Pharmacogenomics is the study of how genetic variations influence an individual’s drug response. It is used to optimize drug choice, dosing, and safety. It also has the potential to reduce costs and improve patient outcomes . Although published evidence has reached conflicting conclusions regarding the clinical effectiveness of pharmacogenomic testing, evidence continues to develop and is expected to identify many effective applications of pharmacogenomics [2–6].
As evidence of the effectiveness of pharmacogenomics increases, implementation of pharmacogenomics has become an area of increased study . Pharmacogenomic implementation barriers examined in prior studies include the need for pharmacogenomics education, strategies to access pharmacogenomic information needed to make clinical decisions, clinical workflow, and workload issues [8–12]. Currently, pharmacogenomic information is mainly included in US Food and Drug Administration (FDA) drug labels or cataloged by organizations such as the Clinical Pharmacogenomics Implementation Consortium (CPIC) and PharmGKB as guidelines and recommendations. These guidelines are complex, lengthy, and difficult for the workflow of busy inpatient physicians.
This study, ‘Implementation of Point-of-Care Pharmacogenomic Decision Support Accounting for Minority Disparities’, sought to implement pharmacogenomics into inpatient practice at three sites: The University of Chicago, Northwestern Memorial Hospital, and the University of Illinois at Chicago . This paper focuses on the barriers we faced in garnering the participation of hospitalists in using pharmacogenomic information for clinical decisions at the University of Chicago.
Results: strategies to enhance implementation
A lack of knowledge about important drug–gene relationships is a major reason why the utilization of pharmacogenomics in clinical practice is low. As a result, the first step in implementation was to identify a platform for pharmacogenomic prescribing that would provide providers with critical information for the use of pharmacogenomic information.
We were fortunate to have an existing solution, the Genomic Prescribing System (GPS), which has been described in previous studies, including ‘1200 Persons Project’ [14,15]. GPS addresses the barriers to available pharmacogenomics information, lack of knowledge, and interpretation difficulty of pharmacogenomic-drug relationships. The GPS is a web-based portal that delivers a clinical interpretation of pharmacogenomic information to providers while simultaneously providing a user-friendly support tool for the interpretation of results presented as traffic lights. The GPS uses genotype data to guide the provider with evidence-based information about medications with pharmacogenomics displayed in the form of electronic clinical consults (Fig. 1). GPS contains 90 drugs and a custom genomic panel including variants of 2D6 gathered from published sources of pharmacogenomic information: FDA, Clinical Pharmacogenetics Implementation Consortium (CPIC), PharmGKB, and published manuscripts. We believe the visual of traffic lights makes an easily recognizable system for providers to proceed with pharmacogenomic information .
Hospital medicine section meetings
The research coordinator introduced the study to the hospitalist group in a monthly section meeting. After the introduction of the study at a section meeting, clinicians completed consent forms without monitoring to respect provider participation preferences. We provided education to build provider interest in pharmacogenomics and familiarize them with the utility of GPS. During multiple meetings, we visually demonstrated GPS function using patient case examples of how precision medicine and associated genomic information can enhance clinical decisions. We continued to attend section meetings and presented cases and reported data on the top five users of the GPS by logins to provide evidence of peer participation. We also used these subsequent educational sessions to recruit non-participating or new providers.
Use of physician champions
Once providers had been introduced to the GPS, physician champions were important for ongoing provider recruitment and GPS utilization. Some providers were unfamiliar (Table 1) with pharmacogenomics or perceived study participation as a workload burden without clinical benefit. To address concerns about value, two identified ‘physician champion’ hospitalists invested in the study were recruited to advocate for the study. The two advocates led implementation efforts, and recruitment on an individual level, and assisted with interpreting genetic test results.
Table 1 -
Encountered barriers and implemented solutions supporting the deployment of pharmacogenomics into inpatient general medicine
|Delivery: pharmacogenomic information and lack of knowledge for interpretation
||Existing system of a web-based portal that provides a convenient, concise system for providers to easily access pharmacogenomic information called the Genomic Prescribing System
|Pharmacogenomic introduction and education
||Presentations and clinical cases at Hospitalist Section Meetings
|Recruitment and buy-in
||Hospitalist section meetings were used to recruit and show clinical cases of how pharmacogenomics helped to encourage buy-in and use. Physician champions who were hospitalists actively recruited and brought the physical presence of support and encouragement
||Initially through paging and attending multi-disciplinary rounds by research coordinators but transitioned to EMR alerts (during prescribing of medications with pharmacogenomic information)
|Hospitalist workflow: Non-integration into EMR and separate passwords
||Research coordinators attend multi-disciplinary rounds to help with passwords and reminders on how to use the system. With the introduction of integration of GPS into EMR and auto login, reduced steps for hospitalists
|Role of hospitalists: focus on acute vs. chronic
||1. Encourage to make changes to chronic medications based on formulary restrictions and documentation with smart phrases
2. Documentation to outpatient providers with smart phrases on discharge
||Creation of smart phrases for concise, easy, and standardized communication in notes
GPS, Genomic Prescribing System; EMR, electronic medical record.
Optimizing system function to fit hospitalist workflow
Hospitalist workflow requires the coordination of numerous responsibilities, including an emphasis on efficiency and throughput. Adding the responsibility to consider pharmacogenomic information in clinical decisions within a constrained workflow was identified as a potential barrier.
Chief among the challenges was the provision of pharmacogenomic information with minimal disruption to workflow. We first tried to reduce barriers to the use of GPS by integrating it into the electronic medical record (EMR) by creating a single sign-in through our EMR. The prior method required entering a different username and password to log into the roster of patients through an external webpage and then searching for the patient. Next, we were able to fully integrate GPS into the EMR by removing the whole sign-in process, when we developed a best practice alert (BPA) which contains a link that allowed direct access to the patient’s pharmacogenomic information (Fig. 2). BPA would open each time there was a medication prescribed that has pharmacogenomic information. We were able to solve a major complaint of remembering another username and password along with multiple steps in a hospitalists’ workflow to obtain genomic information.
Hospitalists rely on progress and handoff notes for communication across shifts and providers. To support effective handoffs in the context of this intervention and encourage hospitalists to use the GPS, we created concise dot phrases (allow text to be easily inserted into notes) that can be used in notes. The dot phrases would document and clarify for future providers what the GPS has provided, whether or not a drug was changed, and the reasoning for that decision from a standard list of options (Fig. 2).
Experience implementing these interventions at the University of Chicago as the first site for the project led to the identification of barriers to implementation (Table 1). Barriers to knowledge requiring education and integration into EMR have been discussed in other studies . We encountered barriers that led to the development of strategies to encourage provider participation by creating more streamlined and efficient processes that would better integrate into the hospitalists’ workflow. We continue to work to integrate pharmacogenomics into inpatient care and workflow, with the intent that it will become the standard of care like checking vitals and labs. We are still collecting data on how access to pharmacogenomic information affects providers’ and patients’ experiences, and its clinical impact. We believe our model and our study is an exciting paradigm shifts in how we practice inpatient medicine and how we approach prescribing in the 21st century.
The study was supported by NIH Grant U54MD010723.
Conflicts of interest
M.J.R. receives royalties related to UGT1A1 genotyping. For the remaining authors, there are no conflicts of interest.
1. Weinshilboum RM, Wang L. Pharmacogenomics: precision medicine
and drug response. Mayo Clin Proc 2017; 92:1711–1722.
2. Billings J, Racsa PN, Bordenave K, Long CL, Ellis JJ. The impact of real-world cardiovascular-related pharmacogenetic testing in an insured population. Int J Clin Pract 2018; 72:e13088.
3. Sorich MJ, Rowland A, McKinnon RA, Wiese MD. CYP2C19 genotype has a greater effect on adverse cardiovascular outcomes following percutaneous coronary intervention and in Asian populations treated with clopidogrel: a meta-analysis. Circ Cardiovasc Genet 2014; 7:895–902.
4. Plumpton CO, Roberts D, Pirmohamed M, Hughes DA. A systematic review of economic evaluations of pharmacogenetic testing for prevention of adverse drug reactions. PharmacoEcon 2016; 34:771–793.
5. Reese ES, Daniel Mullins C, Beitelshees AL, Onukwugha E. Cost-effectiveness of cytochrome P450 2C19 genotype screening for selection of antiplatelet therapy with clopidogrel or prasugrel [published correction appears in Pharmacotherapy. 2012 Jun;32(6):581]. Pharmacotherapy 2012; 32:323–332.
6. Kaufman AL, Spitz J, Jacobs M, Sorrentino M, Yuen S, Danahey K, et al. Evidence for clinical implementation of pharmacogenomics in cardiac drugs. Mayo Clin Proc 2015;90:716–729.
7. van der Wouden CH, Cambon-Thomsen A, Cecchin E, Cheung KC, Dávila-Fajardo CL, Deneer VH, et al. Ubiquitous Pharmacogenomics Consortium. Implementing pharmacogenomics in Europe: design and implementation strategy of the ubiquitous pharmacogenomics consortium [published correction appears in Clin Pharmacol Ther
. 2017 Jul;102(1):152]. Clin Pharmacol Ther 2017; 101:341–358.
8. Bank PCD, Swen JJ, Guchelaar HJ. Implementation of pharmacogenomics in everyday clinical settings. Adv Pharmacol 2018; 83:219–246.
9. Klein ME, Parvez MM, Shin JG. Clinical implementation of pharmacogenomics for personalized precision medicine
: barriers and solutions. J Pharm Sci 2017; 106:2368–2379.
10. Dunnenberger HM, Crews KR, Hoffman JM, Caudle KE, Broeckel U, Howard SC, et al. Preemptive clinical pharmacogenetics implementation: current programs in five US medical centers. Annu Rev Pharmacol Toxicol 2015; 55:89–106.
11. Arwood MJ, Chumnumwat S, Cavallari LH, Nutescu EA, Duarte JD. Implementing pharmacogenomics at your institution: establishment and overcoming implementation challenges. Clin Transl Sci 2016; 9:233–245.
12. Borden BA, Galecki P, Wellmann R, Danahey K, Lee SM, Patrick-Miller L, et al. Assessment of provider-perceived barriers to clinical use of pharmacogenomics during participation in an institutional implementation study. Pharmacogenet Genomics 2019; 29:31–38.
13. Friedman PN, Shaazuddin M, Gong L, Grossman RL, Harralson AF, Klein TE, et al. The ACCOuNT Consortium: a model for the discovery, translation, and implementation of precision medicine
in African Americans. Clin Transl Sci 2019; 12:209–217.
14. O’Donnell PH, Wadhwa N, Danahey K, Borden BA, Lee SM, Hall JP, et al. Pharmacogenomics-based point-of-care clinical decision support significantly alters drug prescribing. Clin Pharmacol Ther 2017; 102:859–869.
15. O’Donnell PH, Danahey K, Jacobs M, Wadhwa NR, Yuen S, Bush A, et al. Adoption of a clinical pharmacogenomics implementation program during outpatient care – initial results of the University of Chicago ‘1,200 Patients Project’. Am J Med Genet C Semin Med Genet 2014; 166C:68–75.