Decision analysis

Question 1

Complex decisions

Answer 1

multiple value dimensions, which may be in conflict

Question 2

Non-compensatory vs. compensatory strategies

Answer 2

1. A negative value on one attribute cannot be compensated by an equal or higher value on another attribute
2. A negative value on one attribute can be compensated by an equal or higher value on another attribute

Question 3

Non-compensatory strategies

Answer 3

Based on decision rules to shortcut or simplify the process “Elimination by Aspect” strategy (Tversky 1972)
-> Elimination of the option with the very low rate on an important attribute from the decisionprocess
“Lexicographic rule” (Svenson 1979)
-> Pick up the option with the best rate on a specific attribute

Question 4

Compensatory strategies

Answer 4

Multi Attribute Utility Theory

MAUT

Question 5

MAUT

Answer 5

Technique to support decision making when there is a limited number of available alternatives
AIM:To assist decision makers make better choices by helping them to achieve greater understanding and insight into the decision they are facing

Question 6

MAUT steps

Answer 6

Step 1. List of defining alternatives and value-relevant attributesStep 2. Evaluating each alternative separately on each attributeStep 3. Assigning relative weights to the attributesStep 4. Aggregating the weights of attributes and the single-attribute evaluations of alternatives (overall evaluation)

Question 7

MAUT step 1

Answer 7

List alternatives and value-relevant attributes. Very demanding and time consuming

• Complete
• Operational
• Decomposable
• Non redundant
• Minimal
  p. s. We can operationalize attributes as we prefer as long as the value scale is satisfied

Question 8

MAUT Step 2

Answer 8

Evaluate each alternative separately on each attribute
Different elicitation methods:
-Direct rating(assigned a score from 0 to 100)
-Difference standard sequences(a sequence of stimuli that are equally spaced in value)
- Bisection method (the most and least preferred option are identified and a midpoint is found that is equidistant in value from both extremes)

Question 9

MAUT Step 3

Answer 9

Assign relative weights to the attributes
Different elicitation methods:
-Ranking(attributes are ranked in order of importance)
-Direct rating(in which a rating from 0 to 100 is allocated to each criteria)
-Point allocation(in which a total budget of 100 points is divided among criteria

Question 10

MAUT Step 4

Answer 10

Obtain the overall evaluation
Multiply the weight * attribute value and sum these weighted attribute values over all attributes(weighted linear additive function)

Question 11

Evidence-based

Answer 11

Explicit use of modern, best evidence in making decisions about the care of individual patients
Integrates clinical experience, patient values, and the best available research information

Question 12

Patient-centered

Answer 12

Actively involves patients and incorporate patient preferences and values

Question 13

MULTI CRITERIA METHODS

Answer 13

ideal because they incorporate multiple considerations into the decision making process
Set of techniques that provides:
▪clarity on which criteria are relevant
▪the importance attached to each criteria
▪how to use this information

Question 14

MCDA modeling approaches:

Answer 14

• Value measurement models
• Outranking methods
• Reference-level modeling

Question 15

Value measurement models

Answer 15

Constructing and comparing numerical scores to identify the degree to which one decision alternative is preferred over another

Question 16

Outranking methods

Answer 16

Making pairwise comparison of alternatives on each criterion, which, in turn are then combined to obtain a measure of support of each alternative being judged the top-ranked alternative overall

Question 17

Reference-level modeling

Answer 17

Searching for the alternative that is closest to attaining predefined minimum level of performance on each criterion

Question 18

Medical decisions

Answer 18

Some are clear-cut (gold standard)

Some depend on patients’ preferences and values (no gold standard) –> Preference-sensitive treatment decisions