Risk

-Decision Analysis (note the capital letters) that is based on estimates of the probabilities of 

each state of nature. It is also based on measures of the overall desirability of each 

outcome, called its utility, so ultimately it is a ONE -criterion approach

-The structure 

of the decision making problem is represented as a decision tree

-

a)Decision Tree:is a type of model of a decision problemt is constructed according to the following conventions: 

1/Imagine a tree lying on its side, with its trunk on the left repeatedly subdividing into branches as you move to the right. The trunk represents the first (and 

sometimes the only) decision point. The tree finishes with the terminal branches on the right, which represent the full set of different possible outcomes 

(combinations of decision options and states of nature). 

2/each subdivision of the tree represents either a decision node (usually represented 

as a square) or a chance node (usually represented by a circle) 

3/branches from a decision node represent the mutually exclusive and exhaustive 

set of possible decision options at that point. 

4/branches leading out of a chance node represent the states of nature that could occur after going down the branch leading in to the node. 

-after taking the ‘no 

radical mastectomy’ branch, the patient may or may not die of cancer.

-each branch is labelled with the probability of the state of nature in question actually occurring.

-the branches leading out of a chance node are a 

mutually exclusive and exhaustive set of outcomes, and so the sum of their probabilities must add up to 1.0. 

5/each terminal branch is labelled with a utility (or disutility). This indicates the 

payoff associated with the outcome represented. 

Note:The data required are the same as for the approach to decision making under uncertainty described in Chapter 5, except that;

-probabilities of each state of nature are also needed

- payoffs or utilities will also have to be estimated, Decis analys is associated with a particular approach to this k.a. the standard gamble. 

1.We need to measure 

outcomes in a way that will allow us to see what chance of one favoured outcome our 

patient will relinquish to obtain another favoured outcome

2.When we have such a measure 

we can combine these utilities with the probabilities in a logical fashion to calculate the 

treatment with the highest expected utility. 

-One of the advantages of Decision Analysis in a clinical context is that it can clarify this 

distinction and ensure that patients’ utilities influence decisions

-In management decisions, 

however, a number of decision makers may have differing but equally legitimate, assessments 

of the probabilities and utilities involved. Sensitivity analysis may be able to show that 

some of these differences of opinion would have little impact on the best decision and 

focus attention on where better data are worth obtaining or differences of opinion have to 

be resolved.

-With goodwill, the result can be better understanding of the problem and a well-informed consensus but this may require skilful facilitation.

-multi-attribute utility analysis

if the outcome at the end of a given branch is best represented by more than one attribute e.g. both mortality rate and a morbidity rate 

-To use the tree in the usual way, this would 

involve turning a multiple criteria problem into a single-criterion one by deriving a 

utility for each combination of attribute scores

-difficulty is that:ut multi-attribute utilities are difficult to obtain in practice. Instead one can use the tree to work out how the ‘expected’ scores for each attribute in turn vary for the different decision options 

1 Graphic representation: the tree provides a model of the decision problem, with the advantages of providing an explicit basis for sharing knowledge and understanding of what is going on.

-

However, the trees can quickly become very big 

and apparently complicated; large sub-trees may be duplicated many times.

2 Incorporates estimates of risk: this is the key characteristic of the method, which is an advantage if there are reasonable estimates of risk available (as in 

a well-researched clinical problem) but arguably a disadvantage if there are not. 

3 Incorporates utilities: the separation of risk and utility is one of the strong points of the method from a theoretical point of view.

-this has practical implications in clinical decision making when, in principle at least, the health care professional can supply the tree and the probabilities and the patient can supply the utilities.

-However, different methods of eliciting utilities can give different results and there may be questions about the validity and reliability of 

the values used. Also there are questions about what to do when there are many stakeholders with different utilities. 

4 Sensitivity analysis: the fact that decision trees lend themselves readily to sensitivity analyses is a strength and in some circumstances can overcome concerns 

about the precision of the probability estimates and the validity of utilities. 

=However, while one-dimensional sensitivity analyses are straightforward to conduct and interpret, the numbers of potential multi-dimensional analyses 

(which address the effect of varying more than one factor at a time) can become very large and the results difficult to present and understand. 

5 A single criterion approach: in most examples of Decision Analysis in the literature the utility of each possible outcome is expressed as a single figure. The task is to choose the option that maximizes the expected value of this utility, so DA is mainly used as a single-criterion decision method.

-but many problems naturally involve more than one outcome criterion or attribute 

Decision conferencing:

-where stakeholders having different values or different implicit models of the situation 

-Decision conferencing provides a controlled setting that helps to focus attention on achieving the decision 

making goal.

-computer-based Decision Analysis is used during a decision workshop to represent the different perspectives of the decision makers, to explore differences of opinion and to develop new and useful 

insights.

-The aim is not to tell the decision makers what to decide but rather to help 

them to understand their beliefs, judgements and preferences in the context of the 

choices facing them. 

-process is typically supported by a facilitator and a decision analyst.

The facilitator leads the workshop, guides the discussion and builds and interprets models of the decision making problem.

The decision analyst handles the details of 

the model building. 

a) the advantages of decision conferencing are that: 

1/All stakeholders have an opportunity to contribute and to share in the ownership of and responsibility for the resulting plan of action. 

2/It facilitates communication between decision makers by focusing on critical aspects of the decision making problem and avoiding non-productive debate of 

irrelevant issues. 

3/The knowledge and creativity brought to bear on th edecision making problem are extended. 

4/Participants are able to reach a shared understanding of the nature of the problem and the reasons for the decision, even if they do not all share the same 

priorities and utilities. 

b)Its disadvantages are that: 

1/Like other methods of decision support, it is not productive in situations where 

the decision makers show little willingness to compromise and pursue a common 

interest. 

2/The time and cost involved make this method unsuitable for day-to-day management 

problems.