make_decision: Make Decision

In jiyingz/clinicalMPI: The novel modified probability interval (mPI) phase I clinical trial design

Make Decision

Description

Make a decision to escalate (E), stay (S), or de-escalate (D) dose level based on the selected model. Prioritizes safety over efficacy in that if the chosen model's estimated chance of toxicity is higher than the toxicity equivalence interval, choose to de-escalate. Otherwise, choose to escalate if chosen model lies above the futility equivalence interval, and stay if it is within or below both equivalence intervals. (Assumes that DU/EU decision, if it is the one to make, has already been taken care of before hitting this function.)

Usage

intervals = make_intervals(PF, PF_tolerance, eta, PT, PT_tolerance, zeta)
pf_matrix = intervals$pf_matrix
pt_matrix = intervals$pt_matrix
selected = select_highestprob_interval(dose_info, pf_matrix, pt_matrix)
best_pf = selected$best_pf
best_pt = selected$best_pt
decision_table[f_count+1, t_count+1] = make_decision(best_pf, best_pt, pf_matrix, pt_matrix,
                                                     PF = 0.2, PF_tolerance = 0.05, PT = 0.2, PT_tolerance = 0.05)

Arguments

best_pf	Futility interval index for chosen model
best_pt	Toxicity interval index for chosen model
pf_matrix	Matrix containing start and end points of futility (pf) intervals
pt_matrix	Matrix containing start and end points of toxicity (pt) intervals
PF	Target tolerable futility probability
PF_tolerance	Half-length of futility equivalence interval (epsilon_F)
PT	Target tolerable toxicity probability
PT_tolerance	Half-length of toxicity equivalence interval (epsilon_T)

Value

String containing decision for next dose ("D", "S", "E")

jiyingz/clinicalMPI documentation built on Jan. 19, 2024, 5:58 a.m.