This was achieved by considering a limited number of patient attributes (e.g. age, sex, current treatment,...) and a limited number of possible values for each attribute (e.g. 14, 18, 35, 75 years for age), and then generating all the possible combinations of attribute values.
Using varying hypothetical utility weights for both TOX and REL health states, the overall Q-TWiST difference between groups ranged from −1.6 to 7.6 weeks and favoured panitumumab+BSC therapy for 22 of 25 (or 88% of possible value combinations) hypothetical utility levels (Table 3; Figure 3).
First, we randomly selected an equal number of simulations for each combination of every possible value of the other two parameters.
It is important to note that each performance value is not exposed to every other possible combination of performance values.
Realistic criteria were applied to narrow down the possible combination of parameter values.
For each set of parameters, a stable behavioural policy was calculated for both individuals, giving the probability for each possible combination of state values that each individual should forage or rest during a period given that it knows both its own state, and that of its co-player.
> -wrap-foot> The combination of 4 parameters with two possible values leads to 16 scenarios.
There is a 17% chance of rolling a 7 each time, because there are a total of 6 ways to get that value between the two dice, out of 36 possible value combinations between the two.
Each row of the table represents a possible combination of truth-values for the component propositions of the compound, and the number of rows is determined by the number of possible combinations.
The fourth equation encodes four counterfactuals, one for each possible combination of values for BT and ~SH.
