In this study, we investigated attitudes, knowledge, and educational needs of 128 mental health professionals practising in England about prescribing psychotropic drugs based on pharmacogenomics tests. We found that most mental health professionals had a positive attitude towards pharmacogenomics, although most did not feel familiar with ordering and interpreting the results of a pharmacogenomics test.
In addition, we found that most participants had learnt about key pharmacogenomics topics, such as genetics and drug metaboliser phenotypes, while at university, although most had not received any formal training for interpreting the results of a pharmacogenomics test. This highlights a knowledge gap that may need addressing to allow for safe implementation of pharmacogenomics in mental health services in England. Notably, participants’ performance in knowledge testing reflected these unmet educational needs, with just 6% answering all four questions correctly (with each question having four answers to choose form, and only one correct answer). We propose that specific training will provide mental health professionals with greater knowledge and confidence to order and interpret pharmacogenomics tests.
When asked what could be preventing participants from implementing pharmacogenomics testing in clinical practice, most required a better understanding of pharmacogenomics topics (including genetics and drug metabolism), as well as the support of their working institution. This suggests that two distinct strands may help to address these concerns. Firstly, tailored education programmes may be developed and offered to mental health professionals, with a focus on providing practical advice to aid interpretation of pharmacogenomics tests. Secondly, institutions (or more specifically, NHS trusts) should be aware that successful inclusion of pharmacogenomics into clinical practice may require their engagement and support. While beyond the scope of this study, NHS trusts that specialise in mental health should consider the types of support that are needed to ensure pharmacogenomics tests are ordered where it may benefit the patient.
Comparing our findings with Just et al. [8], we highlight that their cohort was composed of clinicians from a range of medical specialties participating in the “European pharmacogenomics clinical implementation project Ubiquitous Pharmacogenomics (U-PGx)” and therefore may be more experienced or interested in pharmacogenomics. By contrast, our cohort included a mixture of healthcare professionals, some of whom are involved in pharmacogenomics research. The proportions of physicians, pharmacists, and other professionals enrolled in their study were like ours.
Through a statistical comparison, shown in Table 2, responses from our cohort were not significantly different to Just et al. in questions about experience and attitudes toward pharmacogenomics; pharmacogenomics knowledge; and learning preferences. However, our cohort displayed a significant reduction in self-assessed familiarity with four pharmacogenomics topics, compared with Just et al. This may be specific to mental health services and suggests (as stated above) that attention should be given to developing and providing educational resources for this group of professionals. Moreover, this may also be specific to health services in England, which is where our cohort were drawn from. While no significant differences were observed in the number of correct answers to pharmacogenomics knowledge-testing questions, a lower self-assessed familiarity may result in fewer pharmacogenomics tests being ordered in mental health services. Indeed, while no significant differences were observed in proportions of participants ordering or recommending pharmacogenomics tests, numbers in our study were lower than Just et al. (18.3 and 34.3%, respectively; Table 2).
In addition, our research aligns with Chan et al.’s (2017) study [22], in which mental healthcare professionals were surveyed in Singapore. They found that most respondents (81%) believed pharmacogenomics testing could effectively identify suitable treatments, while 71% believed it could help with medication intolerance. However, like our study, only 46.4% felt confident in their ability to order these tests. Gender, profession (doctors versus pharmacists), and seniority were factors that significantly influenced responses. Additionally, 94.3% of respondents expressed concern about costs and 84.5% were concerned about the lack of clear guidelines for using pharmacogenomics in clinical practice. Indeed, 98.5% of respondents indicated a desire to learn more about the practical applications of pharmacogenomics, with education being preferred in a variety of formats.
Research has also indicated that clinicians face difficulties in incorporating pharmacogenomics information into their practice and are unsure of its impact [7,8,9,10,11]. This was further reinforced by a survey focusing on US nephrologists about their stance on pharmacogenomics testing [23]. Although they accepted the notion that genetic variations can influence course and treatment of chronic kidney disease, most medical professionals reported unease when discussing genetic results with patients whatever their opinion on pharmacogenomics testing may be. Nevertheless, physicians appeared more likely to utilise such tests when more aware of the supporting evidence.
Guo et al. [7] surveyed Chinese healthcare professionals about pharmacogenomics. They found that while 50% of physicians were unaware of the significance of pharmacogenomics in drug therapy, most were hopeful for its future in China, but regulations, sector standards, reporting standardisation, and a knowledge base must be implemented to make implementation of pharmacogenomics successful. These are similar to our study, given that improved knowledge base and institutional support were considered key factors for successful implementation of pharmacogenomics into regular mental health clinical practice.
St Savuer et al. [1] surveyed primary care physicians in the USA and discovered that only 30% made at least one change to their prescribed medication based on the pharmacogenomics alerts they received, while 53% did not find the alert reports useful. Around 45% of the respondents were unsure if they would use this alert system in the future. Furthermore, over half admitted feeling confused, irritated, frustrated or unable to find extra information when presented with a pharmacogenomics alert. Like our study, interpreting pharmacogenomics tests appears to be a key knowledge gap in several clinical settings.
Behr et al. [24] investigated the confidence and knowledge that US primary care and pain management healthcare providers have about pharmacogenomics. Most believed that pharmacogenomics could lead to better medication-related outcomes for patients, but at the same time did not feel prepared to apply the test results in their prescribing decisions due to a lack of experience with pharmacogenomics training and education, as well as inadequate familiarity with the United States Clinical Consortium (CPIC) and FDA resources for this purpose.
Our study was limited by the survey population consisting mainly of physicians, of whom general adult psychiatrists represented nearly half. This could influence the applicability of the results to different psychiatric subspecialties and general practitioners. The convenience sampling used could also limit its generalisability. Furthermore, participants were mainly drawn from two large mental health NHS trusts in London and recruited in the context of academic presentations, which are provided as part of a continuing professional development programme. These local meetings are mainly attended by psychiatrists, thus limiting a more multi-disciplinary sample.
The strengths of this study are that, to our knowledge, this is the first study specifically focusing on mental health professionals in England, aiming to explore their views and experiences around pharmacogenomics. A good degree of comparability with healthcare professionals from multiple clinical backgrounds was guaranteed by adopting the questionnaire by Just et al.
Concluding remarks
This study suggests that mental health professionals have a positive attitude towards pharmacogenomics but face similar challenges to those of other healthcare practitioners when it comes to implementing pharmacogenomics in their clinical practice. It is evident that both a more tailored education program focused on interpreting pharmacogenomics tests and additional support from institutions are likely to be required for mental health professionals, so they can feel more comfortable ordering tests and making sense of results. To address some of these issues, experts from across the NHS have started collaborating with NHS England’s Genomics Education Programme to create GeNotes – genomic notes for clinicians – a ‘just in time’ educational resource about genomics and pharmacogenomics designed specifically for healthcare professionals [25]. We expect GeNotes to be one of many resources supporting pharmacogenomics-informed prescribing over coming years in UK mental health practice.
Research should aim to better characterise unmet educational needs and effective ways to address them, given that a growing body of evidence highlights a gap in knowledge preventing pharmacogenomics from being integrated into clinical practice, regardless of the medical specialisation. Further research is also needed to optimise the usability and implementation of pharmacogenomics test reports.