The incidence of breast cancer is approximately 4 per 1000 women over age 50 in the United States.
The incidence of breast cancer is approximately 4 per 1000 women over age 50 in the United States.
If we assume that, without screening, the incidence of advanced breast cancer is 4 per 1000. Introduction of screening would reduce this 40% to 2.4 per 1000. That is, 1.6 cases of advanced stage breast cancer per 1000 women would be potentially averted. In this simplified example, the expected total payoffs for screening women over 50 years old are:
For every 1000 women screened,
Expected Payoff [screen] = $225 × 1000 women + $150,000 × 2.4 cases = $825,000
Expected Payoff [not screen] = $150,000 × 4 cases = $600,000
Figure 10-3 Breast Cancer Screening Decision Tree
Although the difference is small, the expected payoff rule says that we would choose to screen women over 50. The savings from advanced cases averted compensate for the costs associated with screening.
CONCLUSION
Decision analysis provides a quantitative method for choosing among options in the face of uncertainty. It allows us to specify alternatives and estimate payoffs of decisions based on our assessment of their likelihood of occurrence. It further allows the health care manager to assess how different probabilities of future states of the world affect decision making. Decision analysis has three core component parts: the alternatives, the anticipated or forecasted state of the world that will occur after the decision is made, and the payoffs assigned to each. It is important to note, however, that in healthcare management decision making, expected payoff is not always the only criterion we use for assessing choices. It is important to consider patients’ quality of life and other values important to the community, because it is not just the decision but the consequences of that decision that are ultimately most important.
EXERCISES: 10-2 In the clinic renovation example, what if management thinks that the likelihood of current demand remaining is 30%, the likelihood of a moderate increase is 25%, and the likelihood of a large increase is 45%? What should they do, according to the expected total payoff.
Is this the question you were looking for? If so, place your order here to get started!