Adaptive Decision Making Based on Interval-Censored Data in a Clinical Trial to Optimize Rapid Treatment of Stroke - Interval-Censored-Time-to-Event Data: Methods and Applications - page 339

TABLE 12.3: Simulation Results for the IA tPA Trial. Under each scenario,
u true (c;q) = true mean utility of combination (c;q); and the highest utility
is highlighted in bold. Utilities of unacceptable (c;q) combinations that are
too toxic or have low ecacy have a gray background. Sel = % selected, N =
number of patients.
Scenario q c=0.2 c=0.3 c=0.4 c=0.5 %None
1 0.1u true (c;q) 49.0 54.9 62.4 71.5 2
Safe, Sel(N) 1%(2.6) 1%(1.2) 4%(2.3) 17%(5.9)
highcand0.2u true (c;q) 52.6 58.4 65.9 73.8
q=.2best Sel(N) 1%(0.5) 1%(0.6) 13%(5.3) 60%(16.9)
2 0.1u true (c;q) 61.1 58.7 51.6 48.0 12
Safe, Sel(N) 43%(14.4) 7%(3.3) 6%(3.4) 5%(2.8)
lowcand 0.2u true (c;q) 58.2 53.9 49.5 45.0
q=.1best Sel(N) 22%(5.0) 3%(1.5) 2%(2.2) 1%(0.9)
3 0.1u true (c;q) 44.8 45.2 45.2 45.0 91
Unsafe Sel(N) 4%(6.5) 1%(1.5) 1%(2.0) 1%(2.2)
0.2u true (c;q) 45.2 45.2 45.0 44.3
Sel(N) 1%(0.9) 0%(0.4) 0%(0.8) 0%(0.6)
4 0.1u true (c;q) 38.2 40.0 41.9 43.3 83
Safe,but Sel(N) 0%(2.8) 0%(1.1) 1%(1.6) 7%(5.2)
no(c;q) 0.2u true (c;q) 39.3 41.2 43.1 44.4
acceptable Sel(N) 0%(0.4) 0%(0.4) 1%(1.4) 7%(4.3)