Genetic testing in intellectual disability psychiatry: Opinions and practices of UK child and intellectual disability psychiatrists

Background An increasing number of genetic causes of intellectual disabilities (ID) are identifiable by clinical genetic testing, offering the prospect of bespoke patient management. However, little is known about the practices of psychiatrists and their views on genetic testing. Method We undertook an online survey of 215 psychiatrists, who were contacted via the Royal College of Psychiatrist's Child and Adolescent and Intellectual Disability Psychiatry mailing lists. Results In comparison with child and adolescent psychiatrists, intellectual disability psychiatrists ordered more genetic tests, referred more patients to genetic services, and were overall more confident in the genetic testing process. Respondents tended to agree that genetic diagnoses can help patient management; however, management changes were infrequently found in clinical practice. Conclusions Differences are apparent in the existing views and practices of child and adolescent and intellectual disability psychiatrists. Developing training and collaboration with colleagues working in genetic services could help to reduce discrepancies and improve clinical practice.

WOLFE Et aL. and epilepsy. For example the 22q11.2 deletion syndrome is associated with intellectual disability, but is also is one of the strongest risk factors for psychosis (Schneider et al., 2014).
Individuals with intellectual disability face obstacles accessing both physical and mental health services and health inequalities have been described (Emerson, Baines, Allerton, & Welch, 2012).
Understanding the genetic aetiology of intellectual disability could help to address some of these inequalities by facilitating individualized care plans. For example, there are clinical management guidelines available for the 22q11.2 deletion syndrome. Screening for specific physical health conditions, including cardiac, renal and immunology investigations, and a comprehensive mental health assessment are recommended (Habel et al., 2014). Such guidelines offer good prospects for early intervention and optimized health care, although they are not yet available for every genetic cause of intellectual disability.
In the UK, infants and children presenting with developmental delay are often seen by paediatricians, who can initiate genetic investigations and where appropriate refer onto specialist child and adolescent mental health services (CAMHS). Adult services are generally provided by specialist intellectual disability psychiatrists. Referrals for genetic testing can be made to National Health Service (NHS) Regional Genetics Centres (RGCs), which offer clinical genetics expertise in syndromes, cascade testing and counselling. Inequities in access to genetic testing have, however, encouraged the mainstreaming of genetic investigations, with an increased emphasis placed on medical specialists ordering tests directly (Burton, 2011).
Intellectual disability is often associated with co-morbid psychiatric disorders and/or behavioural problems. Recent estimates from United Kingdom (UK) primary care records show that approximately 21% of individuals with intellectual disability have a psychiatric disorder, 25% have some record of challenging behaviour, and 49% had been prescribed psychotropic drugs (Sheehan et al., 2015). Given that investigation of the cause of intellectual disability predominately occurs at diagnosis in childhood, there is a large cohort of adults, many with later onset psychiatric disorders, who have not had a diagnostic assessment utilizing the latest genetic technologies (Baker, Raymond, & Bass, 2012). We recently recruited 202 adults with idiopathic intellectual disability from UK psychiatry services and found that 11% had undiagnosed clinically relevant CNVs (Wolfe et al., 2017).
Whilst the role of specialist clinicians in ordering/referring for genetic testing is evolving, little is known about their current views and practices. We aimed to explore the attitudes and practices of UK psychiatrists working in CAMHS and adult intellectual disability psychiatry services on genetic testing in intellectual disability.

| METHOD
Psychiatrists from UK CAMHS and adult intellectual disability psychiatry services were surveyed as to their attitudes towards and current use of genetic investigations using an online survey.

| Survey
The questions were developed through consultation with intellectual disability psychiatrists, a clinical geneticist, a genetic counsellor, a genetic researcher and a statistician. Following a pilot, a number of the questions were amended, and the opportunity for open text responses was enabled. The 28-item self-administered survey was composed of yes/no responses, multiple choice Likert scale questions, numeric outcomes and free text responses (available in the Appendix).
The survey was programmed not to force answers to questions and enable completion of the survey with missing responses. It was administered via the online service tool Survey Monkey (SurveyMonkey Inc. Palo Alto, California, USA).

| Participants
The survey was distributed to members of the Faculty of Child and Adolescent Psychiatry and members of the Faculty of Psychiatry of Intellectual Disability via the Royal College of Psychiatrists' mailing list. Psychiatrists were invited by email to participate in the survey.
A participation reminder was sent after 1 week. Respondents were removed from the analysis if they were junior trainees or listed professions other than CAMHS psychiatry and adult intellectual disability psychiatry, if they lived outside the UK and if they had not seen any patients with intellectual disability in the previous 12 months.

| Statistical analysis
Quantitative statistical analyses were undertaken using IBM SPSS Statistics for Windows, Version 22.0 (IBM Corp, Armonk, NY, USA).
The analysis compared CAMHS psychiatrists (referred to henceforth as child psychiatrists) and adult intellectual disability psychiatrists (referred to henceforth as intellectual disability psychiatrists).
Continuous outcome variables were analysed, using a t test where the data were normally distributed, and Mann-Whitney U test for non-normally distributed data. The chi-squared test was utilized to test categorical outcome variables. Binary logistic regression was undertaken to test univariable factors related to ordering a genetic test. For Likert scale responses, the data were collapsed from 5 to 3 scale responses by merging "strongly agree" with "agree" and "strongly disagree" with "disagree" or "very frequently" with "frequently" and "very rarely" with "rarely." To compare clinical confidence ratings, a composite confidence score was generated by assigning the 5-point Likert scale responses a confidence value ranging from 1 for strongly disagree to 5 for strongly agree. These scores were then summed across the eight confidence measures to obtain an overall composite confidence score. Where analyses have been undertaken on a subset of the data set due to missing values the number of respondents in the analysis has been indicated. Significance has been set at 0.006 to account for multiple testing.

| Thematic analysis
Open text responses were thematically coded using Nvivo qualitative data analysis software (QSR International Pty Ltd. Version 10, 2012).
Three open text questions were included in the survey, focusing on the benefits and concerns of genetic testing in clinical practice. For the word cloud, all open text responses were included, and a word frequency analysis was undertaken using Nvivo. Word stemming was undertaken to combine variations of words from the same root (e.g., genetic and genetics). All words mentioned greater than 5 times, excluding common words, were inputted into Wordle™ (http://www. wordle.net/) for the creation of the word cloud.

| Ordering of genetic tests
More intellectual disability psychiatrists (77%), compared with child psychiatrists (56%), had ordered a genetic test in the last 10 years (n = 162, χ² = 8.08, p = .004). Respondent's estimates of the percentage of intellectual disability caused by genetic factors did not influence the likelihood of them ordering a genetic test (n = 157, OR 1.01, 95% CI (0.99-1.03), p = .19). The percentage of patients on respondents' caseloads with an established genetic diagnosis also did not affect the likelihood of ordering a genetic test (n = 156, OR 1.02, 95% CI (0.99-1.05), p = .33).

| Confidence in the genetic testing process
Respondents were asked how confident they felt in eight aspects of the genetic testing process, as presented in Table 1

| Concerns with the genetic testing process
Respondents were asked what their main concerns were in relation to the genetic testing process, see Table 2. Both child and intellectual disability psychiatrists agreed that lack of available treatment was one of the main concerns (58% versus 51% retrospectively). Another main concern was lack of resources, 54% of child and intellectual disability psychiatrists agreed that this was a concern. Implications for insurance were a bigger concern for child psychiatrists in comparison with intellectual disability psychiatrists (50% versus 38%), whereas issues around counselling were a bigger concern for intellectual disability psychiatrists (53% versus 43%).

| Feedback of results and clinical management
As seen in Figure 1 both child and intellectual disability psychiatrists agreed that a genetic diagnosis is more beneficial for family members than patients. In comparison with child psychiatrists, intellectual disability psychiatrists were more inclined to agree that a diagnosis is beneficial for family members (85% versus 78%) ( Figure 1a) and patients (58% versus 50%) (Figure 1b).
Respondents were also asked how they fed back results to their patients with intellectual disability. Of the 146 respondents eight (5%) had utilized videos, 20 (14%) had received input from speech and language therapists, 48 (33%) had used easy read materials, and 98 (67%) had used none of these aids. Responses were comparable for child and intellectual disability psychiatrists.

| Referral to genetics services
Respondents were asked if they had ever ordered a genetic test or made a referral to a clinical genetics service. T A B L E 2 Concerns child and intellectual disability psychiatrists (n = 195) report in ten areas of the genetic testing process (31% child, 38% intellectual disability). The least likely reason for referral was pharmacological treatment (2% both child and intellectual disability).

| Service structure and training
Both intellectual disability and child psychiatrists agreed that closer links with regional genetics services would be helpful (83% versus 72%, n = 197). Respondents were also in agreement that they would prefer to refer to a regional genetics service rather than order a genetic test themselves (child 85%, 77% intellectual disability n = 195).
Finally, there was a consensus that further training in genetics would be beneficial (child 71%, 66% intellectual disability, n = 195). Access to genetics services was mentioned by 22 respondents, who described problems with referring to genetics services and the variable levels of knowledge of professionals involved in the pathway.

| Thematic analysis
There was concern that psychiatrists, who have not specialized in genetics, do not have the skills to refer directly for genetic testing. Good working relationships with genetics services were said to be a valuable resource. Five child psychiatrists stated that they would defer to their paediatric colleagues to make decisions about genetic testing.
Several respondents felt that current training in genetics was insufficient and is not keeping abreast of technological advances.
It was suggested that quick reference guides and screening tools would be valuable resources to support the decision-making process.
See F I G U R E 2 Percentage of child (n = 121) and intellectual disability (n = 94) psychiatrists who feel that that a genetic diagnosis is helpful for patient management (a) and who report that genetic information has helped their patient management (b) (child n = 73, intellectual disability n = 82). [Colour figure can be viewed at wileyonlinelibrary.com].
psychiatrists) and a summary of positive and negative opinions for each of the main themes.

| DISCUSSION
Our results indicate that the majority of child and intellectual disability psychiatrists working with patients with intellectual disability are already ordering genetic tests or making referrals to genetics services.
However, there are several disparities in clinical genetic practices. In comparison with child psychiatrists, intellectual disability psychiatrists reported the following: a higher number of patients with genetic diagnoses, greater confidence in the genetic testing process, higher numbers of tests ordered and more patients referred per year to genetics services.
Respondents were asked to estimate the percentage of intellec- Both child (85%) and intellectual disability (77%) psychiatrists agreed that they would prefer to refer to a Regional Genetics Centre (RGC) rather than directly order a genetic test themselves; however, links with NHS RGCs appeared to be variable. Some respondents reported good links with their local genetics services, whilst others felt that access to services was a barrier to referring for genetic testing. Both intellectual disability (83%) and child psychiatrists (72%) felt that better links with genetics services would be beneficial.
Many of these clinicians felt that they do not have the knowledge or training to order genetic tests directly. This finding is supported by another survey, which found intellectual disability psychiatrists lacked adequate knowledge about genetics and testing processes (de Villiers & Porteous, 2012).
The majority of respondents expressed a wish for further training (71% child, 66% intellectual disability). Neither child and adolescent nor intellectual disability psychiatry curricula currently have learning objectives that specifically cover genetic disorders associated with intellectual disability (http://www.rcpsych.ac.uk/traininpsychiatry/corespecialtytraining/curricula.aspx). The curricula also fail to cover the genetic work-up and basic genetic counselling skills that are required to take more of an active role in identifying and managing patients with genetic disorders. However, there are several recent initiatives to improve the psychiatry curriculum.
For example, the Gatsby-Wellcome initiative aimed to ensure that training focuses more on scientific advances in basic and clinical neurosciences (http://www.rcpsych.ac.uk/traininpsychiatry/corespecialtytraining/neuroscienceproject.aspx). It is, therefore, hoped that future cohorts of psychiatrists will be more confident in utilizing technological advancements in the assessment and management of their patients.
One of the reasons for undertaking genetic investigations is that a genetic diagnosis is likely to provide information about specific associated medical and psychiatric phenotypes and thus could improve treatment plans and clinical management for the patient. Whilst the majority of respondents felt that a genetic diagnosis would help with clinical management, fewer patients on their caseloads had a genetic diagnosis than they would expect, and clinical management changes following genetic diagnoses were not frequently seen in practice.  (Waite et al., 2014). A recent survey of intellectual disability professionals found that nine out of ten professionals interviewed felt that specific knowledge of a neurodevelopmental syndrome should play a key role in healthcare provision. A specific genetic diagnosis was thought to prompt proactive screening for related physical and mental health problems, which is of particular benefit for patients with severe impairments (Redley, Pannebakker, & Holland, 2016). One of the main challenges in practice is that individual syndromes are rare and psychiatrists are unlikely to care for many individuals with the same disorder, although the overall burden of rare syndromic disorders is large.
Both child and intellectual disability psychiatrists agreed that receiving a genetic diagnosis was more beneficial for family members than for the patient. Research has shown that there is a benefit to mothers in receiving a diagnosis for a child with intellectual disability; however, there is a lack of research as to the impact of a genetic diagnosis for adults with intellectual disability (Lingen et al., 2016) Several respondents reported that a diagnosis can help to alleviate guilt for family members, as well as increasing understanding of the patient's syndrome-specific behaviours and enabling valuable access to support groups. It seems that respondents were able to report on a range of psychosocial benefits, which could indirectly improve patient management; however, tangible changes in clinical decision making following a genetic diagnosis were less easy to define.

| Limitations
The survey was self-reported which could have led to biases in estimations. There may have been a selection bias in the clinicians who chose to respond to the survey, perhaps those with more extreme views on genetics were more inclined to respond. This survey specifically focused on psychiatrists, who are one of the medical specialists frequently in contact with patients with intellectual disability in the UK. These findings may not be generalisable to other countries where services are organized differently. However, as the landscape of genetic investigations becomes more complex it is going to be a challenge for psychiatrists to keep pace of developments. Improvements in training and closer links with genetics services would appear to be key areas to address to meet this challenge.

ACKNOWLEDGMENT
Royal College of Psychiatrists, all respondents who completed the survey.

CONFLICT OF INTEREST
There are no conflicts of interest.

| 281
Published for the British Institute of Learning Disabilities WOLFE Et aL. • ______________% of people with ID.

9) I estimate that a significant genetic contribution towards the cause of the ID has been identified in…. % of my patients:
• ______________% of my patients.