AI in Pharmaceutical Outcomes and Policy Research


Artificial intelligence (AI) is the method by which a computer is able to act on data through statistical analysis, enabling it to understand, analyze, and learn from data through specifically designed algorithms. AI has become increasingly prevalent in various sectors of society, including pharmacy and related fields.

Applications of AI methods in POP research include:

  • Enhancement in cohort identification and measurement: using structured and unstructured healthcare data and electronic medical records to improve efficiency and accuracy of identifying patient cohorts and measuring relevant exposures, outcomes, and covariates.
  • Model advancement: applications of AI modeling approaches to handle high-dimensional data and enhance design-driven causal inference studies.
  • Exploratory analysis of drug effects and treatment outcome heterogeneity: signal detection of drug outcome relationships and drug effect modifiers to inform pharmaceutical outcomes research using real-world data.
  • Outcome prediction and forecasting: development and validation of prediction models to provide clinical and policy decision support.
  • Bias mitigation and fairness evaluation: assessment of AI approaches to reduce potential biases and improve effectiveness, safety, fairness, trustworthiness, equity, and interpretability of AI/ML models in improving drug risk-benefit and healthcare access.
  • Prescriptive decision analytic models: development and evaluation of potential interventions that can be offered in conjunction with prediction models in decision support applications.
  • Data mining and information extraction: utilizing state-of-the-art natural language processing methods to identify disease, medications, and symptoms from unstructured clinical data, as well as measuring cognitive ability and exploring social determinants of health.
  • Other areas: data visualization and dashboard generation of real-time data to inform policy and clinical practice, and AI method development to enhance POP research.

Our goals in this focus area include employing and tailoring novel AI methodology on automated healthcare data to answer important questions about clinical outcomes of drug effects, other healthcare interventions, and the policies governing medication use.

This focus area supports the three specialization areas and is aligned with the UF Quality Enhancement Plan (QEP), which places AI as the centerpiece of a major, long-term initiative, combining world-class research infrastructure and a cutting-edge research & transformational approach to a curriculum through five domains: (1) interdisciplinary excellence, (2) unifying force of AI, (3) encourage AI expansion, (4) strength in applications of AI and (5) Incorporate ethics throughout.

AI Focus Area Curriculum

AI Introductory courses:

  • GMS 6850 Foundations of Biomedical Informatics
  • PHA 6935 Introduction to Artificial Intelligence in Pharmacy

AI intermediate-level, hands-on courses:

  • PHA 6935 AI-supported prediction in pharmaceutical outcomes and policy research
  • BME 6938 Biomedical Data Science

Career Profiles

Focusing on the real-world requirements in AI in Pharmaceutical Outcomes and Policy Research, this focus area provides students with the knowledge and skill set in machine learning, data mining, measurement, causal inference, prediction models, bias mitigation, and decision support systems

Track Faculty