Effectiveness of ChatGPT in clinical pharmacy and the role of artificial intelligence in medication therapy management.

BACKGROUND: The use of artificial intelligence (AI) to optimize medication therapy management (MTM) in identifying drug interactions may potentially improve MTM efficiency. ChatGPT, an AI language model, may be applied to identify medication interventions by integrating patient and drug databases. ChatGPT has been shown to be effective in other areas of clinical medicine, from diagnosis to patient management. However, ChatGPT's ability to manage MTM related activities is little known. OBJECTIVES: To evaluate the effectiveness of ChatGPT in MTM services in simple, complex, and very complex cases to understand AI contributions in MTM. METHODS: Two clinical pharmacists rated and validated the difficulty of patient cases from simple, complex, and very complex. ChatGPT's response to the cases was assessed based on 3 criteria: the ability to identify drug interactions, precision in recommending alternatives, and appropriateness in devising management plans. Two clinical pharmacists validated the accuracy of ChatGPT's responses and compared them to actual answers for each complexity level. RESULTS: ChatGPT 4.0 accurately solved 39 out of 39 (100 %) patient cases. ChatGPT successfully identified drug interactions, provided therapy recommendations and formulated general management plans, but it did not recommend specific dosages. Results suggest it can assist pharmacists in formulating MTM plans to improve overall efficiency. CONCLUSION: The application of ChatGPT in MTM has the potential to enhance patient safety and involvement, lower healthcare costs, and assist healthcare providers in medication management and identifying drug interactions. Future pharmacists can utilize AI models such as ChatGPT to improve patient care. The future of the pharmacy profession will depend on how the field responds to the changing need for patient care optimized by AI and automation.

Gamification to Improve Medication Adherence: A Mixed-method Usability Study for MedScrab.

Medication non-adherence is a prevalent healthcare problem with poor health outcomes and added healthcare costs. MedScrab, a gamification-based mHealth app, is the first attempt to deliver crucial life-saving medication information to patients and increase their medication adherence. The paper presents the development of MedScrab and a two-phase mixed-method usability evaluation of MedScrab. Phase I qualitatively evaluated MedScrab using a think-aloud protocol for its usability. With 51 participants, qualitative data analysis of Phase I revealed two themes: positive functionality of the app and four areas of improvement. The improvement recommendations were incorporated into MedScrab's design. Phase I also validated a widely used mHealth App Usability Questionnaire (MAUQ). Quantitative data analysis of Phase I reduced the original 18-item MAUQ scale to a 15-item scale with two factors: ease of use (4 items) and usefulness and satisfaction (11 items). Phase II surveyed 83 participants from Amazon's Mechanical Turk using a modified MAUQ. The modified MAUQ scale showed strong internal consistency (Cronbach alpha = 0.959) and high factor loadings (between 0.623 and 0.987). The study design of the usability evaluation can serve as a methodological guide for designing, evaluating, and improving mHealth apps.The usability study showed that MedScrab was perceived as ease of use (6.24 out of 7) with high usefulness and satisfaction (5.72 out of 7). The quantitative data analysis results support the use of the modified MAUQ as a valid instrument to measure the usability of the MedScrab. However, the instrument should be used with adaptation based on the app's characteristics.

Opportunities to integrate nutrigenomics into clinical practice and patient counseling.

BACKGROUND: Little progress has been made in translating nutrigenomics knowledge into clinical counseling in the past decade. Currently, clinicians are overwhelmed by nutrigenomics information without the proper scientific guidelines on patient counseling. METHODS: In this study, we conducted a scoping review of the primary literature to assess the current evidence of nutrigenomics counseling. A literature search using PRISMA guidelines identified the current challenges and opportunities facing nutrigenomics counseling in clinical practice. RESULTS: We identified four main themes: inadequate training, lack of awareness, underdeveloped nutrigenomics counseling skills, and unreliable evidence-based practice information. Many clinicians did not have the necessary knowledge to perform nutrigenomic counseling and were unaware of the available scientific information source. Moreover, there are no guidelines in the scientific community to counsel patients on nutrigenomics testing. CONCLUSION: Opportunities exist for government and non-government entities to create an evidence-based information platform using clinical guidelines to integrate nutrigenomics knowledge from bench to bedside successfully.

Artificial Intelligent Context-Aware Machine-Learning Tool to Detect Adverse Drug Events from Social Media Platforms.

INTRODUCTION: Pharmacovigilance (PV) has proven to detect post-marketing adverse drug events (ADE). Previous research used the natural language processing (NLP) tool to extract unstructured texts relevant to ADEs. However, texts without context reduce the efficiency of such algorithms. Our objective was to develop and validate an innovative NLP tool, aTarantula, using a context-aware machine-learning algorithm to detect existing ADEs from social media using an aggregated lexicon. METHOD: aTarantula utilized FastText embeddings and an aggregated lexicon to extract contextual data from three patient forums (i.e., MedHelp, MedsChat, and PatientInfo) taking warfarin. The lexicon used warfarin package inserts and synonyms of warfarin ADEs from UMLS and FAERS databases. Data was stored on SQLite and then refined and manually checked by three clinical pharmacists for validation. RESULTS: Multiple organ systems where the most frequent ADE were reported at 1.50%, followed by CNS side effects at 1.19%. Lymphatic system ADEs were the least common side effect reported at 0.09%. The overall Spearman rank correlation coefficient between patient-reported data from the forums and FAERS was 0.19. As determined by pharmacist validation, aTarantula had a sensitivity of 84.2% and a specificity of 98%. Three clinical pharmacists manually validated our results. Finally, we created an aggregated lexicon for mining ADEs from social media. CONCLUSION: We successfully developed aTarantula, a machine-learning algorithmn based on artificial intelligence to extract warfarin-related ADEs from online social discussion forums automatically. Our study shows that it is feasible to use aTarantula to detect ADEs. Future researchers can validate aTarantula on the diverse dataset.

The promise of digital health in healthcare equity and medication adherence in the disadvantaged dementia population.

Tweetable abstract Pharmacogenomics cascade testing in a digital health solution can improve medication adherence in dementia care for disadvantaged populations.

Framework to enable pharmacist access to health care data using Blockchain technology and artificial intelligence.

BACKGROUND: Decentralization and authentication are embedded in blockchain technology, which utilizes artificial intelligence (AI) to ensure seamless sharing of data among different health care providers while safeguarding data privacy. Although community pharmacists are highly accessible to patients and possess robust clinical knowledge, they are limited in the clinical services they can provide owing to their lack of access to patient health records. We proposed a blockchain- and AI-based conceptual framework by performing a scoping review of successful blockchain integration in health systems. OBJECTIVE: To formulate a conceptual framework based on a scoping review to improve access to health care data in the community pharmacy setting through the adoption of blockchain technology and AI. METHODS: We performed a scoping review of literature based on Preferred Reporting Items for Systematic reviews and Meta-Analyses review criteria to identify the specific areas where blockchain can be implemented in health systems. We utilized the Pharmacists' Patient Care Process (PPCP) to identify 2 critical areas for blockchain integration that can support community pharmacists to access patient electronic health records and implement patient-specific information in clinical decision-making. RESULTS: We included 7 articles out of 70 articles in our final review. The 2 areas in the PPCP identified for the use of blockchain on the basis of the literature review were "Assess" and "Implement." Our proposed model involves pharmacists using AI and blockchain technology to comprehensively assess any concerns with the prescribed medication through access to laboratory results for patients and then implement a plan based on a comprehensive assessment of the patient's health record. CONCLUSIONS: Utilizing blockchain to securely share health data with community pharmacies has the potential to improve patient outcomes, optimize medication safety, and amplify pharmacists' roles in patient care. Future research should focus on implementing the model in the real-world settings.

PGxKnow: a pharmacogenomics educational HoloLens application of augmented reality and artificial intelligence.

Aim: To develop and assess an augmented reality tool for pharmacogenomics (PGx) education based on artificial intelligence. Materials & methods: A HoloLens application was developed using feedback from three clinical PGx-trained pharmacists. 15 Participants independently reviewed the application and assessed usability using the system usability scale (SUS). Results & conclusion: Eighteen different frames were developed. Each video module was 2-3 min for the education. The application included textual information and 3D structures of PGx concepts. The mean SUS score for 15 participants (11 pharmacy students and four pharmacists) was 83, with a standard deviation of 6.6. Results suggest that PGxKnow has the potential to bridge the gap in PGx education, further widespread utilization of PGx and boost its impact on precision medicine.

Clinical Implications of Combinatorial Pharmacogenomic Tests Based on Cytochrome P450 Variant Selection.

Despite the potential to improve patient outcomes, the application of pharmacogenomics (PGx) is yet to be routine. A growing number of PGx implementers are leaning toward using combinatorial PGx (CPGx) tests (i.e., multigene tests) that are reusable over patients' lifetimes. However, selecting a single best available CPGx test is challenging owing to many patient- and population-specific factors, including variant frequency differences across ethnic groups. The primary objective of this study was to evaluate the detection rate of currently available CPGx tests based on the cytochrome P450 (CYP) gene variants they target. The detection rate was defined as the percentage of a given population with an "altered metabolizer" genotype predicted phenotype, where a CPGx test targeted both gene variants a prospective diplotypes. A potential genotype predicted phenotype was considered an altered metabolizer when it resulted in medication therapy modification based on Clinical Pharmacogenetics Implementation Consortium (CPIC) guidelines. Targeted variant CPGx tests found in the Genetic Testing Registry (GTR), gene selection information, and diplotype frequency data from the Pharmacogenomics Knowledge Base (PharmGKB) were used to determine the detection rate of each CPGx test. Our results indicated that the detection rate of CPGx tests covering CYP2C19, CYP2C9, CYP2D6, and CYP2B6 show significant variation across ethnic groups. Specifically, the Sub-Saharan Africans have 63.9% and 77.9% average detection rates for CYP2B6 and CYP2C19 assays analyzed, respectively. In addition, East Asians (EAs) have an average detection rate of 55.1% for CYP2C9 assays. Therefore, the patient's ethnic background should be carefully considered in selecting CPGx tests.

Scoping review: the empowerment of Alzheimer's Disease caregivers with mHealth applications.

Alzheimer's Disease (AD) is one of the most prevalent neurodegenerative chronic diseases. As it progresses, patients become increasingly dependent, and their caregivers are burdened with the increasing demand for managing their care. Mobile health (mHealth) technology, such as smartphone applications, can support the need of these caregivers. This paper examines the published academic literature of mHealth applications that support the caregivers of AD patients. Following the PRISMA for scoping reviews, we searched published literature in five electronic databases between January 2014 and January 2021. Twelve articles were included in the final review. Six themes emerged based on the functionalities provided by the reviewed applications for caregivers. They are tracking, task management, monitoring, caregiver mental support, education, and caregiver communication platform. The review revealed that mHealth applications for AD patients' caregivers are inadequate. There is an opportunity for industry, government, and academia to fill the unmet need of these caregiver.

SARS-CoV-2 early infection signature identified potential key infection mechanisms and drug targets.

BACKGROUND: The ongoing COVID-19 outbreak has caused devastating mortality and posed a significant threat to public health worldwide. Despite the severity of this illness and 2.3 million worldwide deaths, the disease mechanism is mostly unknown. Previous studies that characterized differential gene expression due to SARS-CoV-2 infection lacked robust validation. Although vaccines are  now available, effective treatment options are still out of reach. RESULTS: To characterize the transcriptional activity of SARS-CoV-2 infection, a gene signature consisting of 25 genes was generated using a publicly available RNA-Sequencing (RNA-Seq) dataset of cultured cells infected with SARS-CoV-2. The signature estimated infection level accurately in bronchoalveolar lavage fluid (BALF) cells and peripheral blood mononuclear cells (PBMCs) from healthy and infected patients (mean 0.001 vs. 0.958; P < 0.0001). These signature genes were investigated in their ability to distinguish the severity of SARS-CoV-2 infection in a single-cell RNA-Sequencing dataset. TNFAIP3, PPP1R15A, NFKBIA, and IFIT2 had shown bimodal gene expression in various immune cells from severely infected patients compared to healthy or moderate infection cases. Finally, this signature was assessed using the publicly available ConnectivityMap database to identify potential disease mechanisms and drug repurposing candidates. Pharmacological classes of tricyclic antidepressants, SRC-inhibitors, HDAC inhibitors, MEK inhibitors, and drugs such as atorvastatin, ibuprofen, and ketoconazole showed strong negative associations (connectivity score < - 90), highlighting the need for further evaluation of these candidates for their efficacy in treating SARS-CoV-2 infection. CONCLUSIONS: Thus, using the 25-gene SARS-CoV-2 infection signature, the SARS-CoV-2 infection status was captured in BALF cells, PBMCs and postmortem lung biopsies. In addition, candidate SARS-CoV-2 therapies with known safety profiles were identified. The signature genes could potentially also be used to characterize the COVID-19 disease severity in patients' expression profiles of BALF cells.

Pharmacogenomics cascade testing (PhaCT): a novel approach for preemptive pharmacogenomics testing to optimize medication therapy.

The implementation of pharmacogenomics (PGx) has come a long way since the dawn of utilizing pharmacogenomic data in clinical patient care. However, the potential benefits of sharing PGx results have yet to be explored. In this paper, we explore the willingness of patients to share PGx results, as well as the inclusion of family medication history in identifying potential family members for pharmacogenomics cascade testing (PhaCT). The genetic similarities in families allow for identifying potential gene variants prior to official preemptive testing. Once a candidate patient is determined, PhaCT can be initiated. PhaCT recognizes that further cascade testing throughout a family can serve to improve precision medicine. In order to make PhaCT feasible, we propose a novel shareable HIPAA-compliant informatics platform that will enable patients to manage not only their own test results and medications but also those of their family members. The informatics platform will be an external genomics system with capabilities to integrate with patients' electronic health records. Patients will be given the tools to provide information to and work with clinicians in identifying family members for PhaCT through this platform. Offering patients the tools to share PGx results with their family members for preemptive testing could be the key to empowering patients. Clinicians can utilize PhaCT to potentially improve medication adherence, which may consequently help to distribute the burden of health management between patients, family members, providers, and payers.

Artificial Intelligence-Powered Smartphone App to Facilitate Medication Adherence: Protocol for a Human Factors Design Study.

BACKGROUND: Medication Guides consisting of crucial interactions and side effects are extensive and complex. Due to the exhaustive information, patients do not retain the necessary medication information, which can result in hospitalizations and medication nonadherence. A gap exists in understanding patients' cognition of managing complex medication information. However, advancements in technology and artificial intelligence (AI) allow us to understand patient cognitive processes to design an app to better provide important medication information to patients. OBJECTIVE: Our objective is to improve the design of an innovative AI- and human factor-based interface that supports patients' medication information comprehension that could potentially improve medication adherence. METHODS: This study has three aims. Aim 1 has three phases: (1) an observational study to understand patient perception of fear and biases regarding medication information, (2) an eye-tracking study to understand the attention locus for medication information, and (3) a psychological refractory period (PRP) paradigm study to understand functionalities. Observational data will be collected, such as audio and video recordings, gaze mapping, and time from PRP. A total of 50 patients, aged 18-65 years, who started at least one new medication, for which we developed visualization information, and who have a cognitive status of 34 during cognitive screening using the TICS-M test and health literacy level will be included in this aim of the study. In Aim 2, we will iteratively design and evaluate an AI-powered medication information visualization interface as a smartphone app with the knowledge gained from each component of Aim 1. The interface will be assessed through two usability surveys. A total of 300 patients, aged 18-65 years, with diabetes, cardiovascular diseases, or mental health disorders, will be recruited for the surveys. Data from the surveys will be analyzed through exploratory factor analysis. In Aim 3, in order to test the prototype, there will be a two-arm study design. This aim will include 900 patients, aged 18-65 years, with internet access, without any cognitive impairment, and with at least two medications. Patients will be sequentially randomized. Three surveys will be used to assess the primary outcome of medication information comprehension and the secondary outcome of medication adherence at 12 weeks. RESULTS: Preliminary data collection will be conducted in 2021, and results are expected to be published in 2022. CONCLUSIONS: This study will lead the future of AI-based, innovative, digital interface design and aid in improving medication comprehension, which may improve medication adherence. The results from this study will also open up future research opportunities in understanding how patients manage complex medication information and will inform the format and design for innovative, AI-powered digital interfaces for Medication Guides. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/21659.

Application of Human Factors Methods to Understand Missed Follow-up of Abnormal Test Results.

OBJECTIVE: This study demonstrates application of human factors methods for understanding causes for lack of timely follow-up of abnormal test results ("missed results") in outpatient settings. METHODS: We identified 30 cases of missed test results by querying electronic health record data, developed a critical decision method (CDM)-based interview guide to understand decision-making processes, and interviewed physicians who ordered these tests. We analyzed transcribed responses using a contextual inquiry (CI)-based methodology to identify contextual factors contributing to missed results. We then developed a CI-based flow model and conducted a fault tree analysis (FTA) to identify hierarchical relationships between factors that delayed action. RESULTS: The flow model highlighted barriers in information flow and decision making, and the hierarchical model identified relationships between contributing factors for delayed action. Key findings including underdeveloped methods to track follow-up, as well as mismatches, in communication channels, timeframes, and expectations between patients and physicians. CONCLUSION: This case report illustrates how human factors-based approaches can enable analysis of contributing factors that lead to missed results, thus informing development of preventive strategies to address them.

The inclusion of health data standards in the implementation of pharmacogenomics systems: a scoping review.

Background: Despite potential benefits, the practice of incorporating pharmacogenomics (PGx) results in clinical decisions has yet to diffuse widely. In this study, we conducted a review of recent discussions on data standards and interoperability with a focus on sharing PGx test results among health systems. Materials & methods: We conducted a literature search for PGx clinical decision support systems between 1 January 2012 and 31 January 2020. Thirty-two out of 727 articles were included for the final review. Results: Nine of the 32 articles mentioned data standards and only four of the 32 articles provided solutions for the lack of interoperability. Discussions: Although PGx interoperability is essential for widespread implementation, a lack of focus on standardized data creates a formidable challenge for health information exchange. Conclusion: Standardization of PGx data is essential to improve health information exchange and the sharing of PGx results between disparate systems. However, PGx data standards and interoperability are often not addressed in the system-level implementation.

Improving Medication Information Presentation Through Interactive Visualization in Mobile Apps: Human Factors Design.

BACKGROUND: Despite the detailed patient package inserts (PPIs) with prescription drugs that communicate crucial information about safety, there is a critical gap between patient understanding and the knowledge presented. As a result, patients may suffer from adverse events. We propose using human factors design methodologies such as hierarchical task analysis (HTA) and interactive visualization to bridge this gap. We hypothesize that an innovative mobile app employing human factors design with an interactive visualization can deliver PPI information aligned with patients' information processing heuristics. Such an app may help patients gain an improved overall knowledge of medications. OBJECTIVE: The objective of this study was to explore the feasibility of designing an interactive visualization-based mobile app using an HTA approach through a mobile prototype. METHODS: Two pharmacists constructed the HTA for the drug risperidone. Later, the specific requirements of the design were translated using infographics. We transferred the wireframes of the prototype into an interactive user interface. Finally, a usability evaluation of the mobile health app was conducted. RESULTS: A mobile app prototype using HTA and infographics was successfully created. We reiterated the design based on the specific recommendations from the usability evaluations. CONCLUSIONS: Using HTA methodology, we successfully created a mobile prototype for delivering PPI on the drug risperidone to patients. The hierarchical goals and subgoals were translated into a mobile prototype.

Improving Team-Based Decision Making Using Data Analytics and Informatics: Protocol for a Collaborative Decision Support Design.

BACKGROUND: According to the September 2015 Institute of Medicine report, Improving Diagnosis in Health Care, each of us is likely to experience one diagnostic error in our lifetime, often with devastating consequences. Traditionally, diagnostic decision making has been the sole responsibility of an individual clinician. However, diagnosis involves an interaction among interprofessional team members with different training, skills, cultures, knowledge, and backgrounds. Moreover, diagnostic error is prevalent in the interruption-prone environment, such as the emergency department, where the loss of information may hinder a correct diagnosis. OBJECTIVE: The overall purpose of this protocol is to improve team-based diagnostic decision making by focusing on data analytics and informatics tools that improve collective information management. METHODS: To achieve this goal, we will identify the factors contributing to failures in team-based diagnostic decision making (aim 1), understand the barriers of using current health information technology tools for team collaboration (aim 2), and develop and evaluate a collaborative decision-making prototype that can improve team-based diagnostic decision making (aim 3). RESULTS: Between 2019 to 2020, we are collecting data for this study. The results are anticipated to be published between 2020 and 2021. CONCLUSIONS: The results from this study can shed light on improving diagnostic decision making by incorporating diagnostics rationale from team members. We believe a positive direction to move forward in solving diagnostic errors is by incorporating all team members, and using informatics. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/16047.

Identifying complexity in infectious diseases inpatient settings: An observation study.

BACKGROUND: Understanding complexity in healthcare has the potential to reduce decision and treatment uncertainty. Therefore, identifying both patient and task complexity may offer better task allocation and design recommendation for next-generation health information technology system design. OBJECTIVE: To identify specific complexity-contributing factors in the infectious disease domain and the relationship with the complexity perceived by clinicians. METHOD: We observed and audio recorded clinical rounds of three infectious disease teams. Thirty cases were observed for a period of four consecutive days. Transcripts were coded based on clinical complexity-contributing factors from the clinical complexity model. Ratings of complexity on day 1 for each case were collected. We then used statistical methods to identify complexity-contributing factors in relationship to perceived complexity of clinicians. RESULTS: A factor analysis (principal component extraction with varimax rotation) of specific items revealed three factors (eigenvalues>2.0) explaining 47% of total variance, namely task interaction and goals (10 items, 26%, Cronbach's Alpha=0.87), urgency and acuity (6 items, 11%, Cronbach's Alpha=0.67), and psychosocial behavior (4 items, 10%, Cronbach's alpha=0.55). A linear regression analysis showed no statistically significant association between complexity perceived by the physicians and objective complexity, which was measured from coded transcripts by three clinicians (Multiple R-squared=0.13, p=0.61). There were no physician effects on the rating of perceived complexity. CONCLUSION: Task complexity contributes significantly to overall complexity in the infectious diseases domain. The different complexity-contributing factors found in this study can guide health information technology system designers and researchers for intuitive design. Thus, decision support tools can help reduce the specific complexity-contributing factors. Future studies aimed at understanding clinical domain-specific complexity-contributing factors can ultimately improve task allocation and design for intuitive clinical reasoning.

Big-Data Based Decision-Support Systems to Improve Clinicians' Cognition.

Complex clinical decision-making could be facilitated by using population health data to inform clinicians. In two previous studies, we interviewed 16 infectious disease experts to understand complex clinical reasoning. For this study, we focused on answers from the experts on how clinical reasoning can be supported by population-based Big-Data. We found cognitive strategies such as trajectory tracking, perspective taking, and metacognition has the potential to improve clinicians' cognition to deal with complex problems. These cognitive strategies could be supported by population health data, and all have important implications for the design of Big-Data based decision-support tools that could be embedded in electronic health records. Our findings provide directions for task allocation and design of decision-support applications for health care industry development of Big data based decision-support systems.