Qno5 A) Solution

Q no 5
a)
Solution:
To calculate the sample size required per group, we can use the formula:
N = [(Zα/2 + Zβ) * σ / d] ^2
Where:
 N is the required sample size per group.
 Zα/2 is the critical value for a two-tailed test at the desired significance level (α/2). For a
0.05 significance level, Zα/2 is approximately 1.96.
 Zβ is the critical value for the desired power (90% power corresponds to Zβ of
approximately 1.28).
 d is the effect size, which is the difference in infant length between the treatment and
control groups, expressed in standard deviations (0.2 in this case).
 σ is the population standard deviation.
Since we don’t have the population standard deviation (σ), we'll assume a common value of σ =
1 for this calculation.
Now, using the formula:
N = [(1.96 + 1.28) * 1 / 0.2] ^2
Calculating this:
N = (3.24 / 0.2) ^2
N = 16.2^2
N ≈ 262.44
Since you can't have a fraction of a participant, you should round up to the nearest whole
number. Therefore, the sample size required per group is approximately 263. Thus, you would
need 263 participants in each of the two groups for a total sample size of 526 to detect a
difference in infant length of 0.2 standard deviations with 90% power and a significance level of
0.05.
b)
To calculate the required sample size per group while considering non-compliance with
treatment, we need to account for the anticipated dropout rate of 25%.
The formula for sample size estimation with non-compliance is:
N = N_Actual / (1 - p)
Where:
N is the required sample size per group, considering non-compliance.
N_Actual is the desired sample size per group without considering non-compliance.
p is the anticipated dropout rate, expressed as a decimal (25% dropout corresponds to p = 0.25).
In our previous calculation, we determined that the desired sample size per group (without non-
compliance) is approximately 263.
N = 263 / (1 - 0.25)
Calculating this:
N = 263 / 0.75
N ≈ 350.67
Thus, we need 351 participants in each of the two groups, for a total sample size of 702, to
account for a 25% dropout rate and detect a difference in infant length of 0.2 standard deviations
with 90% power and a significance level of 0.05.
c)
Yes, it is reasonable to adjust for the age at which infants were first introduced to solid foods in
the primary analysis. This adjustment is necessary because the trial investigator believes that the
age at which infants were introduced to solid foods is an important predictor of length at 12
months. By adjusting for this variable, the investigator can account for any potential confounding
effects and ensure that the comparison between the treatment groups is more accurate and
unbiased. Adjusting for this variable will help improve the validity and reliability of the study
results.
d)
To maintain an overall Type I error rate of 0.05 (or any other preset significance threshold) in a
situation where several analyses are done throughout a clinical trial, you must employ a strategy
that modulates the alpha level. The Bonferroni correction is a popular technique for this.
In this scenario, in addition to the final analysis, the investigator intends to examine the primary
outcome at three separate time periods (6 months, 12 months, and 18 months). Because they
recommend an alpha level of 0.05 for each of these studies, you should keep the total alpha at
0.05.
To do this, you would employ the Bonferroni correction, with the following calculation for
adjusting the alpha level:
Overall, Alpha / Number of Analyses = Adjusted Alpha Level
So, for this case:
Alpha Level Adjusted = 0.05 / 4 = 0.0125
To keep the overall alpha at 0.05, each individual analysis should have an alpha level of 0.0125.
This change minimises the likelihood of making a Type I mistake across numerous studies, but it
also makes each individual study stricter and makes statistical significance more difficult to
attain. If any of the analyses achieves statistical significance at the adjusted alpha threshold of
0.0125, the null hypothesis is rejected, and the trial is terminated early.

Qno5 A) Solution

Uploaded by

Copyright:

Available Formats

Qno5 A) Solution

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Qno5 A) Solution

Uploaded by

Copyright:

Available Formats

Q no 5

You might also like