• Utility-based Techniques for Eliciting Patient Preferences
• Non-utility-based Techniques
• Constructing Health States for Preference Elicitation

Are used in situations when:

• There is uncertainty about the outcomes or the optimal strategy
• There are meaningful trade-offs to be made
• The choice is non-trivial

Examples of utility-based techniques:

• Decision Analysis
• Time Trade-off
• Standard Gamble
• Willingness to Pay

Decision analysis is a systematic approach to decision making under conditions of uncertainty. To perform a decision analysis one need to construct a decision model, or decision tree, that consists of:

• ALTERNATIVES to represent the available choices
• OUTCOMES that depict potential consequences of these choices
• PROBABILITIES, defining the likelihood of outcomes of each choice
• UTILITIES, or values, assigned to outcomes

A basic decision tree is created from a decision node (square) which represents the decision point. Chance nodes (circles) represent the chance events given that the choice is made. The value at the terminal node (triangle) represents the utility of ending up in the particular outcome state. Utilities can represent any type of value, such as survival or preference.

Decision Tree

A decision tree is analyzed by the process of averaging out and folding back. Probabilities for outcomes are multiplied by their utilities and summed. The choice with the highest expected utility is considered the " best" choice. In the example below, this is Choice A.

Analyzing the Tree

A decision analysis is completed with a sensitivity analysis where the important variables are systematically swept trough a range of plausible values. This helps the decision maker to determine the "robustness" of the optimal choice.

Important considerations in decision analysis:

• View the problem from the correct perspective
• Model the problem in the context of the decision
• Include the appropriate level of detail / relevance
• Model the problem over the appropriate time horizon
• Obtain correct probabilities

The time Trade-off technique measures how much time a subject is willing to trade off to avoid a specific health outcome.

How to:

A TTO begins with presenting a written description of the health state (a scenario) under evaluation. After reading the scenario, the respondent (e.g. patient) is asked to imagine a hypothetical situation in which he or she is confronted with a choice. The options available are to continue living in the state of health described in the scenario for a defined time t (such as the rest of life until age 80), or to choose alternative 2: perfect health for a time x < t (e.g. 75). Time is varied until the respondent is indifferent between the two alternatives. The TTO utility (the indifference point) is the length of remaining life in perfect health divided by the length of remaining life with the evaluated health state. For example, a respondent who is indifferent between living with the described health state for 36 more years and living with perfect health for 18 more years has a TTO utility of .5.

Other examples of TTO:

Laboratory For the Study of Patients' Preferences: An Internet Resource For Studying the Implications of Patients' Health Values Principal Investigator: Leslie A. Lenert, MD Associate Professor of Medicine, University of California: http://prefdev.ucsd.edu/demonstrations/TradeOff.htm

Health Decision Strategies: Online decision tools: http://www.healthstrategy.com/tto/tto.html

The core of the standard gamble is a paired comparison in which the respondent chooses between two alternatives with two possible outcomes: a good outcome with the probability p and a bad outcome with the probability of 1-p. The probability of outcomes is varied until the respondent is indifferent about the two alternatives. The standard gamble is directly derived from utility theory and is thought to be the criterion method for assessing preferences because of its strong axiomatic basis.

What would you choose?
Sure Thing		Gamble
Lifetime in state Y	OR
At this point, the gamble's expected utility is (.75)*(1.00) + (.25)*(0.00) = .75

How to:

The Standard Gamble begins with presenting a written description of the health state (a scenario) under evaluation. After reading the scenario, the respondent is asked to imagine a hypothetical situation in which he or she is confronted with a choice. The options available are to continue living in the state of health described in the scenario, or to take a gamble with 2 possible outcomes, e.g. perfect health and death. The probabilities in the gamble are systematically altered until the respondent cannot choose between the certainty of continued life in the described health state and the gamble. The utility for the evaluated health state = ( probability of outcome A)* (utility of outcome A) + ( probability of outcome B)* (utility of outcome B) .

Other Examples of Standard Gamble techniques:

Laboratory For the Study of Patients' Preferences: An Internet Resource For Studying the Implications of Patients' Health Values Principal Investigator: Leslie A. Lenert, MD Associate Professor of Medicine, University of California: http://prefdev.ucsd.edu/demonstrations/SG.html

Health Decision Strategies: Online decision tools: http://www.healthstrategy.com/sg/sg.html

Examples:

• Rating scales
• Probability Trade-Off Technique
• Balance scale

To be added

To be added