OPTIONAL SHORT COURSES

Parallel Short Course 1
INTRODUCTION TO MEDICAL DECISION ANALYSIS AND
COST-EFFECTIVENESS ANALYSIS
U. Siebert, Austria

Course Level: Beginner (no prerequisites)

Course Description: Objectives:
By the end of this course, participants will

1. understand the key concepts and goals of decision analysis,
2. know the basic methods of decision tree analysis and Markov modeling and be able to choose the appropriate model type for a given research question
3. understand why and when decision-analytic modeling should be used in clinical and economic evaluation, and
4. be able to critically judge the conclusions derived from a model and know the strengths and limitations and of modelling.

Course Description:
Decision-analytic modeling is a systematic approach to decision making under uncertainty that is used widely in clinical decision making, economic evaluation, and health technology assessment.
This half day course provides an introduction into decision-analytic modeling as a tool for medical decision making and economic evaluation. The course consists of lectures and interactive group exercises and discussions.
During the course, participants will develop a basic understanding of:

• Key concepts, definitions and goals of decision analysis
• Creating the structure of a model
• Measuring health effects and costs
• Application of modeling techniques such as decision trees and Markov models
• Perform sensitivity analysis

Using practical examples from the published literature, participants will be guided through the main modeling steps, and the guidelines of the ISPOR-SMDM Joint Modeling Good Research Practices Task Force will be presented.
The intended audience includes researchers from all substance matter fields.
No laptop is needed. Please bring a simple pocket calculator!

Parallel Short Course 2
INFECTIOUS DISEASE MODELING FOR GLOBAL HEALTH INTERVENTIONS
M. Jit, UK and S. Verguet, USA

Course Level: Beginner (very basic knowledge of epidemiology and Excel is helpful)

Course Description:
Increasingly sophisticated models are being developed to capture population dynamic aspects of infectious diseases such as infection transmission, immunity and strain replacement. Many of these models are now being used to address global health issues such as prioritisation of vaccine funding and measures to mitigate the effects of an influenza pandemic. This course introduces the principles behind building dynamic transmission models of infectious diseases, and some of the ways they can be used to inform decisions about global health interventions. We begin with a hands-on tutorial on setting up a simple transmission model in Excel. The model is then used to demonstrate the way questions relevant to global health policy can be answered, such as evaluating the population-level effect of an intervention such as vaccination, estimating the health, budget impact and cost-effectiveness of an intervention, and incorporating several interventions into a single delivery platform. The appropriate use of existing infectious disease modelling tools in different contexts is also discussed.

Participants are advised to bring their own laptops so that they are able to try out the techniques that they learn.

Parallel Short Course 3
SYSTEM DYNAMICS AND FEEDBACK MODELS FOR DECISION SUPPORT
P. Einzinger, G. Zauner, N. Popper, C. Urach, F. Miksch and B. Jahn, Austria

Course Level: Beginner (no prerequisites)

Objectives:
By the end of this course, participants will 1. understand how the feedback structure of systems relates to its behavior 2. be able to formalize qualitative hypotheses about a health care problem in causal loop diagrams 3. know which aspects of a research question makes it suited for System Dynamics modeling and how System Dynamics relates to other modeling approaches 4. be able to understand stock and flow diagrams and create fully working System Dynamics models
Course Description:
System Dynamics is a modeling approach that emphasizes the importance of the global structure of a system and its consequences on the system’s behavior, especially when key variables dynamically influence themselves through feedback. These aspects are important in many health care research areas such as transmission of infectious diseases where the number of infectious people influences its own change through the risk of infection (see the report on dynamic transmission modeling by the ISPOR-SMDM Joint Modeling Good Research Practices Task Force).
This course provides an introduction into System Dynamics for applications in health care research. It addresses all steps of the modeling process from stating the research question over formulating dynamic hypotheses to creating as well as validating the model and using it for analyzing scenarios. Examples from health care applications are provided and participants learn how to use System Dynamics software.

Participants should bring their own laptop computers to the course. In addition participants should install java plug-in for the web browser, and they should install the simulation software AnyLogic from the website http://www.xjtek.com/anylogic/download/ (Educational / University Researcher Version for their specific platform) and activate a trial license. A video guide for the activation process can be found under http://www.xjtek.com/anylogic/download/video-guides/, and the license will work for 30 days.

Parallel Short Course 4
PROPENSITY SCORE ANALYSIS
H.O. Melberg, Norway

Course Level: Beginner

Course description:
This course will give a theoretical and practical introduction in how to use propensity score analysis to measure causal effects based on observational data. Propensity score analysis is a statistical method that adjusts for selection bias that arises in observational data when the respondents are not randomly selected into treatments. Examples include the effect of college education on wages and the effect of moderate alcohol consumption on mortality. Propensity score analysis is one method that can be used to reduce the problem of selection bias in these and similar examples.

Parallel Short Course 5
WHY DO PHYSICIANS NOT MAKE RATIONAL, EVIDENCE-BASED DECISIONS?
PART 1
R.M. Poses and W.R. Smith, USA
Please note that this is a full day course. Part II is Course Number 13 in the afternoon

Course Level: Beginner (knowledge of rudiments of the concepts of evidence-based medicine is helpful).

Course Description:
The principles of evidence-based medicine (EBM) and medical decision making call for physicians to make decisions based on the best available evidence from clinical research that maximize individual patients’ benefits and minimize their harms, according to the patients’ values. Clinical practice guidelines (CPGs) are advocated both to foster such decision making, and to serve as a standard for it. However, evidence suggests that physicians rarely make decisions according to CPGs and many attempts to improve physicians’ decisions have failed.
We will review evidence that physicians fail to follow clinical practice guidelines and otherwise fail to make decisions in accord with the best available evidence. We will then discuss “traditional” theories as to why this may be so. In particular, we will discuss how physicians’ human cognitive limitations foster the use of cognitive heuristics and allow the influence of cognitive biases, and how these biases and heuristics, in turn, may affect their decision making.
To illuminate other barriers to optimal decision making, we will then focus on a case study: the recognition and treatment of depression in primary care. We will discuss “traditional” barriers to adherence to CPGs for this problem and review the latest clinical evidence to develop alternative hypotheses about why physicians fail to follow them.
This will lead to a wider consideration of possible barriers to ideal decision making, including, in some cases, the very CPGs advocated to improve decisions, but which actually may be untrustworthy, and hence should not inspire adherence. We will discuss insidious external influences that may impede ideal decision making, including those intentionally designed to further vested interests: manipulations of the design, implementation, analysis, and dissemination of clinical studies, outright suppression of research, manipulation of education, especially involving stealth marketing, deception used in conventional advertising and marketing, and perverse incentives.
Finally, we will propose some possible solutions, based on both findings from cognitive psychology and review of recent Institute of Medicine reports on conflicts of interest in medicine, and development of trustworthy guidelines. We will finish with a brain-storming session to develop new research ideas.

Parallel Short Course 6
PARAMETRIC SURVIVAL MODELS FOR HEALTH ECONOMIC
EVALUATION: PART 1
J. Lewsey, UK
Please note that this is a full day course. Part II is Course Number 7 in the afternoon

Course Level: Intermediate

Course Description- Background:
The semiparametric Cox model is a popular statistical method to assess the effectiveness of treatments/interventions for time to event outcomes and can also play a role in health economic models. However, it is not specified in a Cox model how the risk of an event changes over time and so its use in time dependent Markov models is limited. Parametric survival models are a class of models that overcome this limitation by specifying a distribution for the time to event.
Course description and objectives:
This course will introduce the key components of parametric survival models with the aid of health economic examples. Specific objectives are:

1. To review the Cox model leading onto the introduction of parametric models
2. To demonstrate how to fit, interpret and choose between parametric survival models using data sets with health economic settings
3. To demonstrate how parametric survival models can be used to predict transition probabilities beyond observed follow-up periods
4. To illustrate how parametric survival models can be applied to the Kaplan-Meier sample-average estimator approach for estimating future costs and morbidity

Format and requirements:
The course will be a mixture of informal teaching and “hands on” computer exercises using the STATA statistics package (an understanding of STATA would be an advantage but is not a pre-requisite). A participant is likely to gain more from the course if they have some knowledge of survival analysis (e.g. Kaplan-Meier, Cox). Participants are required to bring their own laptops.

Parallel Short Course 7
PARAMETRIC SURVIVAL MODELS FOR HEALTH ECONOMIC
EVALUATION: PART 2
J. Lewsey, UK

Course Level: Intermediate

Please note that participation in the morning session (Course 6) is a pre-requisite for participation in Course 7 in the afternoon.

Parallel Short Course 8
INTRODUCTION TO SHARED DECISION MAKING AND DECISION AIDS
A. Stiggelbout, The Netherlands, D. Alden, USA and
Marilyn Schapira, Department of Medicine, University of Pennsylvania, Philadelphia, PA, USA

Course Level: Beginner

The format will involve both didactic lecture and group exercises and discussions. There are no prerequisites for the course. The intended audience includes researchers and practitioners interested in shared decision making and/or in using and designing patient decision aids.
Description and Objectives: This introductory course will first provide a broad overview of professional-patient decision-making models, the elements contained in the various proposed models, and the position of shared decision making (SDM) in health care. SDM is a general approach to health care decision making, which may be facilitated by the use of patient decision aids (PtDAs) or decision support tools. The course will touch briefly upon the International Patient Decision Aid Standards (IPDAS) and will review findings from the literature on the evaluation of PtDAs. Next it will discuss the basics to guide a systematic design of PtDAs. Design steps include deciding on what theory to guide the development, what components to include in the aid, what information to include and how to present it.
Finally, cultural tailoring issues regarding wording of risk statements and value/lifestyle statements within decision aids will be discussed.
Objectives:
By the end of the course, participants will:

1. Appreciate the difference between the extant patient-professional decision models.
2. Understand the elements of the shared decision making model.
3. Appreciate the evidence on the impact of patient decision aids.
4. Understand basic steps to consider when designing patient decision aids.
5. Be aware of the sensitivity of decision aids to cultural issues.

Parallel Short Course 9
ETHICAL AND LEGAL ISSUES OF INFECTIOUS DISEASE MANAGEMENT-
LESSONS FROM JUDICIAL REVIEW COURSE
V. Stühlinger and M. Flatscher-Thöni, Austria

Course Level: Intermediate

Description:
Infectious diseases can spread directly or indirectly from one person to another. In order to prevent spread of infectious diseases public health authorities have to follow a certain strategy, using public health measures. In an increasingly linked global society such public health measures are of central importance not only in a national context but also in an international context. Public health measures – especially in pandemic scenarios – can be intrusive and potentially interfere with human rights if they are not justified, disproportional or unjustifiably cause damages. Thus, measures applied by public authorities should be scientifically proven effective and – under the rule of law – have to be as foreseeable as possible and have to stand up to subsequent judicial review. In recent years courts increasingly dealt with ethical and legal issues in the context of public health decision making managing infectious diseases. Given this development the short course at hand is in a first step aiming at clarifying the underlying ethical and legal framework by discussing key human rights principles in the context of infectious disease management and their implications on public health decision making. In a second step we will present and discuss specific case law dealing with issues in connection with infectious disease management. Summing up, this short course is aiming at providing participants with a fundamental understanding of ethical and legal principles in the context of public health decision making, focusing on managing infectious diseases and at analyzing the lessons we can learn from judicial review for future decisions.

Parallel Short Course 10
AGENT BASED TRANSMISSION MODELS FOR INFECTIOUS DISEASES
F. Miksch, C. Urach, N. Popper, G. Zauner and P. Einzinger, Austria

Course Level: Beginner

Objectives:
By the end of this course, participants will 1. understand the concepts of transmission based epidemic models, 2. be able understand and plan an agent based epidemic model, 3. be able to choose parameters for a given model structure 4. to interpret the results for decision making.
Course description:
Infectious diseases are spread by transmissions; therefore it makes sense to take a closer look at the transmissions, leading to dynamic epidemic models.
Concepts of models that deal with single persons, especially agent based models are a practical approach to simulate such systems. These dynamic transmission models are also subject of interest in the ISPOR-SMDM Modeling Good Research Practices Task Force Working Group, as they are state of the art in advanced dynamic epidemic modeling.
In the first part we explain the main concepts of agent based epidemic models so that participants understand their usability and flexibility. That explanation includes an introduction into a simple model structure and tips for programming.
In the second part a framework for agent based epidemic modeling is provided. Participants are guided through practical examples within that framework that show representative situations and give an idea of epidemic dynamics and characteristics of such an approach.
In the third part we will analyze the results in detail and discuss how the outcomes might be interpreted. Further variations and extensions of the approach are presented to show that complex and detailed real world problems can be simulated with agent based epidemic models.

Participants should bring their own laptop computers to the course.

Parallel Short Course 11
DISCRETE EVENT SIMULATION
B. Jahn and U. Rochau, Austria

Course Level: Beginner

Course Description:
This half day course provides an introduction into Discrete Event Simulation (DES) as a tool for clinical and economic decision analysis as well as for management optimization. The course consists of lectures and interactive hands-on activities. It will also consider the new results and best practice recommendations of the ISPOR-SMDM Joint Modeling Good Research Practice Task Force.
Whereas DES has been successfully applied in industrial engineering since 1960s, it has now increasingly been applied to evaluate various health care technologies (e.g., in cancer, cardiovascular diseases, infectious diseases).
DES is a microsimulation method that allows modelling on the individual (e.g., patient) level. Patients can interact with each other or the health care system and compete for resources (e.g., hospital beds or organ transplants). These Resources can be modelled explicitly. In a DES, time can be managed flexible, instead of using fixed time intervals. An event that can be anything that happens during a simulation (e.g. start of therapy, admission to hospital, change in dose, adverse event, etc.) can occur at any points in time.
Participants will develop a basic understanding of the key concepts of DES (entities, attributes, events, resources and queues). The course provides insight into model application considering recommendations of the ISPOR-SMDM Modeling Task Force. It starts with an introduction to decision modeling.
Based on practical examples participants will be guided through the main modeling steps. The course will combine lectures and hands-on activities.
Models will be constructed using ARENA(TM) or other software. No previous knowledge of is required.
Participants should bring their own lap top computers.

Parallel Short Course 12
COST-VALUE ANALYSIS: INCORPORATING CONCERNS FOR FAIRNESS IN
PUBLIC WILLINGNESS TO PAY FOR HEALTH INTERVENTIONS
E. Nord, Norway

Course Level: Intermediate
(Issues will be at the research front, but the form of presentation will be simple and free of equations.)

Course description:
In cost-value analysis, costs are compared with benefits valued from societal decision makers’ perspective rather than in terms of individual utility. The societal perspective includes concerns for fair distribution. Two main fairness concerns are (a) giving priority to the worse off and (b) not discriminating too strongly against those who have lesser potentials for health. The main aims of the course are to document the strength of such concerns in a number of jurisdictions, to show how they can be represented in terms of values for health states that have a different structure from those used in conventional QALYs and to demonstrate how limits to public willingness to pay for different health interventions and programs can be estimated directly from such alternative health state values. The course will furthermore address discounting of health benefits, which in cost-value analysis has a weaker justification than it has in cost-utility analysis, and the treatment of production gains and non-health consumption costs in evaluation of gained life years. In general, issues will be at the research front, but the form of presentation will be simple and free of equations.

Parallel Short Course 13
WHY DO PHYSICIANS NOT MAKE RATIONAL, EVIDENCE-BASED DECISIONS?
PART 2
R.M. Poses and W.R. Smith, USA

Course Level: Beginner (knowledge of rudiments of the concepts of evidence-based medicine is helpful).

Please note that participation in the morning session (Course 5) is a pre-requisite for participation in Course 13 in the afternoon.