1. Fold back the tree below to calculate the cost-effectiveness ratio comparing “treat none” and “treat all”. Use QALY as the measure for effectiveness. Use a 5% discount rate for all future costs and benefits. Which decision is more favorable? Perform all calculations in Excel.
P1 = prevalence of disease = 0.3
P2 = probability of disease leading to blindness when it is not treated = 0.4
P3 = probability of disease leading to blindness after being treated = 0.05
P4 = probability of having side effects = 0.5
Outcome
|
Costs involved
|
Life expectancy (year)
|
Quality of life
|
Outcome 1
|
C2
|
35
|
0.6
|
Outcome 2
|
0
|
40
|
1
|
Outcome 3
|
0
|
40
|
1
|
Outcome 4
|
C1+C2+Cse
|
35
|
0.55
|
Outcome 5
|
C1+C2
|
35
|
0.6
|
Outcome 6
|
C1+Cse
|
40
|
0.95
|
Outcome 7
|
C1
|
40
|
1
|
Outcome 8
|
C1+Cse
|
40
|
0.95
|
Outcome 9
|
C1
|
40
|
1
|
C1 = cost of the one-time treatment = $50,000
C2 = annual cost of caring for blindness = $2,000
Cse = one-time cost of treating side effects = $1,000 (Side effects last for 6 months)
Quality of life adjustments:
Normal quality of life = 1
Quality of life associated with blindness = 0.6
Quality of life associated with side effects = 0.95
Quality of life associated with both blindness and side effects = 0.55
2. The table below shows the calculations for a Markov model comparing intervention to the absence of intervention. Use QALY as the measure for effectiveness. Use a 5% discount rate for all future costs and benefits. Which decision is more favorable? Perform all calculations in Excel. The table below can be copied directly into Excel.
Time
|
Intervention
|
No intervention
|
All states
|
All states
|
Remission
|
Active
|
Death
|
Remission
|
Active
|
Death
|
0
|
|
100
|
0
|
|
100
|
0
|
1
|
25
|
30
|
45
|
15
|
30
|
55
|
2
|
23
|
19
|
14
|
12
|
17
|
17
|
3
|
18
|
15
|
9
|
8
|
11
|
9
|
4
|
15
|
12
|
7
|
6
|
8
|
6
|
5
|
12
|
9
|
5
|
4
|
5
|
4
|
6
|
9
|
7
|
4
|
3
|
4
|
3
|
7
|
7
|
6
|
3
|
2
|
2
|
2
|
8
|
6
|
5
|
3
|
1
|
2
|
1
|
9
|
5
|
4
|
2
|
1
|
1
|
1
|
10
|
4
|
3
|
2
|
1
|
1
|
1
|
11
|
3
|
2
|
1
|
0
|
1
|
0
|
12
|
2
|
2
|
1
|
0
|
0
|
0
|
13
|
2
|
2
|
1
|
0
|
0
|
0
|
14
|
2
|
1
|
1
|
0
|
0
|
0
|
15
|
1
|
1
|
1
|
0
|
0
|
0
|
16
|
1
|
1
|
0
|
0
|
0
|
0
|
17
|
1
|
1
|
0
|
0
|
0
|
0
|
18
|
1
|
1
|
0
|
0
|
0
|
0
|
19
|
1
|
0
|
0
|
0
|
0
|
0
|
20
|
0
|
0
|
0
|
0
|
0
|
0
|
The length of the cycle is one year.
One-time cost of intervention = $10,000
Cost associated with “active” state = $50,000/year
Cost associated with “remission” state = $1,000/year
Cost associated with death = 0
Quality of life of “active” state = 0.4
Quality of life of “remission” state = 0.9
Quality of life of death = 0
3. For the cost-effectiveness analysis in Question 1, conduct a one-way sensitivity analysis on p1. Show your result graphically. Answer the following questions:
1) Is the conclusion sensitive to p1?
a. The conclusion is sensitive to p1
2) What is the threshold at which you conclusion will change?
a. When the threshold exceeds 0
3) What is the clinical implication of your sensitivity analysis?
4. For the cost-effectiveness analysis in Question 1, conduct a two-way sensitivity analysis on p1 and p2. Show your result graphically. Which area in the graph represents scenarios in which “Intervention” is preferred to “No intervention”?