incidental finding).13

The clinical utility of diagnostic genetic testing is further clouded by the fact that a negative result does not unequivocally rule out a genetic cause because aCGH can only detect copy number changes—it cannot detect changes (variants) within genes.14 Moreover, the detection of a variant in a child or parent does not mean it has necessarily caused the specific problem because the clinical significance of some CNVs is not well understood. For instance, a microdeletion within 15q11.2 is commonly associated with a wide range of features including autistic traits and learning difficulties.15 However, variable expressivity and incomplete penetrance in populations suggests that this microdeletion is not necessarily causal, which has raised concerns that its detection among looked-after children can be misleading and potentially stigmatising.16

Fetal alcohol spectrum disorder (FASD) is a common embryopathy found more frequently among looked-after children.17 Clinical diagnostic criteria include evidence of maternal alcohol, intellectual disability, growth restriction and dysmorphic facial features. Diagnosis is challenging, however, because there is no reliable biomarker, clinical features are non-specific and there may be insufficient information about the pregnancy. Futhermore, the associated dysmorphic features overlap with features found in some chromosomal disorders.9 For these reasons, the use of aCGH is a valid attempt to exclude other possible causes of developmental problems and dysmorphic features, but the absence of an abnormality is not a direct confirmation of FASD.18 The British Medical Association guidelines support the involvement of clinical genetics in gathering evidence of perinatal history and performing careful examination of dysmorphic features before proceeding with genetic testing.19

As chromosomal microarray testing becomes a mainstream diagnostic technology, there are concerns that non-genetics professionals, especially those working within social services, may not be fully informed about the limitations of a genetic test result.16 Some may believe that genetic testing supersedes clinical assessment and removes uncertainty about a child’s present or future health. The aim of this study was to explore the knowledge and expectations of social workers’ and medical advisors’ accounts of genetic testing in adoption.

Methods

Twenty participants were recruited through purposive sampling of social workers (n=8), social work managers (n=4) and medical advisors/community paediatricians (n=8) working in adoption services in Wales. Social workers were contacted through the National Adoption Service, all of whom were self-selecting as having first-hand experience of cases that involved genetic testing. Medical advisors working in each of the local authorities were contacted individually via email. Snowball sampling techniques were used to contact specific participants who were singled out as having relevant experience; this widened our recruitment to England. Medical advisors were all mid-to-late career, social workers were more diverse, ranging from the newly qualified to those with over 30 years’ experience. Table 1 presents participant characteristics including the number of cases they discussed in relation to genetic testing. The asterisk (*) symbol indicates where participants discussed cases generically.

Semi-structured qualitative interviews were used to gather detailed accounts of retrospective cases. Treating research interviews as ‘accounts’ recognises that interviews are more than representations of past events and experiences, but complex social activities oriented to performing certain actions in the present, such as explaining or defending the speaker’s conduct or practical reasoning.20 Interviews explored a range of issues concerning: past cases of preadoption genetic testing, circumstances that trigger a genetic investigation, procedures for recording and communicating genetic information between professionals, practices of sharing information with prospective adopters, the impact of genetic information on postplacement experiences and whether a genetic diagnosis disadvantages a child’s prospects for adoption. All the interviews lasted between 60 and 90 min, were audio-recorded and subsequently transcribed verbatim.

Transcripts were coded via an iterative process of reading and noticing relevant phenomena, allowing the analyst (first author) to arrange data according to differences, commonalities and structures. Coded selections of data were compiled into a subcorpus for group discussion (involving all three authors). Data extracts were then selected in order to identify and illustrate ‘themes’ relating to implicit assumptions and expectations that underpin professionals’ accounts of genetic testing. Each of the major themes represent a range of statements about the reasons given for, and presumed outcomes of, genetic testing in adoption (tables 2 and 3).

Results

Discussion

The study found that social workers often described ‘genetic testing’ as a generic procedure for detecting the presence or absence of a specific aetiology, the results of which were believed to be definitive and mutually exclusive. In the absence of specific knowledge about genetic testing, social workers adopted an essentialist model (a sociocognitive heuristic that assumes simple causality and gene determinism) to make sense of past cases.21 Although some displayed knowledge of gene expression and gene-environment interaction, social workers assumed that genetic testing offered clarity in adoption because it removed uncertainty about the child’s health. Many explained that these uncertainties were unpalatable to a majority of adopters.

It was not unexpected to find that medical advisors were better informed about genetic testing and therefore recognised that many genetic conditions are multifactorial. Indeed, they were far more circumspect and agnostic about the outcomes of genetic testing, especially the results of aCGH. Rather than reducing uncertainties about the child, many advisors explained that uncertainty was an unavoidable outcome of assessment, and often at odds with the knowledge and expectations of social workers and prospective adopters. Cases that involved reporting variants of unknown significance were described as unwelcome outcomes that required careful and sensitive communication.

Medical advisors also reported multiple cases involving inappropriate requests for testing from social workers, special guardians, prospective adopters and legal professionals. There was apparently a general misconception that genetic testing was capable of confirming or excluding a genetic condition in the child. In some cases, testing was pursued for unaffected children on the basis of family history. Medical advisors expressed concerns that testing unaffected children was medically and ethically inappropriate as well as an unsuitable strategy for diagnosing complex, multifactorial conditions.

The study also revealed cases in which social workers and legal professionals believed in the existence of a genetic test for FASD. This may arise from the fact that some medical advisors use microarray to exclude chromosomal abnormalities in children presenting with dysmorphic features. However, several medical advisors stressed that diagnostic testing for FASD should be approached with caution. Where there is clinical suspicion of FASD, children should receive formal clinical assessment as well as detailed assessment of the available family before deciding to pursue a genetic investigation. Even then, the outcomes may have limited clinical utility and may affect the child’s prospects of adoption. A negative aCGH result could be misinterpreted as confirming a diagnosis of FASD, while finding a microdeletion of reduced penetrance could be taken inappropriately as undermining the same diagnosis.

Conclusion

This study explores the ways in which different professional groups make sense of genetic assessment in adoption. Social workers and medical advisors play a crucial role in ensuring that vulnerable children are properly assessed and supported to achieve permanency with an adoptive family. However, recent advances in genetic technologies, notably the shift from genetic to genomic testing, are changing the ways in which professionals understand and tolerate uncertainty. This study reveals cases in which social workers and medical advisors have different understandings and expectations about the clinical utility of genetic testing. The priority for social workers is to use health information to increase the transparency of information about the child in order to secure their placement with a suitable family. In many cases, information regarding medical and developmental uncertainties is construed as obstacles to the adoption process. The priority for medical advisors is to report these uncertainties as and when they apply, to place the risks of the child in context and to protect the child’s future autonomy from unnecessary testing. The findings suggest that social workers would benefit from specific training on the ethical and technical aspects of genetic and genomic testing, while multidisciplinary team meetings would provide a practical forum to discuss the contrasting priorities of adoption professionals.