Subsequently, testing of proposed samples is performed and material parameters are evaluated.
A total of 95 observations were conducted at baseline (out of a proposed sample size of 90) while 96 observations were conducted at endline.
As of August 2004, over 1000 patients had been recruited to this trial of a proposed sample size of 1800.
With the proposed sample size of 210 for each of the two groups (males/females), the study would have 80.3% power to yield a statistically significant result.
Given the exploratory nature of the trial, we estimate that the proposed sample size of 80 is a realistic estimate based on the nature of the intervention and the possible size of the study population, as well as what is feasible within the one year timeframe for the nature of the Phase II element under the MRC Framework.
The proposed sample size of 675 infants in each arm of the trial was estimated for the primary outcome of weight gain at 21 days post-enrollment, as measured by the change in z-score (z-score at Day 21 – z-score at enrollment), using preliminary data from the FICare pilot study available at the time of trial design.
Table 2 shows the power we would have with the proposed sample size of 96 clinicians and an equal number of patients, and also illustrates the limited impact of increasing the number of patients from 1 1 to 1 3.
Assuming a 20% dropout rate at week 26, a proposed sample size of 2138 randomised patients (475 in each of the QVA149, indacaterol, glycopyrronium and tiotropium groups and 238 in the placebo group) was determined.
In order to account for the expected 10% match failures because of the intention-to-treat principle (see below), we randomly generated 1000 datasets, each with the proposed sample size of 620 and the expected matching effect of 30% after 3 years against a background of 30% immune reactions.
Our proposed sample size of 278 per group is sufficient to achieve power of 75%, at the 5% significance level.
