Society for Medical Decision Making

46th Annual North American Meeting
In-Person Short Courses

___________________________________________________________

*Register for short courses with meeting registration*  Additional course registration can be made here.

Sunday, 27 October, 2024
AM Courses: 8:30am - 12:00pm

Health Literacy Tools for Decision-Making Interventions

Course Faculty: Christine Gunn, MA, PhD, Dartmouth College
Ashley Housten, OTD, MSCI, OTR/L, Washington University in St. Louis

Health literacy is a complex skillset associated with a variety of health behaviors and outcomes. Highly relevant to decision sciences, this course will provide didactic and hands-on learning in applying principles and best practices in health literacy for decision-making and educational interventions. Conceptualizations of health literacy continue evolve and our aim is to present cutting-edge research and applications of health literacy to advance research in decision sciences. As such, the overall goal of this support efforts at increasing inclusivity in decision making interventions by incorporating health literacy best practices with end-users in mind.

Attendees will learn about the breath of health literacy definitions, critically appraise conceptualizations of health literacy, evaluate and select health literacy measures that are fit-for-purpose, identify opportunities to incorporate health literacy design principles, and apply tools intended to evaluate the accessibility of informational materials used to support decision making among a variety of populations. There will be an overall emphasis on assessing tradeoffs for the practical implementation of health literacy principles in local community and clinical contexts.
By the end of this course, attendees will be able to identify a working definition of health literacy relevant to their context, evaluate approaches for assessing health literacy, and apply strategies to develop and use health literate materials to support informed decision-making.

Attendees are strongly encouraged to bring projects in development for application during the session.

Learning Objectives:

• Identify the strengths and limitations of health literacy definitions and frameworks.

• Appraise and select relevant health literacy measures for use relative to a specific project or purpose.

• Evaluate and improve the health literacy of materials designed to support health decision-making.

Course Level: Intermediate

Advanced Discrete-Event Simulations in R

Course Faculty: Thomas A. Trikalinos, MD, Brown University
Fernando Alarid-Escudero, PhD, Stanford University
Yuliia Sereda, PhD, Brown University
Stavroula A. Chrysanthopoulou, PhD, Brown University

A discrete event simulation (DES) stitches together several elemental mathematical models, each describing the occurrence of different events. For example, in a DES of a cancer’s natural history, a person may (i) develop zero, one, or more tumors; each tumor may (ii) become malignant and (iii) give symptoms; (iv) symptoms can lead to diagnosis; and death from (v) cancer or (vi) non-cancer causes may occur. Each type of events (i) to (vi), is described by a mathematical object called a point process. Thus, the central task in a DES is simulating point processes in which the rate of event occurrence varies with time.

A widely used model for point processes with time-varying intensity functions is the nonhomogeneous Poisson point process (NHPPP). NHPPPs are satisfactory models for many phenomena and thus are  commonly used in DES applications. If you can use an NHPPP to describe a point process, get a lot of flexibility: Any measurable non-negative bounded function can be used as an intensity function – so, have at it.
In this course, we will abstract DES to the level of sampling from NHPPPs, discuss how to sample from NHPPPs accurately and efficiently, and then tie everything together in a hands-on example that simulates a cancer’s natural history.

Learning Objectives:

• Just enough theory to motivate three properties of NHPPPs (memorylessness, composability, and transmutation by transforming the time axis) that are important for simulation because they enable exact sampling and justify some clever tricks.

• Algorithms to sample event times exactly, fast, and with a small memory footprint, and their implementation with the nhppp R package.

• How to do conditional simulations, such as simulating only those diagnosed with cancer at some point in their lives, by a specific age, or at a specific age.

• How to structure the code of a basic DES.

Prerequisites:

• Some familiarity with DES (what it is, why it is useful).

• Some familiarity with programming in R.

• A computer with an R installation (version >= 4.0.0), and the most recent versions of the nhppp and data.table R packages from the Comprehensive R Archive Network (CRAN), at least for those who wish to follow the exercise during the course.

• Some familiarity with basic calculus is helpful but not required.

Course Level: Advanced

Who am (A)I? Boosting Values Clarification in Shared Decision Making with Artificial Intelligence

Course Faculty: Marieke de Vries, PhD, Radboud University
Marene Dimmendaal, MSc, Radboud University

Artificial Intelligence (AI) is increasingly being implemented in medicine. This creates many opportunities, but also comes with risks and dangers. Some of these risks and dangers are already known, but for the context of values clarification in SDM, conditions for responsible AI are still to be determined. How can we leverage these technologies in such a way that patients and healthcare providers can confidently use new technologies in a safe and user-friendly manner during the values clarification process in SDM? This short course will challenge its participants to design conditions and criteria that can make this happen.

To achieve this, this interactive short course will first provide participants with a concise overview of the SDM process, of the values clarification process and of currently employed AI in medicine. Subsequently, we will discuss the current growth of AI in medicine, system designs, methods, impact and user-autonomy with the group, and the teaching team will share insights from their interdisciplinary, ongoing studies concerning this matter, which involves amongst others a decision psychologist, a gynaecologist, a philosopher and a user experience expert. Simultaneously, a design-session in which the participants will design a prototype that can support the values clarification process in SDM will be facilitated. We will end the short course with a discussion about future directions in research and clinical practice.

Learning Objectives:

• To recite key concepts of shared decision making and the values clarification process and the gap AI could fill here

• To recite examples of currently deployed AI in medicine

• To discuss the current state-of-the-art of AI in medicine and the challenges that AI faces in the context of (values clarification in) SDM

• To critically and constructively formulate ideas, concerns and challenges regarding AI that can support the values clarification process in SDM

• To design (conditions for) a prototype of an AI system that can support the values clarification process in SDM

• To define novel ideas for directions of future research and clinical practice

Course Level: Beginner

Cohort state transition modeling in R

Course Faculty: Petros Pechlivanoglou, PhD, The Hospital for Sick Children
Alexandra Moskalewicz, MSc, The Hospital for Sick Children
Fernando Alarid Escudero, PhD, Stanford University
Hawre Jalal, MD, PhD, University of Ottawa

Cohort state transition models (cSTM), often called Markov models, are frequently used to inform policy makers about how to allocate health care resources under constrained budgets. cSTM are deterministic models that simulate hypothetical cohorts as they transition across health states over time. cSTMs can be classified as either time-homogeneous (time-independent) or time-varying (time-dependent). Time-independent cSTMs have constant transition probabilities, while time-dependent cSTMs have transition probabilities that depend on the time since model start or on time spent on state residence. cSTMs can be solved with matrix algebra, which can be done in R. R is a freely available, open source programming software that can be used to implement, simulate, and analyze the results of a cohort model.

This course will teach participants how to implement cohort state-transition models (cSTMs) in R. We will first give a conceptual overview of cSTMs and the general structure for their implementation in a programming language. This will be followed by a brief review of good coding practices and how to structure your code in an efficient, transparent and reproducible way. We will demonstrate a cSTM in R, followed by hands-on exercises that will help participants implement a cSTM in R. We will also demonstrate how to conduct a cost-effectiveness analysis (CEA) and how to calculate and visually present epidemiological and health economic outcomes using the cSTM. We will shortly cover the implementation of probabilistic sensitivity analysis (PSA) of a cSTM in R. 

Throughout the course, we will highlight good programming principles.

All materials of this short course will be provided to participants after the course for future use.

Required Prerequisites:

• Construct a Cohort state transition model with time dependency since model start in R

Course Level: Beginner

_________________________________

PM Courses: 2:00 - 5:30 pm

Causal Diagrams, Target Trial Emulation and Causal Inference for Modeling and Medical Decision Making

Course Faculty: Uwe Siebert, MD, MPH, MSc, ScD, UMIT TIROL
Douglas Faries, PhD
Felicitas Kuehne, MSc, Dr.phil, UMIT TIROL

This course introduces to the principles of causation, the use of causal diagrams (directed acyclic graphs, DAGs), and methods for causal inference. We will use the framework of "target trial emulation" with counterfactual "clones" to apply causal methods such as “cloning-censoring-weighting” to big real-world observational data and pragmatic trials and link this framework with causal decision-analytic modeling.The course will consist of lectures, case examples drawn from the published literature, polls and interactive discussion.

The intended audience includes researchers from all substance matter fields, statisticians, epidemiologists, outcome researchers, health economists and health policy decision makers interested either in methods of causal analysis or causal interpretation of results based on the underlying method.

Learning Objectives:

• Define causal interventions and actions, draw and interpret causal diagrams, and apply the rules of causal diagrams to distinguish causal from non-causal statistical associations and to identify bias.

• Understand the concept of “target trial emulation” and the use of “clones” from a medical decision making perspective, and develop a target trial protocol.

• Decide which biostatistical/epidemiological methods (e.g., propensity score methods or g-methods) must be used in different situations to derive causal effect parameters.

• Perform a target trial emulation using observational real-world data.

• Describe how causal methods and approaches are applied in big observational real-world data, single arm trials, and pragmatic randomized trials.

• Understand the conceptual approaches of causal modeling.

• Derive causal modeling input parameters to feed decision-analytic models for benefit-harm assessment and health-economic evaluations in health technology assessment.

Course Level: Intermediate

Introduction to Measuring Preferences in Health

Course Faculty: Norah Crossnohere, The Ohio State University College of Medicine
Ilene Hollin, MPH, PhD, Temple University College of Public Health
Semra Ozdemir, PhD, Duke University

This course will introduce the audience to measuring preferences in health, with an emphasis on two common methods: discrete-choice experiments (DCE) and best-worst scaling (BWS). Participants will learn about the value of measuring preferences in health, be exposed to common methods for measuring preferences, and learn practical tips for designing, implementing, and analyzing preference studies.

Learning Objectives:

• Understand what preferences methods are and their appropriate use;

• Recognize and differentiate between discrete-choice experiments and best-worst scaling;

• Gain knowledge of the steps required to design a stated preference survey.

Required Prerequisites:

• Understand what preferences methods are and their appropriate use; recognize and differentiate between discrete-choice experiments and best-worst scaling; and gain knowledge of the steps required to design a stated preference survey.

Course Level: Beginner

An Introduction to Reproducible Programming and Project Management

Course Faculty: Jacob Jameson, M.S.

This beginner-friendly workshop focuses on using R and Git/GitHub to manage the full lifecycle of a scientific project, with a strong emphasis on version control and collaborative working. Participants will learn how to set up projects, manage dependencies, ensure reproducibility, and prepare code for publication, all within a collaborative framework that simulates real-world scientific teamwork. The course combines theoretical concepts with practical, hands-on exercises to build a foundational understanding of managing projects in an open-source environment.

Learning Objectives:

• Effective Project Initiation and Setup: Start projects in R using best practices for file organization and setup, integrating Git from the outset to track changes and manage contributions.

• Advanced Version Control with Git: Dive deep into Git operations such as branching, merging, resolving conflicts, and using remote repositories to facilitate collaboration among multiple contributors.

• Collaborative Workflows: Explore collaborative features of GitHub, including pull requests, code reviews, and using issues for communication, to enhance team cooperation and project transparency.

• Dependency Management: Utilize R’s package managers to handle project dependencies correctly, ensuring that your projects are portable and reproducible across different environments.

• Ensuring Code Reproducibility: Implement reproducibility checks and techniques, such as using RMarkdown and Docker, to ensure that analyses are repeatable and can be run on any computer.

• Publishing and Packaging: Learn how to package your research projects using R packages, publish your code on GitHub, and archive final versions on platforms like Zenodo or Figshare for broader dissemination.

• Interactive Exercises: Participate in group exercises that simulate a collaborative project environment, from initial setup through to publication, emphasizing real-world application of the skills learned.

Required Prerequisites:

• initiate, manage, and collaboratively execute a scientific project using R and Git/GitHub, ensuring that all aspects of the project are reproducible, well-documented, and ready for publication.

Course Level: Beginner

Advanced Computation of Value of Information Measure to Determine Optimal Research Design

Faculty: Natalia Kunst, University of York
Jeremy Goldhaber-Fiebert, Stanford Health Policy
Hawre Jalal, University of Ottawa

This course presents computation methods for the Expected Value of Sample Information (EVSI), a decision-theoretic measure of the monetary value of collecting additional information through potential future research. Participants will discuss EVSI and how it can be used to design research studies. The course will then give a demonstration of how to efficiently compute EVSI in practice with accompanying code provided in R.

Learning Objectives:

The purpose of this course is to introduce EVSI as a tool for research prioritization and study design. The course will introduce several recent methods for the calculation of EVSI alongside R code to calculate and present these measures. By the end of the course, participants will be able to:

• Distinguish four recently developed calculation methods for EVSI

• Decide which EVSI calculation method is suitable for a given health economic decision model

• Calculate EVSI in R for two different health economic models

• Present EVSI analyses using standardized, publication-quality graphics

• Discuss key assumptions for calculating the Expected Net Benefit of Sampling (ENBS)

• Design efficient future research studies by determining their optimal sample sizes

Required Prerequisites:

Participants require experience using R and knowledge of probabilistic health economic modelling and prior exposure to the Value of Information analysis. Some knowledge of Bayesian statistical methods is helpful but not required.

Course Level: Advanced

*Register for short courses with meeting registration*  Additional course registration can be made here.