Abstract
The objective of this paper is to explore and propose how pharmacists can leverage precision medicine to achieve optimal pharmacotherapy and how artificial intelligence (AI) can guide and enhance these efforts across acute care, chronic disease management, and rare diseases. A PubMed literature search was performed (2020-2025) using the following search terms: precision medicine, artificial intelligence, machine learning, predictive analytics, prescriptive analytics, and pharmacogenomics. Abstracts from article bibliographies were reviewed; all relevant English-language studies or in-depth clinical and technical reviews were considered. There were several key findings in this narrative literature review. First, the review emphasizes the importance of incorporating pharmacogenomics into a broader multi-omics evaluation to guide safe and effective therapy. Second, it includes the demonstrated versus emerging applications of AI, clarifies how AI complements genetic testing and specifies pharmacist responsibilities in clinical, operational, and ethical domains. Finally, it highlights the value of leveraging AI, specifically predictive analytics, and machine learning to support and refine pharmacotherapy evaluations. In cases where multi-omics testing is not accessible, the utilization of big data is proposed as a viable alternative for clinical decision support. Given the complexities and nuances of tailored pharmacotherapy, leveraging advanced strategies such as pharmacogenomics and AI can provide pharmacists with valuable tools. As such,
Introduction
The practice of pharmacy is both a science and an art, blending evidence-based knowledge with compassionate care. Years of experience often define a clinician’s ability to recommend optimal pharmacotherapy. In the absence of extensive experience or in the case of treating rare diseases, where experience and research are often scarce, it may be especially challenging to identify the most appropriate therapy. Further, with an aging and increasingly medically complex patient population across the United States, it is crucial to tailor pharmacotherapy recommendations to meet patient-specific needs. 1 Traditionally, clinical guidelines for disease prevention and treatment have been developed based on the expected responses of an average patient; while effective in certain contexts, this approach often fails to account for interindividual variability. 2 Precision medicine seeks to tailor therapies by integrating patient specific factors including clinical history, social determinants, and genomics into treatment decisions. This article reviews how pharmacists can leverage precision medicine to achieve optimal pharmacotherapy and how artificial intelligence (AI), including machine learning and data analytics, can guide and enhance these efforts across acute care, chronic care, and rare disease management. It outlines the demonstrated versus emerging applications of AI, clarifies how AI complements genetic testing, and specifies pharmacist responsibilities in clinical, operational, and ethical domains, including the limitations and risks that must be addressed for safe implementation.
Current State of Precision Medicine: Emphasis on Pharmacogenomics
Precision medicine signifies a pivotal transformation in healthcare delivery, transitioning from broad (one size fits all) guideline recommendations to customized therapies for patients. 3 To date, precision medicine has been mostly enabled by pharmacogenomic testing, by providing insights into how a patient’s genetic composition affects their pharmaceutical responses (“pharmacogenomics”). 4 Despite impressive progress, precision medicine still faces significant and unique challenges in the acute care setting as well as chronic and rare disease management.
Chronic conditions like diabetes and heart disease affect many people, yet each patient’s journey is unique, requiring care that considers their personal genetic makeup and how their body responds to treatments. A well-published example of pharmacogenomics is the use of patient genotyping to guide warfarin therapy. Genotype-guided warfarin dosing illustrates how algorithms can combine pharmacogenomic and clinical variables to inform initial dose selection. While the “clinical art” of ongoing international normalized ratio (INR) titrating remains essential and direct oral anticoagulant (DOAC) adoption has reduced warfarin’s market share, this example demonstrates how pharmacists can implement protocolized, data-informed dosing, educate clinicians and patients, and monitor for safety signals. However, as in the case of warfarin dosing, it is well recognized that environmental factors (eg, diet, concurrent medications) also play a role in guiding optimal therapy. 5
For rare diseases, the hurdles can be even greater due to lack of insights on how to pursue the necessary omics testing and/or insufficient data or treatment options to support true personalized care. Further, a recent study demonstrated that pharmacist participation, especially in the acute care setting, can improve adoption of pharmacogenomics practices, which highlights an area where the pharmacy workforce can provide improved support. 6
While pharmacogenomics is a cornerstone of precision medicine, its implementation remains limited due to cost, infrastructure requirements, and clinical interpretation and application capabilities. 7 A practical barrier is the relatively small set of gene-drug pairs with strong implementable guidance and inconsistent incentives for broad preemptive testing. In this context, pharmacists help translate results into care plans and can champion prescriptive analytics that combine genomics with clinical variables to yield actionable dose or therapy recommendations, especially where genotyping is unavailable or results are indeterminate.
Precision Medicine Supported by AI Enabled Data Analytics
Examples of Publications Describing the Use of Precision Medicine and Artificial Intelligence in the Acute Care, Chronic Disease Management, and Rare Disease Management
Examples of Publications Demonstrating the Role of the Pharmacist in Precision Medicine, Predictive Analytics and Artificial Intelligence
AI is a discipline within the field of Computer Science that aims to mimic human intelligence. Machine learning (ML), a subfield of AI, aims at identifying trends, forecasting outcomes, and decision support through processing and analysis of an extensive amount of data. 11 An application of AI in precision medicine includes the rapid analysis of omic results for a patient, alongside other available data, to provide specific therapeutic recommendations. AI improves this domain, for example, by swiftly evaluating millions of gene-drug interactions in minutes and suggesting the best treatment options. 12 Multi-omic results encompass data from various fields such as genomics, proteomics, transcriptomics, metabolomics, and epigenomics, which study biological systems comprehensively. When these multi-omic results are integrated with AI, it helps analyze complex datasets to identify patterns, predict outcomes, and guide personalized treatments based on multiple layers of patient data. Although their prospective performance and generalizability require further validation, emerging models may assist with treatment selection by synthesizing clinical, laboratory, medication, and social factors. This could enable pharmacists to move beyond the limited or unidimensional data points and instead utilize vast, integrated datasets to support real-time, multidimensional evidence-driven decisions. ML enables a transition from relying on user-generated programming for data analysis and interpretation to enable computer systems to learn independently from data without human intervention. In essence, ML can analyze extensive healthcare datasets to recognize patterns and acquire its own insights, rather than depending solely on human expertise. This is particularly valuable in precision medicine, where incorporating numerous patient-specific factors and deep experience is a critical, yet very complex process. ML algorithms that predict disease progression and generative AI systems that produce tailored clinical guidance can expedite advancements in precision medicine and enhance patient care. One significant example is the use of AI in the absence of omic data, such as automated speech analysis and detection of movement disorders, cognitive impairment, and mental disorders to enhance the early diagnosis and facilitate timely treatment of several disease states (eg, Alzheimer’s disease, Parkinson’s disease). 13
Pharmacists are distinctly positioned to assume a pivotal role in the execution of AI-driven precision medicine. Their proficiency in pharmacology, patient education, and medication safety renders them subject matter experts who can convert AI-generated data into actionable insights. Despite AI models’ ability to provide “actionable” suggestions, pharmacists’ oversight is crucial to contextualize recommendations, resolve conflicts with guidelines or formulary constraints, manage drug-drug and organ function considerations beyond model scope, ensure equity and safety where training data are biased, and communicate risks and benefits to prescribers. These tasks convert the model output into clinically accountable care, thus serving as a pharmacist extender to allow time for more patient interaction and counseling. Further, pharmacists’ presence in hospitals, ambulatory care clinics, retail pharmacies, and the pharmaceutical industry enables them to apply these therapeutic advancement strategies across many patient care settings. 14 With this said, not all pharmacists feel adequately trained in the science of AI and ML, and therefore, hesitancy to pursue such applications may exist. The American Society of Health-System Pharmacists Statement on the Use of AI in Pharmacy Practice describes several areas where the pharmacy workforce can gain experience and support these technologies across several domains. 14 With greater experience, pharmacists can operationalize AI outputs by validating model recommendations against clinical context, counseling patients on risks and benefits, coordinating with prescribers for therapy changes, monitoring outcomes and safety signals, and contributing to protocol development. Overall, expertise in pharmacogenomics and data analytics enables pharmacists to optimize precision medicine.
AI and Data Analytics in the Management of Chronic Diseases
The primary scalable contribution of AI to precision medicine is its prediction skills. ML algorithms can comb through many electronic health records (EHRs) to detect people at elevated risk for chronic diseases such as diabetes, hypertension, and heart failure.15,16 Furthermore, AI can forecast adverse drug reactions (ADRs), hospital readmissions, and non-adherence, facilitating earlier management. Pharmacists can integrate risk scores into medication therapy management (MTM) to schedule targeted follow ups, reconcile interacting regimens that elevate modeled adverse drug event risk, tailor adherence strategies for patients flagged as high risk and escalate care or initiate protocolized therapy adjustments in collaboration with prescribers when models indicate accelerating disease trajectories. 16 In community and ambulatory care pharmacy contexts, predictive algorithms can ascertain patients predisposed to polypharmacy issues, allowing pharmacists to optimize therapy and enhance adherence. 11 Pharmacists frequently leverage descriptive analytics, the approach of using retrospective data to gain historical perspective on appropriateness of therapy (eg, medication use evaluation), to develop protocols or inform future recommendations. The next step in the data analytics maturity curve, predictive analytics, is demonstrated through clinical publications where we may utilize retrospective data to create logistic regression models and further predict which risk factors and/or therapies may align with a given patient population.17,18 By collaborating with data science colleagues, pharmacists can help inform predictive analytics models and simulations that could support precision medicine and medication safety efforts. To achieve optimal foresight, AI platforms are producing prescriptive analytics, for which data feeds are analyzed in near-real time to predict outputs in an agile manner that is not anchored in the past and furthermore, convert such predictions to actionable recommendations. This is preferred over predictive models, which are trained on historical data and as a result, can perpetuate existing biases and may fall into novel scenarios unless actively monitored and recalibrated. Across the various analytics spectrum, pharmacists offer unique insights to maximize such programs and enhance precision medicine capabilities. 17
AI and Predictive Analytics in the Management of Acute Care Illness
Beyond chronic disease management, AI and predictive analytics capabilities in the acute care setting are rapidly expanding. Given the high acuity and potential rapid deterioration in this setting, there is a significant need for enhanced capabilities to more efficiently diagnose and treat acute disease. 19 AI-assisted tools are being explored for early sepsis recognition, clinical deterioration alerts and intensive care unit (ICU) medication optimization. A study published by Li X et al describes a logistic regression machine learning model with the ability to predict hemorrhagic transformation after intravenous thrombolytic therapy in acute ischemic stroke patients, which can support treatment decision-making in these urgent cases. 20 Further, machine learning methods have been designed to identify and impute missing patient data and clinical values which are crucial for clinical decision-making when time is of the essence. 21 Pharmacists contribute to triaging algorithm alerts for medication-related etiologies, recommending time-critical therapy adjustments, validating clinical data, and participating in multidisciplinary response pathways and stewardship programs. Another emerging area of interest is the use of AI to support antimicrobial dosing and therapeutic drug monitoring (TDM), particularly for agents requiring individualized dosing based on renal function, severity of illness, pharmacokinetic variability, and evolving clinical response.22,23 AI-driven models may help synthesize laboratory trends, patient characteristics, and prior dosing responses to optimize antimicrobial therapy. However, these applications remain dependent on data quality, prospective validation, and appropriate clinical oversight. In practice, pharmacist involvement remains critical to ensure that model outputs align with antimicrobial stewardship principles, toxicity monitoring, site-specific resistance patterns, and overall patient context.
AI and Data Analytics in Rare Disease Detection and Management
Diagnoses of rare diseases are often delayed due to their uncommon presentations and low prevalence. AI can explore vast amounts of data to identify nuanced symptomatic patterns that may escape the human eye.24,25 In pediatrics, AI has been employed to identify atypical neurodevelopmental patterns and facilitate early referrals to specialists. 26 For pharmacists in specialty pharmacies or hospital teams, early identification is essential for initiating prompt therapies and enhancing outcomes.27,28 In the case of selecting treatment for rare diseases, it is especially challenging considering the limited evidence available for most of them. A recent study described a novel use of AI in identifying new therapeutic applications from existing medicines; this approach led to repurposing of existing medicines that are candidates for more than 17,000 diseases. 29 While diagnosis is the remit of clinicians, pharmacists add value after diagnosis by coordinated access to orphan therapies, counseling on complex regimens and adverse effect management, navigating financial and distribution barriers, and contributing to clinical trial operations. In drug repurposing contexts, pharmacists support literature appraisal, feasibility assessments and protocol development, recognizing that clinical studies are required before practice can change.13,27
Challenges Related to Ethics, Equity, and Implementation
Although AI offers substantial opportunities, it also introduces ethical and practical challenges. Algorithmic bias is a well-known issue, as models trained on non-representative datasets may produce inaccurate or detrimental suggestions for underrepresented groups. 30 Pharmacists must continually assess AI technologies, promote transparency, and ensure that algorithms equitably serve various patient populations. Furthermore, data privacy constitutes another significant concern. The utilization of AI necessitates substantial quantities of patient data, raising concerns regarding confidentiality, informed consent, and data ownership.31,32 In addition to concerns related to bias and data privacy, AI-supported pharmacotherapy introduces the possibility of inaccurate or fabricated outputs, including incorrect citations, unsupported recommendations, and misinterpretations of clinical information.33,34 These risks are particularly important when AI-generated suggestions are accepted without appropriate clinical validation. Unlike trained clinicians, AI systems may fail to fully account for patient-specific variables such as allergy history, organ dysfunction, contraindications, local formulary restrictions, or social context. Additionally, AI systems may not fully capture clinical nuances such as off-label medication use, evolving guideline interpretations, or individualized risk-benefit considerations that often influence real-world therapeutic decision-making. As a result, AI should function as a clinical decision-support tool rather than an autonomous decision-maker, with pharmacist oversight remaining essential to ensure safe and appropriate application. AI-generated recommendations may also introduce safety concerns if they fail to identify contraindications, drug-drug interactions, or patient-specific risk factors for adverse drug reactions. In such cases, accountability remains with healthcare professionals applying these tools. Pharmacist oversight is therefore essential to validate outputs, identify unsafe recommendations, and ensure that therapeutic decisions remain patient-specific and clinically appropriate. Disparities in access further hamper the deployment of AI. 35 Rural and under-resourced healthcare environments may lack the infrastructure necessary for implementing advanced AI capabilities. In these settings, pharmacists can serve as key advocates for scalable, resource-efficient solutions that provide meaningful clinical benefit. 36 An additional concern is the potential for overreliance on AI systems, which may diminish independent clinical reasoning over time if outputs are accepted without appropriate critical evaluation.
Equipping Pharmacists for AI Integration and Advanced Data Analytics
AI contributes along three main categories of clinical pharmacy practice: descriptive analytics to characterize current therapy quality and safety, predictive analytics to estimate risks such as hospitalization, adverse drug events or non-adherence, and prescriptive analytics that translate predictions into suggested actions such as dose, drug choice and timing. Pharmacists may engage across all tiers by validating model fit, operationalizing recommendations, and monitoring downstream outcomes. To fully leverage the potential of AI in precision medicine, pharmacy education must integrate core concepts related to these capabilities, including data literacy, ethical considerations, and hands-on experience with digital health tools. In addition, pharmacy education should also emphasize that AI functions as a clinical support tool rather than a replacement for clinical reasoning. Training should reinforce critical appraisal of AI-generated outputs, ensuring that students develop the ability to evaluate recommendations within the context of patient specific factors, clinical guidelines, and real-world constraints. Incorporating case-based learning, supervised use of AI-driven tools, and discussion of system limitations may help prevent overreliance while preserving independent clinical judgment. Furthermore, continuing education programs and certifications in AI-enabled healthcare can significantly enhance the ability of practicing pharmacists to assume leadership roles within this field. Interdisciplinary collaboration is essential; pharmacists must collaborate with data scientists, physicians, nurses, administrators, and policymakers to guarantee that AI tools are therapeutically pertinent and ethically robust. Pharmacists are also poised to spearhead organizational transformation. Their participation in formulary management, clinical decision support systems, quality improvement, and medication safety activities positions them as optimal advocates for AI integration. By assuming these leadership positions, pharmacists can influence the prudent application of AI in healthcare, especially in the practice of precision medicine. 37 Further guidance for equipping the pharmacy workforce to support AI applications has been published by ASHP. 13
Conclusion
Artificial intelligence enhances the accessibility, applicability, and efficacy of precision medicine. Pharmacogenomics is well-established but combining it with multi-omic evaluations offers an exciting new frontier. By integrating ML and AI-predictive analytics, pharmacists can better predict outcomes and tailor care, especially in chronic and rare disease management. Pharmacists, as reliable healthcare professionals, must persist in adopting innovation while maintaining ethical standards and promoting equal access. By doing so, they will not only improve patient outcomes but also reinforce their position at the forefront of precision medicine movement. By focusing on chronic disease management and the identification of rare diseases, pharmacists can provide personalized treatment options to a broader population. This approach facilitates the realization of precision medicine in a scalable and equitable manner.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of Conflicting Interests
SSB and AT are employees and shareholders of Becton, Dickinson and Company. RWK and AQ declare no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.