Study: Constructor for a Study

In scientific-computing-solutions/eventPrediction: Event Prediction in Clinical Trials with Time-to-Event Outcomes

Description

Constructor for a Study

Usage

Study(alpha, power, HR, r, N, study.duration, ctrl.median, k, acc.period,
  two.sided, shape = 1, dropout = NULL, lag.settings = NullLag())

Arguments

alpha	Significance level [0,1] (see also two-sided indicator)
power	Power [0,1]
HR	Hazard ratio to be detected
r	Control:Experimental subject balance (1:r), i.e. nE/nC=r. r=1 corresponds to equally many subjects in both arms. 2 means we have twice the number of subjects in the experimental arm.
N	Number of subjects to be recruit (integer)
study.duration	Number of months the study will be going.
ctrl.median	Median time to an event in the control group.
k	non-uniformity of accrual (integer, 1=uniform). Non-uniform accrual is allowed for using the following distribution for the probability of a patient entering the trial at time b within the accrual period [0,B]: F(b)=b_k/B_k; f(b)=k b_{k-1}/B_k where k is the measure of non-uniformity (k>0). k=1 indicates uniform accrual. This implies that during the first half of the accrual period, 1/2^k of the patients will be recruited. Half of the patients will be recruited by time B/2^{1/k}.
acc.period	Accrual time.
two.sided	If TRUE, two sided test will be used (i.e. alpha/2).
shape	The Weibull shape parameter
dropout	if subjects drop out in study (due to competing risks not as there is a finite follow up time) then this argument should contain a list with proportion and time and optionally shape i.e. dropout=list(proportion=0.03,time=12,shape=1.2) meaning in the absence of events 3 will have dropped out after 12 months with a Weibull hazard rate with shape=1.2. If shape is not included then it defaults to 1 (exponential rate). If dropout is NULL then no subjects will drop out
lag.settings	The LaggedEffect object describing any lag effect for the study

scientific-computing-solutions/eventPrediction documentation built on May 29, 2019, 3:44 p.m.