9

dosing/scheduling effects or variations by key factors such as biological sex or race/ethnicity. Overall, Wang and colleagues have made a noteworthy and novel contribution regarding the potential role of vitamin D for psychosis outcomes in AD. The study also provides an example of how systems or ‘omics-based approaches might be used to identify new treatments for NPS, including psychosis, in AD and to uncover molecular origins and therapeutic targets. Although the exploratory nature of the findings indicates the need for replication, the study approach and results provide helpful direction for future work. Aim: To explore primary care physicians’ views of the utility and delivery of direct access to pharmacogenomics (PGx) testing in a community health system. Methods: This descriptive study assessed the perspectives of 15 healthcare providers utilizing qualitative individual interviews. Results: Three main themes emerged: perceived value and utility of PGx testing; challenges to implementation in practice; and provider as well as patient needs. Conclusion: While providers in this study viewed benefits of PGx testing as avoiding side effects, titrating doses more quickly, improving shared decision-making and providing psychological reassurance, challenges will need to be addressed such as privacy concerns, cost, insurance coverage and understanding the complexity of PGx test results. Pharmacogenomics (PGx) testing has the potential to personalize medication management for a wide range of patients [1]. More than 160 medications currently have PGx labeling from the US FDA, ranging from heart disease medications to psychiatric drugs [2]. Identifying genetic variants associated with response to medication may benefit patients for more accurate drug dosing and selection in order to optimize responses and/or to avoid unwanted side effects [3,4]. Given the wide reach of PGx and a shortage of genetic professionals to handle the potential volume, primary care physicians are in a front-line position to review and explain findings to patients from PGx tests [5,6]. Guidelines have become available for a number of medications with applicable PGx testing through the Clinical Pharmacogenetics Implementation Consortium [7–9]. Although physicians may perceive a benefit to using PGx [3,5,9–12], many barriers exist that have prevented implementation on a larger scale such as: physicians lacking sufficient knowledge and finding time to keep up-to-date; uncertainty surrounding clinical utility; lack of confidence or comfort in ordering PGx testing; lack of comfort interpreting results; uncertainty in how to communicate results to patients and having sufficient time to do so; difficulty in managing changes to workflow; determining how to incorporate results into the electronic medical record (EMR); and uncertainty about level of coverage by insurance and reimbursement [3,5,6,9–15]. While various studies have investigated physician knowledge, opinions, comfort level and preparedness for using PGx testing in practice, questions remain regarding the best delivery method of PGx testing and reporting results [16]. Implementation of PGx has been attempted in a variety of settings and using various methods. A multidisciplinary clinic approach utilized medical geneticists, pharmacists, an advanced practice nurse and genetic counselors [17]. Although this model provides the patient a thorough evaluation of the pharmacogenetic variants they are found to carry, such effort lacks scalability to be an efficient sole implementation effort for a health system. Pharmacist-delivered medication management demonstrated feasibility and satisfaction from patients; however, overcoming barriers of continued education and reimbursement still need to be addressed [18]. Other pharmacist-assisted methods have been implemented in community pharmacies which proved feasible and took little time, but patient understanding and satisfaction were not investigated [16]. Clinical decision support methods have been employed in some primary care practices; however, physicians may not always know how to utilize and incorporate pharmacogenomics decision alerts in their medical decision-making [19]. In one study, only

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18